%0 Journal Article %J arXiv preprint arXiv:1606.09610 %D 2016 %T A Crowdsourcing Approach to Collecting 399 Tutorial Videos on Logarithms %A Whitehill, Jacob %A Seltzer, Margo %B arXiv preprint arXiv:1606.09610 %G eng %0 Journal Article %J arXiv preprint arXiv:1602.08610 %D 2016 %T Scalable Bayesian Rule Lists %A Yang, Hongyu %A Rudin, Cynthia %A Seltzer, Margo %B arXiv preprint arXiv:1602.08610 %G eng %0 Conference Proceedings %B International Conference on Extending Database Technology (EDBT), %D 2016 %T Designing Access Methods: The RUM Conjecture %A Athanassoulis, Manos %A Michael S. Kester %A Lukas M. Maas %A Radu Stoica %A Stratos Idreos %A Anastasia Ailamaki %A Mark Callaghan %X The database research community has been building methods to store, access, and update data for more than four decades. Throughout the evolution of the structures and techniques used to access data, access methods adapt to the ever changing hardware and workload requirements. Today, even small changes in the workload or the hardware lead to a redesign of access methods. The need for new designs has been increasing as data generation and workload diversification grow exponentially, and hardware advances introduce increased complexity. New workload requirements are introduced by the emergence of new applications, and data is managed by large systems composed of more and more complex and heterogeneous hardware. As a result, it is increasingly important to develop application-aware and hardware-aware access methods. The fundamental challenges that every researcher, systems architect, or designer faces when designing a new access method are how to minimize, i) read times (R), ii) update cost (U), and iii) memory (or storage) overhead (M). In this paper, we conjecture that when optimizing the read-update-memory overheads, optimizing in any two areas negatively impacts the third. We present a simple model of the RUM overheads, and we articulate the RUM Conjecture. We show how the RUM Conjecture manifests in state-of-the-art access methods, and we envision a trend toward RUM-aware access methods for future data systems. %B International Conference on Extending Database Technology (EDBT), %C Bordeaux, France %@ 978-3-89318-070-7 %G eng %0 Conference Proceedings %B NSDI 2016 %D 2016 %T Polaris: Faster Page Loads Using Fine-grained Dependency Tracking %E Ravi Netravali %E Ameesh Goyal %E James Mickens %E Hari Balakrishnan %X

To load a web page, a browser must fetch and evaluate objects like HTML files and JavaScript source code. Evaluating an object can result in additional objects being fetched and evaluated. Thus, loading a web page requires a browser to resolve a dependency graph; this partial ordering constrains the sequence in which a browser can process individual objects. Unfortunately, many edges in a page’s dependency graph are unobservable by today’s browsers. To avoid violating these hidden dependencies, browsers make conservative assumptions about which objects to process next, leaving the network and CPU underutilized.

We provide two contributions. First, using a new measurement platform called Scout that tracks fine-grained data flows across the JavaScript heap and the DOM, we show that prior, coarse-grained dependency analyzers miss crucial edges: across a test corpus of 200 pages, prior approaches miss 30% of edges at the median, and 118% at the 95th percentile. Second, we quantify the benefits of exposing these new edges to web browsers. We introduce Polaris, a dynamic client-side scheduler that is written in JavaScript and runs on unmodified browsers; using a fully automatic compiler, servers can translate normal pages into ones that load themselves with Polaris. Polaris uses fine-grained dependency graphs to dynamically determine which objects to load, and when. Since Polaris’ graphs have no missing edges, Polaris can aggressively fetch objects in a way that minimizes network round trips. Experiments in a variety of network conditions show that Polaris decreases page load times by 34% at the median, and 59% at the 95th percentile.

%B NSDI 2016 %C Santa Clara, CA %G eng %U http://mickens.seas.harvard.edu/files/mickens/files/polaris.pdf %0 Conference Proceedings %B SIGMOD International Conference on Management of Data %D 2016 %T Adaptive Indexing over Encrypted Numeric Data %A P. Karras %A A. Nikitin %A M. Saad %A R. Bhatt %A D. Antyukhov %A S. Idreos %X Today, outsourcing query processing tasks to remote cloud servers becomes a viable option; such outsourcing calls for encrypting data stored at the server so as to render it secure against eavesdropping adversaries and/or an honest-but-curious server itself. At the same time, to be efficiently managed, outsourced data should be indexed, and even adaptively so, as a side-effect of query pro- cessing. Computationally heavy encryption schemes render such outsourcing unattractive; an alternative, Order-Preserving Encryption Scheme (OPES), intentionally preserves and reveals the order in the data, hence is unattractive from the security viewpoint. In this paper, we propose and analyze a scheme for lightweight and indexable encryption, based on linear-algebra operations. Our scheme provides higher security than OPES and allows for range and point queries to be efficiently evaluated over encrypted numeric data, with decryption performed at the client side. We implement a prototype that performs incremental, query-triggered adaptive indexing over encrypted numeric data based on this scheme, without leaking order information in advance, and without prohibitive overhead, as our extensive experimental study demonstrates. %B SIGMOD International Conference on Management of Data %I ACM %C San Francisco, CA %G eng %R 10.1145/2882903.2882932 %0 Conference Proceedings %B NSDI 2016 %D 2016 %T Polaris: Faster Page Loads Using Fine-grained Dependency Tracking %E Ravi Netravali %E Ameesh Goyal %E James Mickens %E Hari Balakrishnan %X

To load a web page, a browser must fetch and evaluate objects like HTML files and JavaScript source code. Evaluating an object can result in additional objects being fetched and evaluated. Thus, loading a web page requires a browser to resolve a dependency graph; this partial ordering constrains the sequence in which a browser can process individual objects. Unfortunately, many edges in a page’s dependency graph are unobservable by today’s browsers. To avoid violating these hidden dependencies, browsers make conservative assumptions about which objects to process next, leaving the network and CPU underutilized.

We provide two contributions. First, using a new measurement platform called Scout that tracks fine-grained data flows across the JavaScript heap and the DOM, we show that prior, coarse-grained dependency analyzers miss crucial edges: across a test corpus of 200 pages, prior approaches miss 30% of edges at the median, and 118% at the 95th percentile. Second, we quantify the benefits of exposing these new edges to web browsers. We introduce Polaris, a dynamic client-side scheduler that is written in JavaScript and runs on unmodified browsers; using a fully automatic compiler, servers can translate normal pages into ones that load themselves with Polaris. Polaris uses fine-grained dependency graphs to dynamically determine which objects to load, and when. Since Polaris’ graphs have no missing edges, Polaris can aggressively fetch objects in a way that minimizes network round trips. Experiments in a variety of network conditions show that Polaris decreases page load times by 34% at the median, and 59% at the 95th percentile.

%B NSDI 2016 %C Santa Clara, CA %G eng %U http://mickens.seas.harvard.edu/files/mickens/files/polaris.pdf %0 Conference Paper %B Proceedings of the 29th ACM on International Conference on Supercomputing %D 2015 %T Automatically Scalable Computation %A Seltzer, Margo %B Proceedings of the 29th ACM on International Conference on Supercomputing %I ACM %P 283–283 %G eng %0 Journal Article %J Proceedings of the VLDB Endowment %D 2015 %T A scalable distributed graph partitioner %A Daniel Margo %A Seltzer, Margo %B Proceedings of the VLDB Endowment %I VLDB Endowment %V 8 %P 1478–1489 %G eng %N 12 %0 Conference Paper %B 2015 International Conference on Parallel Architecture and Compilation (PACT) %D 2015 %T Towards general-purpose neural network computing %A Eldridge, Schuyler %A Waterland, Amos %A Seltzer, Margo %A Appavoo, Jonathan %A Joshi, Ajay %B 2015 International Conference on Parallel Architecture and Compilation (PACT) %I IEEE %P 99–112 %G eng %0 Conference Paper %B 7th USENIX Workshop on the Theory and Practice of Provenance (TaPP 15) %D 2015 %T Recent advances in computer architecture: the opportunities and challenges for provenance %A Balakrishnan, Nikilesh %A Bytheway, Thomas %A Carata, Lucian %A Chick, Oliver RA %A Snee, James %A Akoush, Sherif %A Sohan, Ripduman %A Seltzer, Margo %A Hopper, Andy %B 7th USENIX Workshop on the Theory and Practice of Provenance (TaPP 15) %G eng %0 Conference Proceedings %B 2015 IEEE 31st International Conference on Data Engineering (ICDE) %D 2015 %T LLAMA: Efficient graph analytics using Large Multiversioned Arrays %A Peter Macko %A Virendra Marathe %A Daniel Margo %A Seltzer, Margo %X We present LLAMA, a graph storage and analysis system that supports mutability and out-of-memory execution. LLAMA performs comparably to immutable main-memory analysis systems for graphs that fit in memory and significantly outperforms existing out-of-memory analysis systems for graphs that exceed main memory. LLAMA bases its implementation on the compressed sparse row (CSR) representation, which is a read-only representation commonly used for graph analytics. We augment this representation to support mutability and persistence using a novel implementation of multi-versioned array snapshots, making it ideal for applications that receive a steady stream of new data, but need to perform whole-graph analysis on consistent views of the data. We compare LLAMA to state-of-the-art systems on representative graph analysis workloads, showing that LLAMA scales well both out-of-memory and across parallel cores. Our evaluation shows that LLAMA’s mutability introduces modest overheads of 3-18% relative to immutable CSR for in-memory execution and that it outperforms state-of-the-art out-of-memory systems in most cases, with a best case improvement of 5x on breadth-first-search. %B 2015 IEEE 31st International Conference on Data Engineering (ICDE) %I IEEE %C Seoul, South Korea %8 04/2015 %G eng %R 10.1109/ICDE.2015.7113298 %0 Conference Proceedings %B Proceedings of the VLDB Endowment %D 2015 %T A Scalable Distributed Graph Partitioner %A Daniel W. Margo %A Margo I. Seltzer %X

We present Scalable Host-tree Embeddings for Efficient Partitioning (Sheep), a distributed graph partitioning algorithm capable of handling graphs that far exceed main memory. Sheep produces high quality edge partitions an order of magnitude faster than both state of the art offline (eg, METIS) and streaming partitioners (eg, Fennel). Sheep’s partitions are independent of the input graph distribution, which means that graph elements can be assigned to processing nodes arbitrarily without affecting the partition quality.

%B Proceedings of the VLDB Endowment %I The VLDB Endowment %C Kohala Coast, Hawaii %V 8 %P 1478-1489 %8 08/2015 %G eng %R 10.14778/2824032.2824046 %0 Conference Paper %B ACM SIGPLAN Notices %D 2014 %T ASC: Automatically scalable computation %A Waterland, Amos %A Angelino, Elaine %A Adams, Ryan P %A Appavoo, Jonathan %A Seltzer, Margo %B ACM SIGPLAN Notices %I ACM %V 49 %P 575–590 %G eng %N 4 %0 Book %D 2014 %T A Framework for Incentivizing Deep Fixes %A Malvika Rao %A David C Parkes %A Seltzer, Margo %A Bacon, David F %I WIT-EC %V 14 %G eng %0 Journal Article %J Communications of the ACM %D 2014 %T A primer on provenance %A Carata, Lucian %A Akoush, Sherif %A Balakrishnan, Nikilesh %A Bytheway, Thomas %A Sohan, Ripduman %A Seltzer, Margo %A Hopper, Andy %B Communications of the ACM %I ACM %V 57 %P 52–60 %G eng %N 5 %0 Journal Article %J arXiv preprint arXiv:1403.7265 %D 2014 %T Accelerating MCMC via parallel predictive prefetching %A Angelino, Elaine %A Kohler, Eddie %A Waterland, Amos %A Seltzer, Margo %A Adams, Ryan P %B arXiv preprint arXiv:1403.7265 %G eng %0 Journal Article %J IEEE Computer Society %D 2014 %T Avoiding the top 10 software security design flaws %A Arce, Iván %A Clark-Fisher, Kathleen %A Daswani, Neil %A DelGrosso, Jim %A Dhillon, Danny %A Kern, Christoph %A Kohno, Tadayoshi %A Landwehr, Carl %A McGraw, Gary %A Schoenfield, Brook %A others %B IEEE Computer Society %G eng %0 Journal Article %J Proc. Neuromorphic Architectures (NeuroArch) Workshop at 41th International Symposium on Computer Architecture (ISCA-41) %D 2014 %T Programmable smart machines: A hybrid neuromorphic approach to general purpose computation %A Appavoo, Jonathan %A Waterland, Amos %A Zhao, Katherine %A Eldridge, Schuyler %A Joshi, Ajay %A Seltzer, Margo %A Homer, Steve %B Proc. Neuromorphic Architectures (NeuroArch) Workshop at 41th International Symposium on Computer Architecture (ISCA-41) %G eng %0 Conference Paper %B Proceedings of the 6th International Systems and Storage Conference %D 2013 %T Computational caches %A Waterland, Amos %A Angelino, Elaine %A Cubuk, Ekin D %A Efthimios Kaxiras %A Adams, Ryan P %A Appavoo, Jonathan %A Seltzer, Margo %B Proceedings of the 6th International Systems and Storage Conference %I ACM %P 8 %G eng %0 Conference Paper %B PHARMACOEPIDEMIOLOGY AND DRUG SAFETY %D 2013 %T Automated Use of Electronic Health Record Text Data To Improve Validity in Pharmacoepidemiology Studies %A Rasssen, Jeremy A %A Wahl, Peter M %A Angelino, Elaine %A Seltzer, Margo I %A Rosenman, Marc D %A Schneeweiss, Sebastian %B PHARMACOEPIDEMIOLOGY AND DRUG SAFETY %I WILEY-BLACKWELL 111 RIVER ST, HOBOKEN 07030-5774, NJ USA %V 22 %P 376–376 %G eng %0 Conference Paper %B Proceedings of the 22nd ACM international conference on Information & Knowledge Management %D 2013 %T Local clustering in provenance graphs %A Peter Macko %A Daniel Margo %A Seltzer, Margo %B Proceedings of the 22nd ACM international conference on Information & Knowledge Management %I ACM %P 835–840 %G eng %0 Journal Article %D 2013 %T Local clustering in provenance graphs (extended version) %A Peter Macko %A Margo, Daniel Wyatt %A Seltzer, Margo I %G eng %0 Conference Paper %B Proceedings of the 6th International Systems and Storage Conference %D 2013 %T Performance introspection of graph databases %A Peter Macko %A Daniel Margo %A Seltzer, Margo %B Proceedings of the 6th International Systems and Storage Conference %I ACM %P 18 %G eng %0 Journal Article %J Computer %D 2013 %T Applying KISS to Healthcare Information Technology %A Herzlinger, Regina %A Seltzer, Margo %A Mark Gaynor %B Computer %I IEEE %P 72–74 %G eng %N 11 %0 Conference Paper %B Presented as part of the 2013 USENIX Annual Technical Conference (USENIX ATC 13) %D 2013 %T Flash caching on the storage client %A Holland, David A %A Angelino, Elaine %A Wald, Gideon %A Seltzer, Margo I %B Presented as part of the 2013 USENIX Annual Technical Conference (USENIX ATC 13) %P 127–138 %G eng %0 Journal Article %J IEEE Transactions on Visualization and Computer Graphics %D 2013 %T Evaluation of filesystem provenance visualization tools %A Michelle A. Borkin %A Chelsea S. Yeh %A Madelaine Boyd %A Peter Macko %A Krzysztof Z. Gajos %A Seltzer, Margo %A Hanspeter Pfister %B IEEE Transactions on Visualization and Computer Graphics %I IEEE %V 19 %P 2476–2485 %G eng %N 12 %0 Conference Paper %B Presented as part of the 4th USENIX Workshop on Hot Topics in Parallelism %D 2012 %T Parallelization by simulated tunneling %A Waterland, Amos %A Appavoo, Jonathan %A Seltzer, Margo %B Presented as part of the 4th USENIX Workshop on Hot Topics in Parallelism %G eng %0 Journal Article %J University of California, Santa Cruz, Tech. Rep. UCSC-SSRC-12-01 %D 2012 %T Making sense of file systems through provenance and rich metadata %A Parker-Wood, Aleatha %A Long, Darrell DE %A Miller, Ethan L %A Seltzer, Margo %A Tunkelang, Daniel %B University of California, Santa Cruz, Tech. Rep. UCSC-SSRC-12-01 %G eng %0 Conference Paper %B Proceedings of the ACM EuroSys Conference (EuroSys 2012) %D 2012 %T Cache Craftiness for Fast Multicore Key-Value Storage %A Yandong Mao %A Kohler, Eddie %A Robert Morris %B Proceedings of the ACM EuroSys Conference (EuroSys 2012) %C Bern, Switzerland %8 April %G eng %0 Conference Paper %B TaPP %D 2012 %T A General-Purpose Provenance Library. %A Peter Macko %A Seltzer, Margo %B TaPP %G eng %0 Journal Article %D 2012 %T Mining the Web for Medical Hypothesis: A Proof-of-Concept System %A Diana Maclean %A Seltzer, Margo I %G eng %0 Journal Article %J TaPP %D 2012 %T BURRITO: Wrapping Your Lab Notebook in Computational Infrastructure. %A Guo, Philip J %A Seltzer, Margo %B TaPP %V 12 %P 7–7 %G eng %0 Conference Proceedings %B ACM/IEEE 11th Conference on Information Processing in Sensor Networks (IPSN 2012) %D 2012 %T Simbeeotic: A Simulator and Testbed for Micro-Aerial Vehicle Swarm Experiments %A Bryan Kate %A Jason Waterman %A Karthik Dantu %A Welsh, Matt %X Micro-aerial vehicle (MAV) swarms are an emerging class of mobile sensing systems. Simulation and staged deployment to prototype testbeds are useful in the early stages of large-scale system design, when hardware is unavailable or deployment at scale is impractical. To faithfully represent the problem domain, a MAV swarm simulator must be able to model the key aspects of the system: actuation, sensing, and communication. We present Simbeeotic, a simulation framework geared toward modeling swarms of MAVs. Simbeeotic enables algorithm development and rapid MAV prototyping through pure simulation and hardware-in-the-loop experimentation. We demonstrate that Simbeeotic provides the appropriate level of fidelity to evaluate prototype systems while maintaining the ability to test at scale. %B ACM/IEEE 11th Conference on Information Processing in Sensor Networks (IPSN 2012) %I IEEE/ACM %C Beijing, China %8 04/2012 %G eng %U http://www.eecs.harvard.edu/ bkate/pubs/simbeeotic-ipsn12.pdf %0 Journal Article %J Commun. ACM %D 2011 %T Making information flow explicit in HiStar %A Zeldovich, Nickolai %A Boyd-Wickizer, Silas %A Kohler, Eddie %A Mazières, David %B Commun. ACM %I ACM %C New York, NY, USA %V 54 %P 93–101 %G eng %U http://doi.acm.org/10.1145/2018396.2018419 %N 11 %R 10.1145/2018396.2018419 %0 Conference Paper %B Proceedings of the 13th USENIX conference on Hot topics in operating systems %D 2011 %T Benchmarking file system benchmarking: it *IS* rocket science %A Tarasov, Vasily %A Bhanage, Saumitra %A Zadok, Erez %A Seltzer, Margo %B Proceedings of the 13th USENIX conference on Hot topics in operating systems %S HotOS’13 %I USENIX Association %C Berkeley, CA, USA %P 9–9 %G eng %U http://dl.acm.org/citation.cfm?id=1991596.1991609 %0 Journal Article %D 2011 %T Collecting provenance via the Xen hypervisor %A Seltzer, Margo I %A Peter Macko %A Chiarini, Marc A %I USENIX Association %G eng %0 Journal Article %D 2011 %T Provenance Integration Requires Reconciliation %A Seltzer, Margo I %A Angelino, Elaine Lee %A Braun, Uri Jacob %A Holland, David A %A Peter Macko %A Margo, Daniel Wyatt %G eng %0 Journal Article %D 2011 %T Provenance map orbiter: interactive exploration of large provenance graphs %A Seltzer, Margo I %A Peter Macko %I USENIX Association %G eng %0 Journal Article %D 2011 %T USENIX Association Board of Directors Meeting August 12, 2011 Westin St. Francis Hotel Minutes 9/15/11 %A Seltzer, Margo %A Provos, Niels %A Blaze, Matt %A Couch, Alva %A Arrasjid, John %A Blank-Edelman, David %A Noble, Brian %A Young, Ellie %A Dickison, Anne %A Long, Jane-Ellen %A others %G eng %0 Journal Article %D 2011 %T Addressing underspecified lineage queries on provenance %A Margo, Daniel Wyatt %A Peter Macko %A Seltzer, Margo I %G eng %0 Conference Paper %B TaPP %D 2011 %T Provenance Map Orbiter: Interactive Exploration of Large Provenance Graphs. %A Peter Macko %A Seltzer, Margo %B TaPP %G eng %0 Conference Paper %B TaPP %D 2011 %T Collecting Provenance via the Xen Hypervisor. %A Peter Macko %A Chiarini, Marc %A Seltzer, Margo %A Harvard, SEAS %B TaPP %G eng %0 Journal Article %J Communications of the ACM %D 2011 %T The magazine archive includes every article published in Communications of the ACM for over the past 50 years. %A Kroeker, Kirk L %B Communications of the ACM %V 54 %P 11–14 %G eng %N 10 %0 Journal Article %D 2011 %T Multicore OSes %A Holland, David A %A Seltzer, Margo I %G eng %0 Conference Paper %B Proceedings of the 13th USENIX conference on Hot topics in operating systems %D 2011 %T Multicore OSes: looking forward from 1991, er, 2011 %A David A. Holland %A Margo I. Seltzer %B Proceedings of the 13th USENIX conference on Hot topics in operating systems %S HotOS’13 %I USENIX Association %C Berkeley, CA, USA %P 33–33 %G eng %U http://dl.acm.org/citation.cfm?id=1991596.1991640 %0 Conference Paper %B Proceedings of the 48th Design Automation Conference %D 2011 %T Dimetrodon: processor-level preventive thermal management via idle cycle injection %A Bailis, Peter %A Reddi, Vijay Janapa %A Gandhi, Sanjay %A David Brooks %A Seltzer, Margo %B Proceedings of the 48th Design Automation Conference %S DAC ’11 %I ACM %C New York, NY, USA %P 89–94 %@ 978-1-4503-0636-2 %G eng %U http://doi.acm.org/10.1145/2024724.2024745 %R 10.1145/2024724.2024745 %0 Conference Paper %B TaPP %D 2011 %T Provenance Integration Requires Reconciliation. %A Angelino, Elaine %A Uri Braun %A Holland, David A %A Margo, Daniel W %B TaPP %G eng %0 Conference Paper %B 9th ACM Conference on Embedded Networked Sensor Systems (SenSys 2011) %D 2011 %T Programming micro-aerial vehicle swarms with Karma %A Karthik Dantu %A Bryan Kate %A Jason Waterman %A Bailis, Peter %A Welsh, Matt %X Research in micro-aerial vehicle (MAV) construction, control, and high-density power sources is enabling swarms of MAVs as a new class of mobile sensing systems. For efficient operation, such systems must adapt to dynamic environments, cope with uncertainty in sensing and control, and operate with limited resources. We propose a novel system architecture based on a hive-drone model that simplifies the functionality of an individual MAV to a sequence of sensing and actuation commands with no in-field communication. This decision simplifies the hardware and software complexity of individual MAVs and moves the complexity of coordination entirely to a central hive computer. We present Karma, a system for programming and managing MAV swarms. Through simulation and testbed experiments we demonstrate how applications in Karma can run on limited resources, are robust to individual MAV failure, and adapt to changes in the environment. %B 9th ACM Conference on Embedded Networked Sensor Systems (SenSys 2011) %I ACM %C Seattle, WA %8 11/2011 %@ 978-1-4503-0718-5 %G eng %U http://www.eecs.harvard.edu/ kar/files/karma-sensys11.pdf %R 10.1145/2070942.2070956 %0 Journal Article %J SIGOPS Oper. Syst. Rev. %D 2010 %T Provenance as first class cloud data %A Kiran-Kumar Muniswamy-Reddy %A Seltzer, Margo %B SIGOPS Oper. Syst. Rev. %I ACM %C New York, NY, USA %V 43 %P 11–16 %8 January %G eng %U http://doi.acm.org/10.1145/1713254.1713258 %R http://doi.acm.org/10.1145/1713254.1713258 %0 Journal Article %J ACM Trans. Sen. Netw. %D 2010 %T The Tenet architecture for tiered sensor networks %A Paek, Jeongyeup %A Greenstein, Ben %A Gnawali, Omprakash %A Jang, Ki-Young %A Joki, August %A Vieira, Marcos %A Hicks, John %A Estrin, Deborah %A Govindan, Ramesh %A Kohler, Eddie %B ACM Trans. Sen. Netw. %I ACM %C New York, NY, USA %V 6 %P 34:1–34:44 %G eng %U http://doi.acm.org/10.1145/1777406.1777413 %R http://doi.acm.org/10.1145/1777406.1777413 %0 Conference Paper %B Proceedings of the Workshop on Programmable Routers for Extensible Services of Tomorrow %D 2010 %T Evaluating the suitability of server network cards for software routers %A Manesh, Maziar %A Argyraki, Katerina %A Dobrescu, Mihai %A Egi, Norbert %A Fall, Kevin %A Iannaccone, Gianluca %A Kohler, Eddie %A Ratnasamy, Sylvia %B Proceedings of the Workshop on Programmable Routers for Extensible Services of Tomorrow %S PRESTO ’10 %I ACM %C New York, NY, USA %P 7:1–7:6 %@ 978-1-4503-0467-2 %G eng %U http://doi.acm.org/10.1145/1921151.1921161 %R http://doi.acm.org/10.1145/1921151.1921161 %0 Conference Paper %B Proceedings of the 8th USENIX conference on File and storage technologies %D 2010 %T Provenance for the cloud %A Kiran-Kumar Muniswamy-Reddy %A Peter Macko %A Seltzer, Margo %B Proceedings of the 8th USENIX conference on File and storage technologies %S FAST’10 %I USENIX Association %C Berkeley, CA, USA %P 15–14 %G eng %U http://dl.acm.org/citation.cfm?id=1855511.1855526 %0 Conference Paper %B TaPP %D 2010 %T Using Provenance to Extract Semantic File Attributes. %A Margo, Daniel W %A Robin Smogor %B TaPP %G eng %0 Conference Paper %B Proceedings of the 8th USENIX conference on File and storage technologies %D 2010 %T Tracking back references in a write-anywhere file system %A Peter Macko %A Seltzer, Margo %A Keith A. Smith %B Proceedings of the 8th USENIX conference on File and storage technologies %S FAST’10 %I USENIX Association %C Berkeley, CA, USA %P 2–2 %G eng %U http://dl.acm.org/citation.cfm?id=1855511.1855513 %0 Conference Paper %B Proceedings of the 2nd conference on Theory and practice of provenance %D 2010 %T Towards query interoperability: PASSing PLUS %A Braun, Uri J. %A Margo I. Seltzer %A Chapman, Adriane %A Blaustein, Barbara %A Allen, M. David %A Seligman, Len %B Proceedings of the 2nd conference on Theory and practice of provenance %S TAPP’10 %I USENIX Association %C Berkeley, CA, USA %P 3–3 %G eng %U http://dl.acm.org/citation.cfm?id=1855795.1855798 %0 Conference Paper %B Proceedings of the 7th International Conference on Swarm Intelligence %D 2010 %T Positional communication and private information in honeybee foraging models %A Bailis, Peter %A Radhika Nagpal %A Werfel, Justin %B Proceedings of the 7th International Conference on Swarm Intelligence %S ANTS’10 %I Springer-Verlag %C Berlin, Heidelberg %P 263–274 %@ 978-3-642-15460-7 %G eng %U http://dl.acm.org/citation.cfm?id=1884958.1884982 %0 Conference Paper %B International Provenance and Annotation Workshop %D 2010 %T StarFlow: A script-centric data analysis environment %A Angelino, Elaine %A Yamins, Daniel %A Seltzer, Margo %B International Provenance and Annotation Workshop %I Springer Berlin Heidelberg %P 236–250 %G eng %0 Journal Article %J arXiv preprint arXiv:0902.4730 %D 2009 %T Minimal Economic Distributed Computing %A Saul Youssef %A John Brunelle %A John Huth %A David C Parkes %A Seltzer, Margo %A Jim Shank %B arXiv preprint arXiv:0902.4730 %G eng %0 Conference Paper %B Proceedings of the 12th conference on Hot topics in operating systems %D 2009 %T Hierarchical file systems are dead %A Seltzer, Margo %A Nicholas Murphy %B Proceedings of the 12th conference on Hot topics in operating systems %S HotOS’09 %I USENIX Association %C Berkeley, CA, USA %P 1–1 %G eng %U http://dl.acm.org/citation.cfm?id=1855568.1855569 %0 Journal Article %J ACM Transactions on Storage (TOS) %D 2009 %T Introduction to special issue FAST 2009 %A Seltzer, Margo %A Wheeler, Ric %B ACM Transactions on Storage (TOS) %I ACM %V 5 %P 11 %G eng %N 4 %0 Conference Paper %B Proceedings of the 24th ACM SIGPLAN conference companion on Object oriented programming systems languages and applications %D 2009 %T Provenance: a future history %A Cheney, James %A Stephen Chong %A Foster, Nate %A Seltzer, Margo %A Vansummeren, Stijn %B Proceedings of the 24th ACM SIGPLAN conference companion on Object oriented programming systems languages and applications %I ACM %P 957–964 %G eng %0 Journal Article %J IEEE Transactions on Information Technology in Biomedicine %D 2009 %T Monitoring Motor Fluctuations in Patients With Parkinson’s Disease Using Wearable Sensors %A Shyamal Patel %A Konrad Lorincz %A Richard Hughes %A Nancy Huggins %A John Growden %A David Standaert %A Metin Akay %A Dy, Jennifer %A Welsh, Matt %A Paolo Bonato %B IEEE Transactions on Information Technology in Biomedicine %V 13 %8 2009 %G eng %U /files/syrah/files/parkinsons-titb09.pdf %N 6 %0 Conference Paper %B 1st Workshop on the Theory and Practice of Provenance (TaPP’09) %D 2009 %T The Case for Browser Provenance %A Daniel Margo %A Seltzer, Margo %B 1st Workshop on the Theory and Practice of Provenance (TaPP’09) %8 February 2009 %G eng %0 Conference Paper %B 7th USENIX Conference on File and Storage Technologies (FAST ’09) %D 2009 %T Causality-Based Versioning %A Kiran-Kumar Muniswamy-Reddy %A David A. Holland %B 7th USENIX Conference on File and Storage Technologies (FAST ’09) %8 February 2009 %G eng %U /files/syrah/files/fast09.pdf %0 Conference Paper %B 2009 USENIX Annual Technical Conference %D 2009 %T Layering in Provenance Systems %A Kiran-Kumar Muniswamy-Reddy %A Uri Braun %A David A. Holland %A Peter Macko %A Diana Maclean %A Daniel Margo %A Seltzer, Margo %A Robin Smogor %X Digital provenance describes the ancestry or history of a digital object. Most existing provenance systems, however, operate at only one level of abstraction: the system call layer, a workflow specification, or the high-level constructs of a particular application. The provenance collectable in each of these layers is different, and all of it can be important. Single-layer systems fail to account for the different levels of abstraction at which users need to reason about their data and processes. These systems cannot integrate data provenance across layers and cannot answer questions that require an integrated view of the provenance. We have designed a provenance collection structure facilitating the integration of provenance across multiple levels of abstraction, including a workflow engine, a web browser, and an initial runtime Python provenance tracking wrapper. We layer these components atop provenance-aware network storage (NFS) that builds upon a Provenance-Aware Storage System (PASS). We discuss the challenges of building systems that integrate provenance across multiple layers of abstraction, present how we augmented systems in each layer to ntegrate provenance, and present use cases that demonstrate how provenance spanning multiple layers provides functionality not available in existing systems.Our evaluation shows that the overheads imposed by layering provenance systems are reasonable. %B 2009 USENIX Annual Technical Conference %C San Diego, California %8 June 2009 %G eng %U /files/syrah/files/usenix09.pdf %0 Conference Paper %B 1st Workshop on the Theory and Practice of Provenance (TaPP’09) %D 2009 %T Making a Cloud Provenance-Aware %A Kiran-Kumar Muniswamy-Reddy %A Peter Macko %A Seltzer, Margo %B 1st Workshop on the Theory and Practice of Provenance (TaPP’09) %8 February 2009 %G eng %U /files/syrah/files/tapp09-cloud.pdf %0 Conference Paper %B Proceedings of the 7th ACM Conference on Embedded Networked Sensor Systems (SenSys’09) %D 2009 %T Mercury: A Wearable Sensor Network Platform for High-Fidelity Motion Analysis %A Konrad Lorincz %A Bor-Rong Chen %A Geoffrey Werner Challen %A Atanu Roy Chowdhury %A Shyamal Patel %A Paolo Bonato %A Welsh, Matt %B Proceedings of the 7th ACM Conference on Embedded Networked Sensor Systems (SenSys’09) %C Berkeley, CA %8 November %G eng %0 Conference Paper %B 12th Workshop on Hot Topics in Operating Systems (HotOS-XII) (2009) %D 2009 %T Peloton: Coordinated Resource Management for Sensor Networks %A Jason Waterman %A Geoffrey Werner Challen %A Welsh, Matt %B 12th Workshop on Hot Topics in Operating Systems (HotOS-XII) (2009) %8 May %G eng %U /files/syrah/files/peloton-hotos09.pdf %0 Conference Paper %B ACM SIGCOMM 2009 %D 2009 %T White Space Networking with Wi-Fi like Connectivity %A Paramvir Bahl %A Ranveer Chandra %A Thomas Moscibroda %A Rohan Murty %A Welsh, Matt %B ACM SIGCOMM 2009 %C Barcelona %8 August %G eng %U /files/syrah/files/whitefi-sigcomm09.pdf %0 Journal Article %J Communications of the ACM %D 2008 %T Beyond relational databases %A Seltzer, Margo %B Communications of the ACM %I ACM %V 51 %P 52–58 %G eng %N 7 %0 Journal Article %J ; login:: the magazine of USENIX & SAGE %D 2008 %T The present and future of SAN/NAS: interview with Dave Hitz and Brian Pawlowsky of NetApp %A Seltzer, Margo %B ; login:: the magazine of USENIX & SAGE %I USENIX Association %V 33 %P 39–44 %G eng %N 3 %0 Journal Article %J Concurrency and computation: practice and experience %D 2008 %T Special issue: The first provenance challenge %A Moreau, Luc %A Ludäscher, Bertram %A Altintas, Ilkay %A Barga, Roger S %A Bowers, Shawn %A Callahan, Steven %A Chin, George %A Clifford, Ben %A Cohen, Shirley %A Cohen-Boulakia, Sarah %A others %B Concurrency and computation: practice and experience %I Wiley Online Library %V 20 %P 409–418 %G eng %N 5 %0 Journal Article %J Target %D 2008 %T Proxy network coordinates %A Jonathan Ledlie %A Seltzer, Margo %A Peter Pietzuch %B Target %V 22 %P 25 %G eng %0 Conference Paper %B Proceedings of the 2nd International Provenance and Annotation Workshop %D 2008 %T Choosing a Data Model and Query Language for Provenance %A David A. Holland %A Uri Braun %A Diana Maclean %A Kiran-Kumar Muniswamy-Reddy %A Seltzer, Margo %B Proceedings of the 2nd International Provenance and Annotation Workshop %C Salt Lake City, Utah %8 June 2008 %G eng %U /files/syrah/files/ipaw08.pdf %0 Journal Article %D 2008 %T Layering in Provenance-Aware Storage Systems %A Kiran-Kumar Muniswamy-Reddy %A Joseph Barillari %A Uri Braun %A David A. Holland %A Diana Maclean %A Seltzer, Margo %A Stephen D. Holland %G eng %U /files/syrah/files/pass-TR-04-08.pdf %N TR-04-08 %0 Journal Article %J Concurrency and Control: Practice and Experience %D 2008 %T PASSing the provenance challenge %A David A. Holland %A Seltzer, Margo %A Uri Braun %A Kiran-Kumar Muniswamy-Reddy %B Concurrency and Control: Practice and Experience %V 20 %G eng %U /files/syrah/files/ccpe.pdf %& 531-540 %0 Conference Paper %B Proceedings of the 3rd USENIX Workshop on Hot Topics in Security (HotSec) %D 2008 %T Securing Provenance %A Uri Braun %A Avraham Shinnar %A Seltzer, Margo %B Proceedings of the 3rd USENIX Workshop on Hot Topics in Security (HotSec) %C San Jose, California %8 July 2008 %G eng %U /files/syrah/files/hotsec08.pdf %0 Conference Paper %B Proc. Seventh ACM Workshop on Hot Topics in Networks (HotNets-VII) %D 2008 %T An Architecture for Extensible Wireless LANs %A Rohan Murty %A Jitendra Padhye %A Alec Wolman %A Welsh, Matt %B Proc. Seventh ACM Workshop on Hot Topics in Networks (HotNets-VII) %8 October %G eng %U /files/syrah/files/trantor-hotnets08.pdf %0 Conference Paper %B 2008 IEEE International Conference on Technologies for Homeland Security %D 2008 %T CitySense: An Urban-Scale Wireless Sensor Network and Testbed %A Rohan Murty %A Geoffrey Mainland %A Ian Rose %A Atanu Roy Chowdhury %A Abhimanyu Gosain %A Josh Bers %A Welsh, Matt %B 2008 IEEE International Conference on Technologies for Homeland Security %8 May %G eng %U /files/syrah/files/citysense-ieeehst08.pdf %0 Conference Paper %B 5th USENIX/ACM Symposium on Networked Design and Implementation (NSDI’08) %D 2008 %T Designing High Performance Enterprise Wi-Fi Networks %A Rohan Murty %A Jitendra Padhye %A Alec Wolman %A Brian Zill %B 5th USENIX/ACM Symposium on Networked Design and Implementation (NSDI’08) %8 April %G eng %0 Conference Paper %B 13th ACM SIGPLAN International Conference on Functional Programming (ICFP 2008) %D 2008 %T Flask: Staged Functional Programming for Sensor Networks %A Geoffrey Mainland %A Greg Morrisett %A Welsh, Matt %B 13th ACM SIGPLAN International Conference on Functional Programming (ICFP 2008) %8 September %G eng %U /files/syrah/files/flask-icfp08.pdf %0 Conference Paper %B 6th ACM Conference on Embedded Networked Sensor Systems (SenSys’08) %D 2008 %T Lance: Optimizing High-Resolution Signal Collection in Wireless Sensor Networks %A Geoff Werner-Allen %A Stephen Dawson-Haggerty %A Welsh, Matt %B 6th ACM Conference on Embedded Networked Sensor Systems (SenSys’08) %8 November %G eng %U /files/syrah/files/lance-sensys08.pdf %0 Conference Paper %B 4th IEEE/ACM International Conference on Distributed Computing in Sensor Systems (DCOSS 2008) %D 2008 %T LiveNet: Using Passive Monitoring to Reconstruct Sensor Network Dynamics %A Bor-Rong Chen %A Geoffrey Peterson %A Geoff Mainland %A Welsh, Matt %B 4th IEEE/ACM International Conference on Distributed Computing in Sensor Systems (DCOSS 2008) %8 June %G eng %U /files/syrah/files/livenet-dcoss08.pdf %0 Conference Paper %B Fifth Workshop on Embedded Networked Sensors (HotEmNets’08) %D 2008 %T Pixie: An Operating System for Resource-Aware Programming of Embedded Sensors %A Konrad Lorincz %A Bor-Rong Chen %A Jason Waterman %A Geoff Werner-Allen %A Welsh, Matt %B Fifth Workshop on Embedded Networked Sensors (HotEmNets’08) %8 June %G eng %U /files/syrah/files/pixie-emnets08.pdf %0 Conference Paper %B 6th ACM Conference on Embedded Networked Sensor Systems (SenSys’08) %D 2008 %T Resource Aware Programming in the Pixie OS %A Konrad Lorincz %A Bor-Rong Chen %A Jason Waterman %A Geoff Werner-Allen %A Welsh, Matt %B 6th ACM Conference on Embedded Networked Sensor Systems (SenSys’08) %8 November %G eng %U /files/syrah/files/pixie-sensys08.pdf %0 Conference Paper %B 2008 IEEE International Conference on Technologies for Homeland Security %D 2008 %T Wireless Medical Sensor Networks in Emergency Response: Implementation and Pilot Results %A Tia Gao %A Christopher Pesto %A Leo Selavo %A Chen, Yin %A JeongGil Ko %A JongHyun Lim %A Andreas Terzis %A Andrew Watt %A James Jeng %A Bor-Rong Chen %A Konrad Lorincz %A Welsh, Matt %B 2008 IEEE International Conference on Technologies for Homeland Security %8 May %G eng %U /files/syrah/files/aidn-ieeehst08.pdf %0 Journal Article %J Queue %D 2007 %T A Conversation with Jamie Butler %A Stanik, John %B Queue %I ACM %V 5 %P 16–23 %G eng %N 1 %0 Journal Article %J IEEE Data Eng. Bull. %D 2007 %T Berkeley DB: A Retrospective. %A Seltzer, Margo I %B IEEE Data Eng. Bull. %V 30 %P 21–28 %G eng %N 3 %0 Journal Article %J NSDI %D 2007 %T Network Coordinates in the Wild. %A Jonathan Ledlie %A Gardner, Paul %A Seltzer, Margo I %B NSDI %V 7 %P 299–311 %G eng %0 Conference Paper %B IPTPS %D 2007 %T Wired Geometric Routing. %A Jonathan Ledlie %A Michael Mitzenmacher %A Seltzer, Margo I %A Peter Pietzuch %B IPTPS %G eng %0 Conference Paper %B Proceedings of the 16th International Conference on Parallel Architecture and Compilation Techniques %D 2007 %T Improving performance isolation on chip multiprocessors via an operating system scheduler %A Alexandra Fedorova %A Seltzer, Margo %A Smith, Michael D %B Proceedings of the 16th International Conference on Parallel Architecture and Compilation Techniques %I IEEE Computer Society %P 25–38 %G eng %0 Conference Paper %B First EuroSys Workshop on Virtualization for High-Performance Computing %D 2007 %T Can a File System Virtualize Processors? %A Lex Stein %A David A. Holland %A Seltzer, Margo %A Zhang, Zheng %X Supercomputers are comprising more and more processors and these processors are increasingly heterogeneous, with differing performance characteristics. The conventional programming models assume that all nodes run in lockstep. Thus, applications run at the speed of the least powerful processor. We introduce DesyncFS, a new programming model based on the block abstraction of traditional file systems. It virtualizes the performance characteristics of processors; this allows their heterogeneity to be hidden. We show that DesyncFS allows cluster throughput to scale with average processor throughput instead of being limited by the slowest processor. %B First EuroSys Workshop on Virtualization for High-Performance Computing %8 March 2007 %G eng %U /files/syrah/files/hpcvirt-07.pdf %0 Conference Paper %B 29th IEEE EMBS Annual International Conference %D 2007 %T Analysis of Feature Space for Monitoring Persons with Parkinson’s Disease With Application to a Wireless Wearable Sensor System %A Shyamal Patel %A Konrad Lorincz %A Richard Hughes %A Nancy Huggins %A John H. Growdon %A Welsh, Matt %A Paolo Bonato %B 29th IEEE EMBS Annual International Conference %8 August %G eng %U /files/syrah/files/parkinsons-embs07.pdf %0 Journal Article %D 2007 %T CitySense: A Vision for an Urban-Scale Wireless Networking Testbed %A Rohan Murty %A Abhimanyu Gosain %A Matthew Tierney %A Andrew Brody %A Amal Fahad %A Josh Bers %A Welsh, Matt %I Harvard University %8 September %G eng %U /files/syrah/files/citysense-techrept07.pdf %0 Conference Paper %B 4th USENIX/ACM Symposium on Networked Systems Design and Implementation (NSDI 2007) %D 2007 %T Cobra: Content-based Filtering and Aggregation of Blogs and RSS Feeds %A Ian Rose %A Rohan Murty %A Peter Pietzuch %A Jonathan Ledlie %A Mema Roussopoulos %A Welsh, Matt %B 4th USENIX/ACM Symposium on Networked Systems Design and Implementation (NSDI 2007) %8 April %G eng %U /files/syrah/files/cobra-nsdi07.pdf %0 Journal Article %D 2007 %T LiveNet: Using Passive Monitoring to Reconstruct Sensor Network Dynamics %A Bor-Rong Chen %A Geoffrey Peterson %A Geoff Mainland %A Welsh, Matt %I Harvard University %8 August %G eng %U /files/syrah/files/livenet-techrept07.pdf %0 Conference Paper %B First International Workshop on Systems and Networking Support for Healthcare and Assisted Living Environments (HealthNet’07) %D 2007 %T Participatory User Centered Design Techniques for a Large Scale Ad-Hoc Health Information System %A Tia Gao %A Tammara Massey %A Leo Selavo %A Welsh, Matt %A Majid Sarrafzadeh %B First International Workshop on Systems and Networking Support for Healthcare and Assisted Living Environments (HealthNet’07) %8 June %G eng %U /files/syrah/files/aidn-healthnet07.pdf %0 Conference Paper %B International Conference on Information Processing in Sensor Networks (IPSN’07) %D 2007 %T The Regiment Macroprogramming System %A Ryan Newton %A Greg Morrisett %A Welsh, Matt %B International Conference on Information Processing in Sensor Networks (IPSN’07) %8 April %G eng %U /files/syrah/files/regiment-ipsn07.pdf %0 Conference Paper %B First International Workshop on Agent Technology for Sensor Networks (ATSN-07) %D 2007 %T A Utility-Based Approach to Bandwidth Allocation and Link Scheduling in Wireless Networks %A Qicheng Ma %A David C. Parkes %A Welsh, Matt %B First International Workshop on Agent Technology for Sensor Networks (ATSN-07) %8 May %G eng %U /files/syrah/files/utility-atsn07.pdf %0 Conference Paper %B 26th IEEE International Conference on Distributed Computing Systems Workshops (ICDCSW'06) %D 2006 %T Network-aware overlays with network coordinates %A Peter Pietzuch %A Jonathan Ledlie %A Michael Mitzenmacher %A Seltzer, Margo %B 26th IEEE International Conference on Distributed Computing Systems Workshops (ICDCSW'06) %I IEEE %P 12–12 %G eng %0 Conference Paper %B WORLDS %D 2006 %T Data Management for Internet-Scale Single-Sign-On. %A Perl, Sharon E %A Seltzer, Margo %B WORLDS %G eng %0 Conference Paper %B 26th IEEE International Conference on Distributed Computing Systems (ICDCS'06) %D 2006 %T Stable and accurate network coordinates %A Jonathan Ledlie %A Peter Pietzuch %A Seltzer, Margo %B 26th IEEE International Conference on Distributed Computing Systems (ICDCS'06) %I IEEE %P 74–74 %G eng %0 Journal Article %D 2006 %T Cache-fair thread scheduling for multicore processors %A Alexandra Fedorova %A Seltzer, Margo I %A Smith, Michael D %G eng %0 Journal Article %J Proceedings of the Workshop on the Interaction between Operating Systems and Computer Architecture, in conjunction with ISCA %D 2006 %T A non-work-conserving operating system scheduler for SMT processors %A Alexandra Fedorova %A Seltzer, Margo %A Smith, Michael D %B Proceedings of the Workshop on the Interaction between Operating Systems and Computer Architecture, in conjunction with ISCA %V 33 %P 10–17 %G eng %0 Journal Article %D 2006 %T Egg: An extensible and economics-inspired open grid computing platform %A Brunelle, John A %A Peter Hurst %A John Huth %A Laura Kang %A Chaki Ng %A David C Parkes %A Seltzer, Margo I %A Jim Shank %A Saul Youssef %I World Scientific Publishing %G eng %0 Conference Paper %B Proceedings of the 2006 International Provenance and Annotation Workshop %D 2006 %T Issues in Automatic Provenance Collection %A Uri Braun %A Simson Garfinkel %A David A. Holland %A Kiran-Kumar Muniswamy-Reddy %A Seltzer, Margo %B Proceedings of the 2006 International Provenance and Annotation Workshop %C Chicago, Illinois %8 May 2006 %G eng %U /files/syrah/files/ipaw06.pdf %0 Conference Paper %B Proceedings of the 2006 USENIX Annual Technical Conference %D 2006 %T Provenance-Aware Storage Systems %A Kiran-Kumar Muniswamy-Reddy %A David A. Holland %A Uri Braun %A Seltzer, Margo %X A Provenance-Aware Storage System (PASS) is a storage system that automatically collects and maintains provenance or lineage, the complete history or ancestry of an item. We discuss the advantages of treating provenance as meta-data collected and maintained by the storage system, rather than as manual annotations stored in a separately administered database. We describe a PASS implementation, discussing the challenges it presents, performance cost it incurs, and the new functionality it enables. We show that with reasonable overhead, we can provide useful functionality not available in today’s file systems or provenance management systems. %B Proceedings of the 2006 USENIX Annual Technical Conference %C Boston, Massachusetts %8 June 2006 %G eng %U /files/syrah/files/usenix06.pdf %0 Conference Paper %B Second ACM/Sigmobile workshop on Multi-hop Ad Hoc Networks: From Theory to Reality (REALMAN’06) %D 2006 %T Ad-Hoc Multicast Routing on Resource-Limited Sensor Nodes %A Bor-Rong Chen %A Kiran-Kumar Muniswamy-Reddy %A Welsh, Matt %B Second ACM/Sigmobile workshop on Multi-hop Ad Hoc Networks: From Theory to Reality (REALMAN’06) %8 May %G eng %U /files/syrah/files/tinyadmr-realman06.pdf %0 Conference Paper %B International Workshop on Wearable and Implantable Body Sensor Networks (BSN 2006) %D 2006 %T Communicating Data from Wireless Sensor Networks using the HL7v3 Standard %A S. Baird %A Stephen Dawson-Haggerty %A Dan Myung %A Mark Gaynor %A Welsh, Matt %A Steve Moulton %B International Workshop on Wearable and Implantable Body Sensor Networks (BSN 2006) %8 April %G eng %0 Conference Paper %B American Medical Informatics Association Annual Conference (AMIA 2006) %D 2006 %T Design of a Decentralized Electronic Triage System %A Tammara Massey %A Tia Gao %A Welsh, Matt %A Jonathan Sharp %B American Medical Informatics Association Annual Conference (AMIA 2006) %8 November %G eng %0 Journal Article %D 2006 %T Flask: A Language for Data-driven Sensor Network Programs %A Geoffrey Mainland %A Welsh, Matt %A Greg Morrisett %I Harvard University %8 May %G eng %U /files/syrah/files/flask-TR-13-06.pdf %0 Conference Paper %B 22nd International Conference on Data Engineering (ICDE 2006) %D 2006 %T Network-Aware Operator Placement for Stream-Processing Systems %A Peter Pietzuch %A Jonathan Ledlie %A Jeffrey Shneidman %A Mema Roussopoulos %A Welsh, Matt %A Seltzer, Margo %B 22nd International Conference on Data Engineering (ICDE 2006) %8 April %G eng %U /files/syrah/files/sbon-icde06.pdf %0 Conference Paper %B Second Workshop on Hot Topics in Dependability (HotDep’06) %D 2006 %T Towards a Dependable Architecture for Internet-Scale Sensing %A Rohan Murty %A Welsh, Matt %B Second Workshop on Hot Topics in Dependability (HotDep’06) %8 November %G eng %U /files/syrah/files/iss-hotdep06.pdf %0 Conference Paper %B 7th USENIX Symposium on Operating Systems Design and Implementation (OSDI 2006) %D 2006 %T Fidelity and Yield in a Volcano Monitoring Sensor Network %A Geoff Werner-Allen %A Konrad Lorincz %A Jeff Johnson %A Jonathan Lees %A Welsh, Matt %B 7th USENIX Symposium on Operating Systems Design and Implementation (OSDI 2006) %8 November %G eng %U /files/syrah/files/volcano-osdi06.pdf %0 Journal Article %J Queue %D 2005 %T Beyond relational databases %A Seltzer, Margo %B Queue %I ACM %V 3 %P 50–58 %G eng %N 3 %0 Journal Article %J WORLDS %D 2005 %T Supporting Network Coordinates on PlanetLab. %A Peter Pietzuch %A Jonathan Ledlie %A Seltzer, Margo %B WORLDS %V 5 %P 19–24 %G eng %0 Conference Paper %B Proceedings IEEE 24th Annual Joint Conference of the IEEE Computer and Communications Societies. %D 2005 %T Distributed, secure load balancing with skew, heterogeneity and churn %A Jonathan Ledlie %A Seltzer, Margo %B Proceedings IEEE 24th Annual Joint Conference of the IEEE Computer and Communications Societies. %I IEEE %V 2 %P 1419–1430 %G eng %0 Conference Paper %B International conference on computational science %D 2005 %T A dynamic, data-driven, decision support system for emergency medical services %A Mark Gaynor %A Seltzer, Margo %A Steve Moulton %A Freedman, Jim %B International conference on computational science %I Springer Berlin Heidelberg %P 703–711 %G eng %0 Journal Article %D 2005 %T Modeling the Effects of Memory Hierarchy Performance On Throughput of Multithreaded Processors %A Alexandra Fedorova %A Seltzer, Margo I %A Smith, Michael D %G eng %0 Journal Article %D 2005 %T Performance of multithreaded chip multiprocessors and implications for operating system design %A Alexandra Fedorova %A Seltzer, Margo I %A Small, Christopher A %A Nussbaum, Daniel %G eng %0 Conference Paper %B Proceedings of NetDB 2005 %D 2005 %T Provenance-Aware Sensor Data Storage %A Jonathan Ledlie %A Chaki Ng %A David A. Holland %A Kiran-Kumar Muniswamy-Reddy %A Uri Braun %A Seltzer, Margo %B Proceedings of NetDB 2005 %C Tokyo, Japan %8 April 2005 %G eng %U /files/syrah/files/netdb05.pdf %0 Journal Article %D 2005 %T Provenance-Aware Storage Systems %A Seltzer, Margo %A Kiran-Kumar Muniswamy-Reddy %A David A. Holland %A Uri Braun %A Jonathan Ledlie %8 July 2005 %G eng %U /files/syrah/files/pass-TR-18-05.pdf %N TR-18-05 %0 Book %B Unwired Business: Cases in Mobile Business %D 2005 %T A Prehospital Database System for Emergency Medical Services %A Nada Hashmi %A Mark Gaynor %A Marissa Pepe %A Welsh, Matt %A William W. Tollefsen %A Steven Moulton %A Dan Myung %A Stuart Barnes %A Eusebio Scornavacca %B Unwired Business: Cases in Mobile Business %I IRM Press %G eng %0 Conference Paper %B Fourth International Conference on Information Processing in Sensor Networks (IPSN’05) %D 2005 %T Building up to Macroprogramming: An Intermediate Language for Sensor Networks %A Ryan Newton %A Arvind %A Welsh, Matt %B Fourth International Conference on Information Processing in Sensor Networks (IPSN’05) %8 April %G eng %U /files/syrah/files/tm-ipsn05.pdf %0 Conference Paper %B First IEEE International Workshop on Networking Meets Databases (NetDB’05) %D 2005 %T A Cost-Space Approach to Distributed Query Optimization in Stream Based Overlays %A Jeff Shneidman %A Peter Pietzuch %A Welsh, Matt %A Seltzer, Margo %A Mema Roussopoulos %B First IEEE International Workshop on Networking Meets Databases (NetDB’05) %8 April %G eng %U /files/syrah/files/sbon-netdb05.pdf %0 Conference Paper %B Second USENIX/ACM Symposium on Networked Systems Design and Implementation (NSDI 2005) %D 2005 %T Decentralized, Adaptive Resource Allocation for Sensor Networks %A Geoff Mainland %A David C. Parkes %A Welsh, Matt %B Second USENIX/ACM Symposium on Networked Systems Design and Implementation (NSDI 2005) %8 May %G eng %U /files/syrah/files/sora-nsdi05.pdf %0 Conference Paper %B Fourth International Workshop on Peer-to-Peer Systems (IPTPS’05) %D 2005 %T Evaluating DHT-Based Service Placement for Stream-Based Overlays %A Peter Pietzuch %A Jeff Shneidman %A Jonathan Ledlie %A Welsh, Matt %A Seltzer, Margo %A Mema Roussopoulos %B Fourth International Workshop on Peer-to-Peer Systems (IPTPS’05) %8 February %G eng %U /files/syrah/files/sbon-iptps05.pdf %0 Conference Paper %B International Conference on Information Communication Technologies in Health %D 2005 %T Improving Patient Monitoring and Tracking in Emergency Response %A Tia Gao %A Dan Greenspan %A Welsh, Matt %B International Conference on Information Communication Technologies in Health %8 July %G eng %U /files/syrah/files/monitoring-icicth05.pdf %0 Conference Paper %B Fourth International Conference on Information Processing in Sensor Networks (IPSN’05), Special Track on Platform Tools and Design Methods for Network Embedded Sensors (SPOTS) %D 2005 %T MoteLab: A Wireless Sensor Network Testbed %A Geoff Werner-Allen %A Patrick Swieskowski %A Welsh, Matt %B Fourth International Conference on Information Processing in Sensor Networks (IPSN’05), Special Track on Platform Tools and Design Methods for Network Embedded Sensors (SPOTS) %8 April %G eng %U /files/syrah/files/motelab-spots05.pdf %0 Conference Paper %B International Workshop on Location- and Context-Awareness (LoCA 2005) %D 2005 %T MoteTrack: A Robust, Decentralized Approach to RF-Based Location Tracking %A Konrad Lorincz %A Welsh, Matt %B International Workshop on Location- and Context-Awareness (LoCA 2005) %8 May %G eng %U /files/syrah/files/motetrack-loca05.pdf %0 Journal Article %D 2005 %T Sensor Networks for Medical Care %A Victor Shnayder %A Bor-Rong Chen %A Konrad Lorincz %A Thaddeus R. F. Fulford-Jones %A Welsh, Matt %I Harvard University %8 April %G eng %U /files/syrah/files/codeblue-techrept05.pdf %0 Conference Paper %B 27th IEEE EMBS Annual International Conference %D 2005 %T Vital Signs Monitoring and Patient Tracking Over a Wireless Network %A Tia Gao %A Dan Greenspan %A Welsh, Matt %A Radford R. Juang %A Alex Alm %B 27th IEEE EMBS Annual International Conference %8 September %G eng %U /files/syrah/files/monitoring-embs05.pdf %0 Conference Paper %B Third ACM Conference on Embedded Networked Sensor Systems (SenSys’05) %D 2005 %T Firefly-Inspired Sensor Network Synchronicity with Realistic Radio Effects %A Geoff Werner-Allen %A Geetika Tewari %A Ankit Patel %A Welsh, Matt %A Radhika Nagpal %B Third ACM Conference on Embedded Networked Sensor Systems (SenSys’05) %8 November %G eng %U /files/syrah/files/firefly-sensys05.pdf %0 Journal Article %J ACM SIGMETRICS Performance Evaluation Review %D 2004 %T Using probabilistic reasoning to automate software tuning %A Sullivan, David G %A Seltzer, Margo I %A Avi Pfeffer %B ACM SIGMETRICS Performance Evaluation Review %I ACM %V 32 %P 404–405 %G eng %N 1 %0 Journal Article %J Tr, Harvard University %D 2004 %T Path optimization in stream-based overlay networks %A Peter Pietzuch %A Jeffrey Shneidman %A Welsh, Matt %A Seltzer, Margo %A Mema Roussopoulos %B Tr, Harvard University %G eng %0 Journal Article %D 2004 %T File Classification in Self-* Storage Systems (CMU-PDL-04-101) %A Michael Mesnier %A Eno Thereska %A Daniel Ellard %A Ganger, Gregory R %A Seltzer, Margo %G eng %0 Journal Article %J Third Workshop on Novel Uses of System Area Networks (SAN-3)(Held in conjunction with HPCA-10) %D 2004 %T The case against user-level networking %A Kostas Magoutis %A Seltzer, Margo %A Eran Gabber %B Third Workshop on Novel Uses of System Area Networks (SAN-3)(Held in conjunction with HPCA-10) %G eng %0 Journal Article %D 2004 %T DATA ms %A Ganger, Gregory R %A Seltzer, Margo %G eng %0 Conference Paper %B Proceedings of the 11th workshop on ACM SIGOPS European workshop %D 2004 %T Chip multithreading systems need a new operating system scheduler %A Alexandra Fedorova %A Christopher Small %A Nussbaum, Daniel %A Seltzer, Margo %B Proceedings of the 11th workshop on ACM SIGOPS European workshop %I ACM %P 9 %G eng %0 Journal Article %J Work-in-Progress Reports, 6th Symposium on Operating Systems Design and Implementation %D 2004 %T Implementing an OS Scheduler for Multithreaded Chip Multiprocessors %A Fedorova, A %A Seltzer, M %A Small, C %A Nussbaum, D %B Work-in-Progress Reports, 6th Symposium on Operating Systems Design and Implementation %G eng %0 Conference Paper %B 2004 Workshop on Architectural Support for Security and Anti-Virus %D 2004 %T An Architecture A Day Keeps The Hacker Away %A David A. Holland %A Ada T. Lim %A Margo I. Seltzer %X System security as it is practiced today is a losing battle. In this paper, we outline a possible comprehensive solution for binary-based attacks, using virtual machines, machine descriptions, and randomization to achieve broad heterogeneity at the machine level. This heterogeneity increases the ‘‘cost’’ of broad-based binary attacks to a sufficiently high level that they cease to become feasible. The convergence of several recent technologies appears to make our approach achievable at a reasonable cost, with only moderate run-time overhead. %B 2004 Workshop on Architectural Support for Security and Anti-Virus %C Boston, Massachusetts %8 October 2004 %G eng %U /files/syrah/files/wassa-04.pdf %0 Conference Paper %B 2004 International Conference on Autonomic Computing (ICAC-04) %D 2004 %T File classification in self-* storage systems %A Michael Mesnier %A Eno Thereska %A Daniel Ellard %A Gregory R. Ganger %A Seltzer, Margo %X To tune and manage themselves, file and storage systems must understand key properties (e.g., access pattern, lifetime, size) of their various files. This paper describes how systems can automatically learn to classify the properties of files (e.g., read-only access pattern, short-lived, small in size) and predict the properties of new files, as they are created, by exploiting the strong associations between a file’s properties and the names and attributes assigned to it. These associations exist, strongly but differently, in each of four real NFS environments studied. Decision tree classifiers can automatically identify and model such associations, providing prediction accuracies that often exceed 90%. Such predictions can be used to select storage policies (e.g., disk allocation schemes and replication factors) for individual files. Further, changes in associations can expose information about applications, helping autonomic system components distinguish growth from fundamental change. %B 2004 International Conference on Autonomic Computing (ICAC-04) %8 May 2004 %G eng %U /files/syrah/files/icac-04.pdf %0 Conference Paper %B MobiSys 2004 Workshop on Applications of Mobile Embedded Systems (WAMES 2004) %D 2004 %T CodeBlue: An Ad Hoc Sensor Network Infrastructure for Emergency Medical Care %A David Malan %A Thaddeus Fulford-Jones %A Welsh, Matt %A Steve Moulton %B MobiSys 2004 Workshop on Applications of Mobile Embedded Systems (WAMES 2004) %8 June %G eng %U /files/syrah/files/codeblue-bsn04.pdf %0 Conference Paper %B First USENIX/ACM Symposium on Networked Systems Design and Implementation (NSDI ’04) %D 2004 %T Programming Sensor Networks Using Abstract Regions %A Welsh, Matt %A Geoff Mainland %B First USENIX/ACM Symposium on Networked Systems Design and Implementation (NSDI ’04) %8 March %G eng %U /files/syrah/files/regions-nsdi04.pdf %0 Book %B Ambient Intelligence %D 2004 %T TinyOS: An Operating System for Wireless Sensor Networks %A Philip Levis %A Sam Madden %A Joseph Polastre %A Robert Szewczyk %A Kamin Whitehouse %A Alec Woo %A David Gay %A Hill, Jason %A Welsh, Matt %A Eric Brewer %A David Culler %A W. Weber, J. Rabaey, %A E. Aarts %B Ambient Intelligence %I Springer-Verlag %G eng %U http://www.springeronline.com/sgw/cda/frontpage/0,11855,4-40109-22-36346812-0,00.html %0 Book %B Wireless Sensor Networks: A Systems Perspective %D 2004 %T A Whole-Network Approach to Sensor Network Programming %A Welsh, Matt %A Sam Madden %A Nirupama Bulusu %A Sanjay Jha %B Wireless Sensor Networks: A Systems Perspective %I Artech House %G eng %U http://www.amazon.com/exec/obidos/tg/detail/-/1580538673/qid=1117048636/sr=8-1/ref=sr_8_xs_ap_i1_xgl14/103-8686274-9396647?v=glance&s=books&n=50784 %0 Conference Paper %B 11th ACM SIGOPS European Workshop %D 2004 %T Open Problems in Data Collection Networks %A Jonathan Ledlie %A Jeff Shneidman %A Welsh, Matt %A Mema Roussopoulos %A Seltzer, Margo %B 11th ACM SIGOPS European Workshop %8 September %G eng %U /files/syrah/files/hg-sigops04.pdf %0 Conference Paper %B 26th IEEE EMBS Annual International Conference %D 2004 %T A Portable, Low-Power, Wireless Two-Lead EKG System %A Thaddeus R. F. Fulford-Jones %A Gu-Yeon Wei %A Welsh, Matt %B 26th IEEE EMBS Annual International Conference %8 September %G eng %U /files/syrah/files/ekg-embs04.pdf %0 Conference Paper %B First IEEE International Conference on Sensor and Ad hoc Communications and Networks (SECON) %D 2004 %T A Public-Key Infrastructure for Key Distribution in TinyOS Based on Elliptic Curve Cryptography %A David Malan %A Welsh, Matt %A SMITH, Michael %B First IEEE International Conference on Sensor and Ad hoc Communications and Networks (SECON) %8 October %G eng %U /files/syrah/files/ecc-secon04.pdf %0 Conference Paper %B First International Workshop on Data Management for Sensor Networks (DMSN) %D 2004 %T Region Streams: Functional Macroprogramming for Sensor Networks %A Ryan Newton %A Welsh, Matt %B First International Workshop on Data Management for Sensor Networks (DMSN) %8 August %G eng %U /files/syrah/files/regiment-dmsn04.pdf %0 Conference Paper %B Second ACM Conference on Embedded Networked Sensor Systems (SenSys’04) %D 2004 %T Simulating the Power Consumption of Large-Scale Sensor Network Applications %A Victor Shnayder %A Hempstead, Mark %A Bor-Rong Chen %A Geoff Werner-Allen %A Welsh, Matt %B Second ACM Conference on Embedded Networked Sensor Systems (SenSys’04) %8 November %G eng %U /files/syrah/files/ptossim-sensys04.pdf %0 Conference Paper %B 11th ACM SIGOPS European Workshop %D 2004 %T Using Virtual Markets to Program Global Behavior in Sensor Networks %A Geoff Mainland %A Laura Kang %A Sebastien Lahaie %A David C. Parkes %A Welsh, Matt %B 11th ACM SIGOPS European Workshop %8 September %G eng %U /files/syrah/files/mbm-sigops04.pdf %0 Conference Paper %B Americas Conference on Information Systems (AMCIS 2004) %D 2004 %T Wireless Sensor Network Applications %A Mark Gaynor %A Steve Moulton %A Welsh, Matt %A Austin Rowan %A Ed LaCombe,and John Wynne %B Americas Conference on Information Systems (AMCIS 2004) %8 August %G eng %0 Journal Article %D 2003 %T Overcoming rational manipulation in distributed mechanism implementations %A Jeffrey Shneidman %A David C Parkes %A Seltzer, Margo I %G eng %0 Journal Article %D 2003 %T Strategyproof computing: Systems infrastructures for self-interested parties %A Chaki Ng %A David C Parkes %A Seltzer, Margo I %G eng %0 Conference Paper %B Proceedings of the 4th ACM Conference on Electronic Commerce %D 2003 %T Virtual worlds: Fast and strategyproof auctions for dynamic resource allocation %A Chaki Ng %A David C Parkes %A Seltzer, Margo %B Proceedings of the 4th ACM Conference on Electronic Commerce %I ACM %P 238–239 %G eng %0 Conference Paper %B FAST %D 2003 %T Making the Most Out of Direct-Access Network Attached Storage. %A Kostas Magoutis %A Salimah Addetia %A Alexandra Fedorova %A Seltzer, Margo I %B FAST %G eng %0 Conference Paper %B International Workshop on Peer-to-Peer Systems %D 2003 %T Scooped, again %A Jonathan Ledlie %A Jeff Shneidman %A Seltzer, Margo %A John Huth %B International Workshop on Peer-to-Peer Systems %I Springer Berlin Heidelberg %P 129–138 %G eng %0 Journal Article %J Harvard University Computer Science, Tech. Rep %D 2003 %T Reliability-and capacity-based selection in distributed hash tables %A Jonathan Ledlie %A Jeff Shneidman %A Amis, Matthew %A Michael Mitzenmacher %A Seltzer, Margo %B Harvard University Computer Science, Tech. Rep %G eng %0 Journal Article %D 2003 %T Scooped, Again %A John Huth %A Jonathan Ledlie %A Jeff Shneidman %A Seltzer, Margo %I Springer Verlag %G eng %0 Journal Article %D 2003 %T Making the Most out of Direct-Access Network Attached Storage %A Alexandra Fedorova %A Kostas Magoutis %A Salimah Addetia %A Seltzer, Margo %I USENIX %G eng %0 Journal Article %D 2003 %T Attribute-Based Prediction of File Properties %A Daniel Ellard %A Michael Mesnier %A Eno Thereska %A Gregory R. Ganger %A Seltzer, Margo %X We present evidence that attributes that are known to the file system when a file is created, such as its name, permission mode, and owner, are often strongly related to future properties of the file such as its ultimate size, lifespan, and access pattern. More importantly, we show that we can exploit these relationships to automatically generate predictive models for these properties, and that these predictions are sufficiently accurate to enable optimizations. %C Cambridge, Massachusetts %8 December 2003 %G eng %U /files/syrah/files/tr-14-03.pdf %N 14-03 %0 Journal Article %D 2003 %T DISP: Practical, Efficient, Secure and Fault Tolerant Data Storage for Distributed Systems %A Daniel Ellard %A James Megquier %X We present DISP, a practical, efficient and secure client/server protocol for data storage and retrieval in a distributed environment and show how this protocol can tolerate Byzantine failure. We discuss variations on DISP that can be used as building blocks for different applications, and measure the performance of DISP on commodity hardware. %C Cambridge, Massachusetts %8 December 2003 %G eng %U /files/syrah/files/tr-17-03.pdf %N 17-03 %0 Journal Article %D 2003 %T The File System Interface is an Anachronism %A Daniel Ellard %X Contemporary file systems implement a set of abstractions and semantics that are suboptimal for many (if not most) purposes. The philosophy of using the simple mechanisms of the file system as the basis for a vast array of higher-level mechanisms leads to inefficient and incorrect implementations. We propose several extensions to the canonical file system model, including explicit support for lock files, indexed files, and resource forks, and the benefit of session semantics for write updates. We also discuss the desirability of application-level file system transactions and file system support for versioning. %C Cambridge, Massachusetts %8 January 2003 %G eng %U http://www.eecs.harvard.edu/syrah/papers/tr-15-03/ %N 15-03 %0 Journal Article %D 2003 %T An NFS Trace Player for File System Evaluation %A Ningning Zhu %A Jiawu Chen %A Tzi-cker Chiueh %A Daniel Ellard %X File system traces have been used in simulation of specific design techniques such as disk scheduling, in workload characterization and modeling, and in identifying interesting file access patterns for performance optimizations. Suprisingly they are rarely used to test the correctness and to evaluate the performance of an actuaal file system or server. The main reason is that up until now there did not exist a flexible and easy-to-use trace player that, given an input trace, can properly initialize the test file system and replay the trace to the test system in such a way that respects dependency constraints among file access requests in the trace. This paper describes the design, implementation, and evaluation of an NFS trace play-back tool called FEUT (File system Evaluation Using Traces), which can automatically derive the initial file system image from a trace, speed up or slow down a trace play-back using temporal or spatial scaling, and features a highly efficient implementation that minimizes the CPU and disk I/O overhead during trace play-back. Experiments using a large NFS trace set show that trace-driven file system evaluation can indeed produce substantially different throughput and latency measurements than synthetic benchmarks such as SPECsfs, and FEUT’s trace player is actually more efficient than SPECsfs’s workload generator despite the fact that the former requires more CPU computation and disk I/O accesses. %C Cambridge, Massachusetts %8 December 2003 %G eng %U /files/syrah/files/tr-16-03.pdf %N 16-03 %0 Journal Article %D 2003 %T The Utility of File Names %A Daniel Ellard %A Jonathan Ledlie %A Seltzer, Margo %X For typical workloads and file naming conventions, the size, lifespan, read/write ratio, and access pattern of nearly all files in a file system are accurately predicted by the name given to the file when it is created. We discuss some name-related properties observed in three contemporary NFS workloads, and present a method for automatically creating name-based models to predict interesting file properties of new files, and analyze the accuracy of these models for our workloads. Finally, we show how these predictions can be used as hints to optimize the strategies used by the file system to manage new files when they are created. %C Cambridge, Massachusetts %8 March 2003 %G eng %U /files/syrah/files/tr-05-03.pdf %N 05-03 %0 Journal Article %D 2003 %T Application Performance on the Direct Access File System %A Alexandra Fedorova %A Seltzer, Margo %A Kostas Magoutis %A Salimah Addetia %X The Direct Access File System (DAFS) is a distributed file system built on top of direct-access transports (DAT). Direct-access transports are characterized by using remote direct memory access (RDMA) for data transfer and user-level networking. The motivation behind the DAT-enabled distributed file system architecture is the reduction of the CPU overhead on the I/O data path. In collaboration with Duke University we have created and made available an open-source implementation of DAFS for the FreeBSD platform. In this paper we describe a performance evaluation study of DAFS that was performed with this software. The goal of this work is to determine whether the architecture of DAFS brings any fundamental performance benefits to applications compared to traditional distributed file systems. In our study we compare DAFS to a version of NFS optimized to reduce the I/O overhead. We conclude that DAFS can accomplish superior performance for latency-sensitive applications, outperforming NFS by up to a factor of 2. Bandwidth-sensitive applications do equally well on both systems, unless they are CPU-intensive, in which case they perform better on DAFS. We also found that RDMA is a less restrictive mechanism to achieve copy avoidance than that used by the optimized NFS. %C Cambridge, Massachusetts %8 January 2003 %G eng %U /files/syrah/files/tr-01-03.ps %N 01-03 %0 Conference Paper %B 2003 Large Installation Systems Administration Conference %D 2003 %T New NFS Tracing Tools and Techniques for System Analysis %A Daniel Ellard %A Seltzer, Margo %X Passive NFS traces provide an easy and unobtrusive way to measure, analyze, and gain an understanding of an NFS workload. Historically, such traces have been used primarily by file system researchers in an attempt to understand, categorize, and generalize file system workloads. However, because such traces provide a wealth of detailed information about how a specific system is actually used, they should also be of interest to system administrators. We introduce a new open-source toolkit for passively gathering and summarizing NFS traces and show how to use this toolkit to perform analyses that are difficult or impossible with existing tools. %B 2003 Large Installation Systems Administration Conference %8 October 2003 %G eng %U /files/syrah/files/lisa-03.pdf %0 Conference Paper %B Second International Workshop on Peer-to-Peer Systems (IPTPS 2003) %D 2003 %T Rationality and Self-Interest in Peer to Peer Networks %A Jeffrey Shneidman %A David C. Parkes %X Much of the existing work in peer to peer networking assumes that users will follow prescribed protocols without deviation. This assumption ignores the user’s ability to modify the behavior of an algorithm for self-interested reasons. We advocate a different model in which peer to peer users are expected to be rational and self-interested. This model is found in the emergent fields of Algorithmic Mechanism Design (AMD) and Distributed Algorithmic Mechanism Design (DAMD), both of which introduce game-theoretic ideas into a computational system. We, as designers, must create systems (peer to peer search, routing, distributed auctions, resource allocation, etc.) that allow nodes to behave rationally while still achieving good overall system outcomes. This paper has three goals. The first is to convince the reader that rationality is a real issue in peer to peer networks. The second is to introduce mechanism design as a tool that can be used when designing networks with rational nodes. The third is to describe three open problems that are relevant in the peer to peer setting but are unsolved in existing AMD/DAMD work. In particular, we consider problems that arise when a networking infrastructure contains rational agents. %B Second International Workshop on Peer-to-Peer Systems (IPTPS 2003) %8 February 2003 %G eng %U /files/syrah/files/iptps-03.pdf %0 Conference Paper %B 2003 USENIX Technical Conference Freenix Track %D 2003 %T NFS Tricks and Benchmarking Traps %A Daniel Ellard %A Seltzer, Margo %X We describe two modifications to the FreeBSD 4.6 NFS server to increase read throughput by improving the read-ahead heuristic to deal with reordered requests and stride access patterns. We show that for some stride access patterns, our new heuristics improve end-to-end NFS throughput by nearly a factor of two. We also show that benchmarking and experimenting with changes to an NFS server can be a subtle and challenging task, and that it is often difficult to distinguish the impact of a new algorithm or heuristic from the quirks of the underlying software and hardware with which they interact. We discuss these quirks and their potential effects. %B 2003 USENIX Technical Conference Freenix Track %8 June 2003 %G eng %U /files/syrah/files/freenix-03.pdf %0 Conference Paper %B 2003 USENIX Conference on File and Storage Technologies %D 2003 %T Passive NFS Tracing of Email and Research Workloads %A Daniel Ellard %A Jonathan Ledlie %A Pia Malkani %A Seltzer, Margo %X We present an analysis of a pair of NFS traces of contemporary email and research workloads. We show that although the research workload resembles previously-studied workloads, the email workload is quite different. We also perform several new analyses that demonstrate the periodic nature of file system activity, the effect of out-of-order NFS calls, and the strong relationship between the name of a file and its size, lifetime, and access pattern. %B 2003 USENIX Conference on File and Storage Technologies %8 March 2003 %G eng %U /files/syrah/files/fast-03-tracing.pdf %0 Conference Paper %B Second ACM Workshop on Hot Topics in Networks (HotNets-II) %D 2003 %T Exposing Resource Tradeoffs in Region-Based Communication Abstractions for Sensor Networks %A Welsh, Matt %B Second ACM Workshop on Hot Topics in Networks (HotNets-II) %8 November %G eng %U /files/syrah/files/regions-hotnets03.pdf %0 Conference Paper %B Programming Language Design and Implementation (PLDI 2003) %D 2003 %T The nesC Language: A Holistic Approach to Networked Embedded Systems %A David Gay %A Phil Levis %A Rob von Behren %A Welsh, Matt %A Eric Brewer %A David Culler %B Programming Language Design and Implementation (PLDI 2003) %8 June %G eng %U /files/syrah/files/nesc-pldi-2003.pdf %0 Conference Paper %B First ACM Conference on Embedded Networked Sensor Systems (SenSys’03) %D 2003 %T TOSSIM: Accurate and Scalable Simulation of Entire TinyOS Applications %A Philip Levis %A Nelson Lee %A Welsh, Matt %A David Culler %B First ACM Conference on Embedded Networked Sensor Systems (SenSys’03) %8 November %G eng %U /files/syrah/files/tossim-sensys03.pdf %0 Conference Paper %B Reliable Distributed Systems, 2002. Proceedings. 21st IEEE Symposium on %D 2002 %T Building a reliable mutable file system on peer-to-peer storage %A Stein, Christopher A %A Tucker, Michael J %A Seltzer, Margo I %B Reliable Distributed Systems, 2002. Proceedings. 21st IEEE Symposium on %I IEEE %P 324–329 %G eng %0 Journal Article %D 2002 %T Reliable and fault-tolerant peer-to-peer block storage %A Stein, CA %A Tucker, Michael J %A Seltzer, Margo I %G eng %0 Journal Article %J Harvard University %D 2002 %T Collecting data for one hundred years %A Jeffrey Shneidman %A Choi, Bryan %A Seltzer, Margo %B Harvard University %G eng %0 Conference Paper %B WIESS %D 2002 %T Tree Houses and Real Houses: Research and Commercial Software. %A LoVerso, Susan J %A Seltzer, Margo I %B WIESS %P 55–68 %G eng %0 Journal Article %D 2002 %T Everything You Always Wanted to Know about NFS Trace Analysis, but Were Afraid to Ask Dan Ellard %A Jonathan Ledlie %A Pia Malkani %A Seltzer, Margo %G eng %0 Journal Article %D 2002 %T Clilets: Web Applications with Secure Client-Side Storage %A Fischer, Robert %A Seltzer, Margo I %G eng %0 Journal Article %D 2002 %T Everything you always wanted to know about NFS trace analysis, but were afraid to ask %A Ellard, Dan %A Jonathan Ledlie %A Pia Malkani %A Seltzer, Margo I %G eng %0 Journal Article %D 2002 %T SOS Technical Report: Everything You Always Wanted to Know about NFS Trace Analysis, but Were Afraid to Ask %A Ellard, Dan %A Jonathan Ledlie %A Pia Malkani %A Seltzer, Margo %G eng %0 Conference Paper %B 2002 Workshop on Computer Architecture Education %D 2002 %T On the Design of a New CPU Architecture for Pedagogical Purposes %A Daniel Ellard %A David A. Holland %A Nicholas Murphy %A Seltzer, Margo %X Ant-32 is a new processor architecture designed specifically to address the pedagogical needs of teaching many subjects, including assembly language programming, machine architecture, compilers, operating systems, and VLSI design. This paper discusses our motivation for creating Ant-32 and the philosophy we used to guide our design decisions and gives a high-level description of the resulting design. %B 2002 Workshop on Computer Architecture Education %8 May 2002 %G eng %U /files/syrah/files/wcae-02.pdf %0 Conference Paper %B 2002 USENIX Annual Technical Conference %D 2002 %T Structure and Performance of the Direct Access File System %A Kostas Magoutis %A Salimah Addetia %A Alexandra Fedorova %A Margo I. Seltzer %A Jeffrey S. Chase %A Andrew J. Gallatin %A Richard Kisley %A Rajiv G. Wickremesinghe %A Eran Gabber %X

The Direct Access File System (DAFS) is an emerging industrial standard for network-attached storage. DAFS takes advantage of new user-level network interface standards. This enables a user-level file system structure in which client-side functionality for remote data access resides in a library rather than in the kernel. This structure addresses longstanding performance problems stemming from weak integration of buffering layers in the network transport, kernel-based file systems and applications. The benefits of this architecture include lightweight, portable and asynchronous access to network storage and improved application control over data movement, caching and prefetching.

This paper explores the fundamental performance characteristics of a user-level file system structure based on DAFS. It presents experimental results from an open-source DAFS prototype and compares its performance to a kernel-based NFS implementation optimized for zero-copy data transfer. The results show that both systems can deliver file access throughput in excess of 100 MB/s, saturating network links with similar raw bandwidth. Lower client overhead in the DAFS configuration can improve application performance by up to 40% over optimized NFS when application processing and I/O demands are well-balanced.

%B 2002 USENIX Annual Technical Conference %8 June 2002 %G eng %U /files/syrah/files/usenix-02-dafs.pdf %0 Conference Paper %B 2002 ACM SIGCSE Conference on Computer Science Education %D 2002 %T A New Instructional Operating System %A David A. Holland %A Ada T. Lim %A Margo I. Seltzer %X This paper presents a new instructional operating system, OS/161, and simulated execution environment, System/161, for use in teaching an introductory undergraduate operating systems course. We describe the new system, the assignments used in our course, and our experience teaching using the new system. %B 2002 ACM SIGCSE Conference on Computer Science Education %8 2002 %G eng %U /files/syrah/files/sigcse-02.pdf %0 Conference Paper %B Tenth ACM SIGOPS European Workshop %D 2002 %T Self-Organization in Peer-to-Peer Systems %A Jonathan Ledlie %A Jacob M. Taylor %A Laura Serban %A Seltzer, Margo %X

This paper addresses the problem of forming groups in peer-to-peer (P2P) systems and examines what dependability means in decentralized distributed systems. Much of the literature in this field assumes that the participants form a local picture of global state, yet little research has been done discussing how this state remains stable as nodes enter and leave the system. We assume that nodes remain in the system long enough to benefit from retaining state, but not sufficiently long that the dynamic nature of the problem can be ignored. We look at the components that describe a system’s dependability and argue that next-generation decentralized systems must explicitly delineate the information dispersal mechanisms (e.g., probe, event-driven, broadcast), the capabilities assumed about constituent nodes (bandwidth, uptime, re-entry distributions), and distribution of information demands (needles in a haystack vs. hay in a haystack [lv02gnutella]. We evaluate two systems based on these criteria: Chord [stoica01chord] and a heterogeneous-node hierarchical grouping scheme [ledlie02namequeries]. The former gives a greater than 1 failed request rate under normal P2P conditions and a prototype of the latter a similar rate under more strenuous conditions with an order of magnitude more organizational messages. This analysis suggests several methods to greatly improve the prototype.

%B Tenth ACM SIGOPS European Workshop %8 September 2002 %G eng %U /files/syrah/files/sigops-02.pdf %0 Journal Article %J Concurrency and Computation: Practice and Experience %D 2001 %T HBench: Java: An application-specific benchmarking framework for Java Virtual Machines %A Zhang, Xiaolan %A Seltzer, Margo %B Concurrency and Computation: Practice and Experience %I Wiley Online Library %V 13 %P 775–792 %G eng %N 8-9 %0 Journal Article %J COOTS %D 2001 %T HBench: JGC-An Application-Specific Benchmark Suite for Evaluating JVM Garbage Collector Performance. %A Zhang, Xiaolan %A Seltzer, Margo I %B COOTS %V 1 %P 4–4 %G eng %0 Conference Paper %B USENIX Annual Technical Conference, General Track %D 2001 %T Unifying File System Protection. %A Stein, Christopher A %A Howard, John H %A Seltzer, Margo I %B USENIX Annual Technical Conference, General Track %P 79–90 %G eng %0 Journal Article %D 2001 %T Scaling Filename Queries in a Large-Scale Distributed File System %A Jonathan Ledlie %A Laura Serban %A Dafina Toncheva %X

We have examined the tradeoffs in applying regular and Compressed Bloom filters to the name query problem in distributed file systems and developed and tested a novel mechanism for scaling queries as the network grows large. Filters greatly reduced query messages when using Fan’s "Summary Cache" in web cache hierarchies [fan00summary], a similar albeit smaller, searching problem. We have implemented a testbed that models a distributed file system and run experiments that test various configurations of the system to see if Bloom filters could provide the same kind of improvements. In a realistic system, where the chance that a randomly queried node holds the file being searched for is low, we show that filters always provide lower bandwidth/search and faster time/search, as long as the rates of change of the files stored at the nodes is not extremely high relative to the number of searches. In other words, we confirm the intuition that keeping some state about the contents of the rest of the system will aid in searching as long as acquiring this state is not overly costly and it does not expire too quickly.

The grouping topology we have developed divides n nodes into log(n) groups, each of which has a representative node that aggregates a composite filter for the group. All nodes not in that group use this low-precision filter to weed out whole collections of nodes by probing these filters, only sending a search to be proxied by a member of the group if the probe of the group filter returns positively. Proxied searches are then carried out within a group, where more precise (more bits per file) filters are kept and exchanged between the n/(log(n)) nodes in a group. Experimental results show that both bandwidth/search and time/search are improved with this novel grouping topology.

%C Cambridge, Massachusetts %8 January 2001 %G eng %U /files/syrah/files/tr-03-02.pdf %N 03-02 %0 Conference Paper %B Ninth Workshop on Hot Topics in Operating Systems %D 2001 %T Research Issues in No-Futz Computing %A David A. Holland %A William Josephson %A Kostas Magoutis %A Margo I. Seltzer %A Christopher A. Stein %A Ada Lim %X At the 1999 Workshop on Hot Topics in Operating Systems (HotOS VII), the attendees reached consensus that the most important issue facing the OS research community was "No-Futz" computing; eliminating the ongoing "futzing" that characterizes most systems today. To date, little research ahs been accomplished in this area. Our goal in writing this paper is to focus the research community on the challenges we face if we are to design systems that are truly futz-free, or even low-futz. %B Ninth Workshop on Hot Topics in Operating Systems %C Schoss Elmau, Germany %8 May 2001 %G eng %U /files/syrah/files/hotos-01.ps %0 Journal Article %D 2000 %T OPER AT INGSYSTEMSUPPORTFORMU LTI-USER, REMOTE, GR AP HICALINTERACTION %A Wong, Alexander Ya-li %A Seltzer, Margo %G eng %0 Conference Paper %B USENIX Annual Technical Conference, General Track %D 2000 %T Journaling Versus Soft Updates: Asynchronous Meta-data Protection in File Systems. %A Seltzer, Margo I %A Ganger, Gregory R %A McKusick, Marshall K %A Smith, Keith A %A Soules, Craig AN %A Stein, Christopher A %B USENIX Annual Technical Conference, General Track %P 71–84 %G eng %0 Journal Article %D 2000 %T JOURNALINGVERSUSSOFTU PD AT ES: ASYNCHRONOUSME TA-DA TA PROTECTION INFILESYSTEMS %A Seltzer, Margo I %A Ganger, Gregory R %A McKusick, M Kirk %A Smith, Keith A %A Soules, Craig AN %A Stein, Christopher A %G eng %0 Conference Paper %B ACM SIGMETRICS Performance Evaluation Review %D 2000 %T Improving interactive performance using TIPME %A Yasuhiro Endo %A Seltzer, Margo %B ACM SIGMETRICS Performance Evaluation Review %I ACM %V 28 %P 240–251 %G eng %N 1 %0 Conference Paper %B Unpublished addendum %D 2000 %T Addendum to the paper "Isolation with Flexibility: A Resource Management Framework for Central Servers" %A David G. Sullivan %A Seltzer, Margo %X

As discussed in our 2000 USENIX paper, lottery scheduling’s currency abstraction imposes lower limits on the resource allocations that clients of a system can receive. In our USENIX paper, we proposed one method for overcoming these lower limits. In this addendum to the paper, we discuss a limitation of our original solution to the lower-limits problem, and we propose an alternate solution that overcomes the shortcomings of the original solution.

%B Unpublished addendum %8 June 2000 %G eng %U /files/syrah/files/usenix-00-addendum.ps %0 Conference Paper %B 2000 USENIX Annual Technical Conference %D 2000 %T Isolation with Flexibility: A Resource Management Framework for Central Servers %A David G. Sullivan %A Seltzer, Margo %X

Proportional-share resource management is becoming increasingly important in today’s computing environments. In particular, the growing use of the computational resources of central service providers argues for a proportional-share approach that allows resource principals to obtain allocations that reflect their relative importance. In such environments, resource principals must be isolated from one another to prevent the activities of one principal from impinging on the resource rights of others. However, such isolation limits the flexibility with which resource allocations can be modified to reflect the actual needs of applications. We present extensions to the lottery-scheduling resource management framework that increase its flexibility while preserving its ability to provide secure isolation. To demonstrate how this extended framework safely overcomes the limits imposed by existing proportional-share schemes, we have implemented a prototype system that uses the framework to manage CPU time, physical memory, and disk bandwidth. We present the results of experiments that evaluate the prototype, and we show that our framework has the potential to enable server applications to achieve significant gains in performance.

%B 2000 USENIX Annual Technical Conference %C San Diego, CA %8 June 2000 %G eng %U /files/syrah/files/usenix-00-resmgmt.pdf %0 Conference Paper %B Proceedings of the 3rd conference on USENIX Windows NT Symposium-Volume 3 %D 1999 %T HACC: An architecture for cluster-based web servers %A Zhang, Xiaolan %A Barrientos, Michael %A Chen, J Bradley %A Seltzer, Margo %B Proceedings of the 3rd conference on USENIX Windows NT Symposium-Volume 3 %I USENIX Association %P 16–16 %G eng %0 Journal Article %D 1999 %T HACC: ANARCHITECTUREFOR CLUSTER-BASEDWEBSERVERS %A Zhang, Xiaolan %A Barrientos, Michael %A Chen, J Bradley %A Seltzer, Margo %G eng %0 Journal Article %J Master's thesis, Harvard University and MIT %D 1999 %T I/O automaton model of operating system primitives %A Yates, Daniel %A Lynch, Nancy %A Luchangco, Victor %A Seltzer, Margo %B Master's thesis, Harvard University and MIT %G eng %0 Journal Article %D 1999 %T TERMINALSERVERPERFORM ANCE %A Wong, Alexander Ya-li %A Seltzer, Margo I %G eng %0 Conference Paper %B Hot Topics in Operating Systems, 1999. Proceedings of the Seventh Workshop on %D 1999 %T Tickets and currencies revisited: Extensions to multi-resource lottery scheduling %A Sullivan, David G %A Robert Haas %A Seltzer, Margo I %B Hot Topics in Operating Systems, 1999. Proceedings of the Seventh Workshop on %I IEEE %P 148–152 %G eng %0 Conference Paper %B Hot Topics in Operating Systems, 1999. Proceedings of the Seventh Workshop on %D 1999 %T The case for application-specific benchmarking %A Seltzer, Margo %A Krinsky, David %A Smith, Keith %A Zhang, Xiaolan %B Hot Topics in Operating Systems, 1999. Proceedings of the Seventh Workshop on %I IEEE %P 102–107 %G eng %0 Conference Paper %B USENIX Workshop on Embedded Systems %D 1999 %T Challenges in Embedded Database System Administration. %A Seltzer, Margo I %A Olson, Michael A %B USENIX Workshop on Embedded Systems %G eng %0 Journal Article %J USENIX Conference %D 1999 %T Gizmo databases %A Seltzer, Margo %B USENIX Conference %G eng %0 Conference Paper %B USENIX Annual Technical Conference, FREENIX Track %D 1999 %T Berkeley DB. %A Olson, Michael A %A Bostic, Keith %A Seltzer, Margo I %B USENIX Annual Technical Conference, FREENIX Track %P 183–191 %G eng %0 Journal Article %D 1999 %T We are happy to present the proceedings for the third Symposium on Operating System Design and Implementation, or OSDI III. To paraphrase a line of movie dialog:" Once is happenstance, twice is coincidence, the third time it's a tradition". With the third %A Leach, Paul J %A Seltzer, Margo %G eng %0 Conference Paper %B 1998 Workshop on Computer Architecture Education %D 1999 %T The ANT Architecture – An Architecture for CS1 %A Daniel Ellard %A Penelope Ellard %A James Megquier %A J. Bradley Chen %A Seltzer, Margo %X

A central goal in high-level programming languages, such as those we use to teach introductory computer science courses, is to provide an abstraction that hides the complexity and idiosyncrasies of computer hardware. Although programming languages are very effective at achieving this goal, certain properties of computer hardware cannot be hidden, or are useful to know about. As a consequence, many of the greatest conceptual challenges for beginning programmers arise from a lack of understanding of the basic properties of the hardware upon which computer programs execute.

To address this problem, we have developed a simple virtual machine called ANT for use in our introductory computer science (CS1) curriculum. ANT is designed to be simple enough that a CS1 student can quickly understand it, while at the same time providing an accurate model of many important properties of computer hardware. After two years of experience with ANT in our CS1 course, we believe it is a valuable tool for helping young students understand how programs and data are actually represented in a computer system.

This paper gives a short introduction to ANT. We start with a brief description of the architecture, and then describe how we use ANT in our CS1 course. We include specific examples that demonstrate how ANT can give students intuition about pointers, the representation of data in memory, and other key concepts.

%B 1998 Workshop on Computer Architecture Education %8 February 1999 %G eng %U /files/syrah/files/wcae-98.pdf %0 Journal Article %D 1999 %T A Resource Management Framework for Central Servers %A David G. Sullivan %A Seltzer, Margo %X

Proportional-share resource management is becoming increasingly important in today’s computing environments. In particular, the growing use of the computational resources of central service providers argues for a proportional-share approach that allows resource principals to obtain allocations that reflect their relative importance. In such environments, resource principals must be isolated from one another to prevent the activities of one principal from impinging on the resource rights of others. However, such isolation limits the flexibility with which resource allocations can be modified to reflect the actual needs of applications. We present extensions to the lottery-scheduling resource management framework that increase its flexibility while preserving its ability to provide secure isolation. To demonstrate how this extended framework safely overcomes the limits imposed by existing proportional-share schemes, we have implemented a prototype system that uses the framework to manage CPU time, physical memory, and disk bandwidth. We present the results of experiments that evaluate the prototype, and we show that our framework has the potential to enable server applications to achieve significant gains in performance.

(This is an extended version of our USENIX 2000 paper. It contains additional text and data that could not be included in the published paper due to space constraints.)

%C Cambridge, Massachusetts %8 December 1999 %G eng %U /files/syrah/files/tr-13-99.pdf %N 13-99 %0 Conference Paper %B Seventh Workshop on Hot Topics in Operating Systems %D 1999 %T Tickets and Currencies Revisited: Extending Multi-Resource Lottery Scheduling %A David G. Sullivan %A Robert Haas %A Seltzer, Margo %X

Lottery scheduling’s ticket and currency abstractions provide a resource management framework that allows for both flexible allocation and insulation between groups of processes. We propose extensions to this framework that enable greater flexibility while preserving the ability to isolate groups of processes. In particular, we present a mechanism that allows processes to modify their own resource rights by exchanging resource-specific tickets with other processes. Ticket exchanges limit the effects of the changed allocations to the participants in the exchange, and they allow applications to coordinate with each other in ways that are mutually beneficial. Application-specific "negotiators" can be used to initiate exchanges based on an application’s quality-of-service requirements and the current state of the system. We also propose flexible access controls for currencies through extensible "brokers" that solve such problems as the inability of users isolated by currencies to renice background jobs. Finally, we suggest using extensibility to allow users to install specialized allocation mechanisms for their processes. Together, these extensions enable an application-centered approach to resource management that is both secure and effective.

%B Seventh Workshop on Hot Topics in Operating Systems %C Rio Rico, Arizona %8 March 1999 %G eng %U /files/syrah/files/hotos-99-lotsched.pdf %0 Journal Article %J IEEE concurrency %D 1998 %T MiSFIT: Constructing safe extensible systems %A Christopher Small %A Seltzer, Margo %B IEEE concurrency %I IEEE %V 6 %P 34–41 %G eng %N 3 %0 Conference Paper %B ACM SIGMETRICS Performance Evaluation Review %D 1998 %T A self-scaling and self-configuring benchmark for Web servers %A Stephen Manley %A Seltzer, Margo %A Courage, Michael %B ACM SIGMETRICS Performance Evaluation Review %I ACM %V 26 %P 270–271 %G eng %N 1 %0 Book %D 1998 %T Vitis Propulsion: Theory and Practice %A Baker, Mary %A Seltzer, Margo %I Stanford University %G eng %0 Journal Article %D 1998 %T Using PS %A Aversa, Rocco %A Mazzeo, Antonino %A Mazzocca, Nicola %A Villano, Umberto %A Devarakonda, Murthy %A Christopher Small %A Seltzer, Margo I %G eng %0 Conference Paper %B 6.857 Final Project %D 1998 %T Receipt-Free Secure Elections %A Daniel Ellard %A David Alpert %A Ognjen Kavazovic %A Marc Scheff %X This paper begins by describing Benaloh and Tuinstra’s protocol for receipt-free elections, and then shows how this protocol, although correct, makes unrealistic assumptions and is impractical to implement. We then show how to modify this protocol to make it very practical to implement, by adding an additional "uncoercible" party to the protocol. %B 6.857 Final Project %8 December 1998 %G eng %U /files/syrah/files/6857-98.pdf %0 Journal Article %D 1997 %T Does Systems Research Measure Up? %A Small, Christopher A %A Ghosh, Narendra %A Saleeb, Hany %A Seltzer, Margo I %A Smith, Keith %G eng %0 Conference Paper %B ACM SIGMETRICS Performance Evaluation Review %D 1997 %T File system aging�increasing the relevance of file system benchmarks %A Smith, Keith A %A Seltzer, Margo I %B ACM SIGMETRICS Performance Evaluation Review %I ACM %V 25 %P 203–213 %G eng %N 1 %0 Journal Article %D 1997 %T Issues in extensible operating systems %A Seltzer, Margo I %A Yasuhiro Endo %A Small, Christopher A %A Smith, Keith A %G eng %0 Conference Paper %B Proceedings of the annual conference on USENIX Annual Technical Conference %D 1997 %T Measuring computer systems: how to measure performance %A Seltzer, Margo %A Brown, Aaron %B Proceedings of the annual conference on USENIX Annual Technical Conference %I USENIX Association %P 27–27 %G eng %0 Journal Article %J Proceedings of the 17th IEEE International Conference on Distributed Computing Systems %D 1997 %T An analysis of geographical push-caching %A James Gwertzman %A Seltzer, Margo %B Proceedings of the 17th IEEE International Conference on Distributed Computing Systems %G eng %0 Journal Article %J First USENIX Windows NT Workshop %D 1997 %T Measuring Windows NT�Possibilities and Limitations %A Yasuhiro Endo %A Seltzer, Margo I %B First USENIX Windows NT Workshop %G eng %0 Journal Article %J ACM SIGMETRICS Performance Evaluation Review %D 1997 %T Operating system benchmarking in the wake of lmbench: A case study of the performance of NetBSD on the Intel x86 architecture %A Brown, Aaron B %A Seltzer, Margo I %B ACM SIGMETRICS Performance Evaluation Review %I ACM %V 25 %P 214–224 %G eng %N 1 %0 Journal Article %D 1997 %T The mug-shot search problem %A Baker, Ellie %A Seltzer, Margo I %G eng %0 Conference Paper %B 1997 USENIX Symposium on Internet Technologies and Systems %D 1997 %T Web Facts and Fantasy %A Stephen Manley %A Seltzer, Margo %X There is a great deal of research about improving Web server performance and building better, faster servers, but little research in characterizing servers and the load imposed upon them. While some tremendously popular and busy sites, such as netscape.com, playboy.com, and altavista.com, receive several million hits per day, most servers are never subjected to loads of this magnitude. This paper presents the analysis of internet Web server logs for a variety of different types of sites. We present a taxonomy of the different types of Web sites and characterize their access patterns and, more importantly, their growth. We then use our server logs to address some common perceptions about the Web. We show that, on a variety of sites, contrary to popular belief, the use of CGI does not appear to be increasing and that long latencies are not necessarily due to server loading. We then show that, as expected, persistent connections are generally useful, but that dynamic time-out intervals may be unnecessarily complex and that allowing multiple persistent connections per client may actually hinder resource utilization compared to allowing only a single persistent connection. %B 1997 USENIX Symposium on Internet Technologies and Systems %8 December 1997 %G eng %U /files/syrah/files/sits-97.ps %0 Conference Paper %B Sixth Workshop on Hot Topics in Operating Systems %D 1997 %T Self-Monitoring and Self-Adapting Operating Systems %A Seltzer, Margo %A Christopher Small %X

Extensible operating systems allow applications to modify kernel behavior by providing mechanisms for application code to run in the kernel address space. Extensibility enables a system to efficiently support a broader class of applications than is currently supported. This paper discusses the key challenge in making extensible systems practical: determining which parts of the system need to be extended and how. The determination of which parts of the system need to be extended requires self-monitoring, capturing a significant quantity of data about the performance of the system. Determing how to extend the system requires self-adaptation. In this paper, we describe how an extensible operating system (VINO) can use in situ simulation to explore the efficacy of policy changes. This automatic exploration is applicable to other extensible operating systems and can make these systems self-adapting to workload demands.

%B Sixth Workshop on Hot Topics in Operating Systems %8 May 1997 %G eng %U /files/syrah/files/hotos-97.ps %0 Conference Paper %B 1997 Conference on Object-Oriented Technologies and Systems %D 1997 %T A Tool For Constructing Safe Extensible C++ Systems %A Christopher Small %X

The boundary between application and system is becoming increasingly permeable. Extensible applications, such as web browsers, database systems, and operating systems, demonstrate the value of allowing end-users to extend and modify the behavior of what was formerly considered to be a static, inviolate system. Unfortunately, flexibility often comes with a cost: systems unprotected from misbehaved end-user extensions are fragile and prone to instability.

Object-oriented programming models are a good fit for the development of this kind of system. An extension can be designed as a refinement of an existing class, and loaded into a running system. In our model, when code is downloaded into the system, it is used to replace a virtual function on an existing C++ object. Because our tool is source-language neutral, it can be used to build safe extensible systems written in other languages as well.

There are three methods commonly used to make end-user extensions safe: restict the extension language (e.g., Java), interpret the extension language (e.g., Tcl), or combine run-time checks with a trusted environment. The third technique is the one discussed here; it offers the twin benefits of the flexibility to implement extensions in an unsafe language, such as C++, and the performance of compiled code.

MiSFIT, the Minimal i386 Software Fault Isolation Tool, can be used as the central component of a tool set for building safe extensible systems in C++. MiSFIT transforms C++ code, compiled by g++, into safe binary code. Combined with a runtime support library, the overhead of MiSFIT is an order of magnitude lower than the overhead of interpreted Java, and permits safe extensible systems to be written in C++.

%B 1997 Conference on Object-Oriented Technologies and Systems %8 June 1997 %G eng %U /files/syrah/files/coots-97.ps %0 Conference Paper %B USENIX annual technical conference %D 1996 %T A Comparison of FFS Disk Allocation Policies. %A Smith, Keith A %A Seltzer, Margo I %B USENIX annual technical conference %P 15–26 %G eng %0 Journal Article %J Proceedings of the... USENIX Technical Conference %D 1996 %T James Gwertzman, Microsoft Corporation %A Seltzer, Margo %B Proceedings of the... USENIX Technical Conference %I Usenix Association %P 141 %G eng %0 Conference Paper %B Proceedings of the 41st IEEE International Computer Conference %D 1996 %T People, places, and things: the next generation Web %A James Gwertzman %A Seltzer, Margo %B Proceedings of the 41st IEEE International Computer Conference %I IEEE Computer Society %P 65 %G eng %0 Journal Article %J ACM Transactions on Computer Systems (TOCS) %D 1996 %T The measured performance of personal computer operating systems %A Chen, J Bradley %A Yasuhiro Endo %A Chan, Kee %A Mazieres, David %A Dias, Antonio %A Seltzer, Margo %A Smith, Michael D %B ACM Transactions on Computer Systems (TOCS) %I ACM %V 14 %P 3–40 %G eng %N 1 %0 Journal Article %J ACM SIGOPS Operating Systems Review %D 1996 %T Using latency to evaluate interactive system performance %A Yasuhiro Endo %A Wang, Zheng %A Chen, J Bradley %A Seltzer, Margo I %B ACM SIGOPS Operating Systems Review %V 30 %P 185–199 %G eng %N si %0 Conference Paper %B 1996 USENIX Annual Technical Conference %D 1996 %T A Comparison of OS Extension Technologies %A Christopher Small %A Seltzer, Margo %X The current trend in operating systems research is to allow applications to dynamically extend the kernel to improve application performance or extend functionality, but the most effective approach to extensibility remains unclear. Some systems use safe languages to permit code to be downloaded directly into the kernel; other systems provide in-kernel interpreters to execute extension code; still others use software techniques to ensure the safety of kernel extensions. The key characteristics that distinguish these systems are the philosophy behind extensibility and the technology used to implement extensibility. This paper presents a taxonomy of the types of extensions that might be desirable in an extensible operating system, evaluates the performance cost of various extension techniques currently being employed, and compares the cost of adding a kernel extension to the benefit of having the extension in the kernel. Our results show that compiled technologies (e.g. Modula-3 and software fault isolation) are good candidates for implementing general-purpose kernel extensions, but that the overhead of interpreted languages is sufficiently high that they are inappropriate for this use. %B 1996 USENIX Annual Technical Conference %8 January 1996 %G eng %U /files/syrah/files/usenix-96-bakeoff.ps %0 Conference Paper %B 1996 Workshop on Compiler Support for Systems Software %D 1996 %T Symbiotic Systems Software %A Seltzer, Margo %A Christopher Small %A Michael D. Smith %X

Historically, advances in compiler technology have been driven by the characteristics of applications, particularly those that comprise the SPEC benchmark suite. To achieve high performance, many of the most promising compilation techniques rely on accurate profile information to direct the optimization and instruction scheduling process. Prior studies have shown that for these applications it is possible to generate representative data sets that are suitable for profile-based compilation algorithms. Though we would like to apply sophisticated compiler techniques to operating systems (and eventually other multi-threaded software systems such as database management systems), we are presented with two significant problems. First, the process of profiling the operating system is more difficult than profiling an application, because of the complexity of the operating system and the interaction between the operating system and its applications. Second, even when one can profile the operating system, there seems to be no general agreement of what constitutes a representative data set for such a large and complex system. This implies that the optimization of the operating system in isolation is not the correct approach. A better approach for improving the combined system and application performance is to allow the compiler to symbiotically optimize the operating system in the context of the applications that use it. This paper proposes a structuring of the compiler and operating system so that it is possible to perform this type of symbiotic, cross-address space compile-time optimization.

%B 1996 Workshop on Compiler Support for Systems Software %8 February 1996 %G eng %U /files/syrah/files/wcsss-96.ps %0 Conference Paper %B 1996 USENIX Annual Technical Conference %D 1996 %T World Wide Web Cache Consistency %A James Gwertzman %A Seltzer, Margo %X The bandwidth demands of the World Wide Web continue to grow at a hyper-exponential rate. Given this rocketing growth, caching of web objects as a means to reduce network bandwidth consumption is likely to be a necessity in the very near future. Unfortunately, many Web caches do not satisfactorily maintain cache consistency. This paper presents a survey of contemporary cache consistency mechanisms in use on the Internet today and examines recent research in Web cache consistency. Using trace-driven simulation, we show that a weak cache consistency protocol (the one used in the Alex ftp cache) reduces network bandwidth consumption and server load more than either time-to-live fields or an invalidation protocol and can be tuned to return stale data less than 5% of the time. %B 1996 USENIX Annual Technical Conference %8 January 1996 %G eng %U /files/syrah/files/usenix-96-webcache.ps %0 Conference Paper %B 1996 Symposium on Operating Systems Design and Implementation %D 1996 %T Dealing with Disaster: Surviving Misbehaved Kernel Extensions %A Seltzer, Margo %A Yasuhiro Endo %A Christopher Small %A Keith A. Smith %X

Today’s extensible operating systems allow applications to modify kernel behavior by providing mechanisms for application code to run in the kernel address space. The advantage of this approach is that it provides improved application flexibility and performance; the disadvantage is that buggy or malicious code can jeopardize the integrity of the kernel. It has been demonstrated that it is feasible to use safe languages, software fault isolation, or virtual memory protection to safeguard the main kernel. However, such protection mechanisms do not address the full range of problems, such as resource hoarding, that can arise when application code is introduced into the kernel. In this paper, we present an analysis of extension mechanisms in the VINO kernel. VINO uses software fault isolation as its safety mechanism and a lightweight transaction system to cope with resource-hoarding. We explain how these two mechanisms are sufficient to protect against a large class of errant or malicious extensions, and we quantify the overhead that this protection introduces. We find that while the overhead of these techniques is high relative to the cost of the extensions themselves, it is low relative to the benefits that extensibility brings.

%B 1996 Symposium on Operating Systems Design and Implementation %C Seattle, WA %8 October 1996 %G eng %U /files/syrah/files/osdi-96.ps %0 Journal Article %D 1995 %T Scheduler activations on BSD: sharing thread management between kernel and application %A Small, Christopher A %A Seltzer, Margo I %G eng %0 Conference Paper %B Proceedings of the USENIX 1995 Technical Conference Proceedings %D 1995 %T File system logging versus clustering: A performance comparison %A Seltzer, Margo %A Smith, Keith A %A Hari Balakrishnan %A Chang, Jacqueline %A McMains, Sara %A Padmanabhan, Venkata %B Proceedings of the USENIX 1995 Technical Conference Proceedings %I USENIX Association %P 21–21 %G eng %0 Journal Article %J ACM SIGOPS Operating Systems Review %D 1995 %T Autonomous replication across wide-area internetworks %A James Gwertzman %A Seltzer, Margo %B ACM SIGOPS Operating Systems Review %V 29 %P 234 %G eng %N 5 %0 Conference Paper %B Hot Topics in Operating Systems, 1995.(HotOS-V), Proceedings., Fifth Workshop on %D 1995 %T The case for geographical push-caching %A Gwertzman, James S %A Seltzer, Margo %B Hot Topics in Operating Systems, 1995.(HotOS-V), Proceedings., Fifth Workshop on %I IEEE %P 51–55 %G eng %0 Journal Article %D 1995 %T MOSS: A Mobile Operating System Substrate %A Chen, J Bradley %A Kung, HT %A Seltzer, Margo I %G eng %0 Journal Article %D 1995 %T The Impact of Operating System Structure on Personal Computer Performance %A Chen, J Bradley %A Endo, Yashuhiro %A Chan, Kee %A Mazières, David %A Dias, Antonio %A Seltzer, Margo I %A Smith, Michael D %G eng %0 Conference Paper %B ACM SIGOPS Operating Systems Review %D 1995 %T The measured performance of personal computer operating systems %A Chen, J Bradley %A Yasuhiro Endo %A Chan, Kee %A Mazieres, David %A Dias, Antonio %A Seltzer, Margo %A Smith, Michael D %B ACM SIGOPS Operating Systems Review %I ACM %V 29 %P 299–313 %G eng %N 5 %0 Conference Paper %B USENIX %D 1995 %T Heuristic Cleaning Algorithms in Log-Structured File Systems. %A Blackwell, Trevor %A Harris, Jeffrey %A Seltzer, Margo I %B USENIX %P 277–288 %G eng %0 Journal Article %D 1995 %T The Case for Extensible Operating Systems %A Seltzer, Margo %A Christopher Small %A Keith A. Smith %X

Many of the performance improvements cited in recent operating systems research describe specific enhancements to normal operating system functionality that improve performance in a set of designated test cases. Global changes of this sort can improve performance for one application, at the cost of decreasing performance for others. We argue that this flurry of global kernel tweaking is an indication that our current operating system model is inappropriate. Existing interfaces do not provide the flexibility to tune the kernel on a per-application basis, to suit the variety of applications that we now see.

We have failed in the past to be omniscient about future operating system requirements; there is no reason to believe that we will fare any better designing a new, fixed kernel interface today. Instead, the only general-purpose solution is to build an operating system interface that is easily extended. We present a kernel framework designed to support the application-specific customization that is beginning to dominate the operating system literature. We show how this model enables easy implementation of many of the earlier research results. We then analyze two specific kernel policies: page read-ahead and locking-granting. We show that application-control over read-ahead policy produces performance improvements of up to 16%. We then show how application-control over the lock-granting policy can choose between fairness and response time. Reader priority algorithms produce lower read response time at the cost of writer starvation. FIFO algorithms avoid the starvation problem, but increase read response time.

%C Cambridge, Massachusetts %8 1995 %G eng %U /files/syrah/files/tr-16-95.ps %N 16-95 %0 Conference Paper %B 1995 International Workshop on Object Orientation in Operating Systems %D 1995 %T Structuring the Kernel as a Toolkit of Extensible, Reusable Components %A Christopher Small %A Seltzer, Margo %X Applications often require functionality that is implemented in the kernel, but is not directly available to the user level. While extensible operating systems allow kernel functionality to be augmented, we believe that the emphasis on extensibility is misplaced. Applications should be able to reuse kernel code directly and the emphasis should be placed on designing a kernel with that reuse in mind. The advantage of structuring the kernel as a set of reusable, extensible tools is that applications can avoid re-implementing functionality that is already present in the kernel. This will lead to smaller applications, fewer lines of total code, and a more unified computing environment that will be easier to maintain and perform better. %B 1995 International Workshop on Object Orientation in Operating Systems %8 August 1995 %G eng %U /files/syrah/files/iwooos-95.ps %0 Journal Article %D 1994 %T File layout and file system performance %A Smith, Keith %A Seltzer, Margo I %G eng %0 Journal Article %D 1994 %T Vino: An integrated platform for operating system and database research %A Small, Christopher A %A Seltzer, Margo I %G eng %0 Journal Article %D 1994 %T VINO: The 1994 Fall Harvest %A Seltzer, Margo I %A Yasuhiro Endo %A Small, Christopher A %A Smith, Keith %G eng %0 Journal Article %D 1994 %T Infrastructure for Research towards Ubiquitous Information Systems %A Grosz, Barbara J %A Kung, HT %A Seltzer, Margo I %A Shieber, Stuart Merrill %A Smith, Michael D %G eng %0 Journal Article %D 1994 %T VINO: The 1994 fall harvest %A Yasuhiro Endo %A Seltzer, Margo %A James Gwertzman %A Christopher Small %A Smith, Keith A %A Tang, Diane %I Citeseer %G eng %0 Journal Article %J Proceedings of the First Annual Conference on Telecommunications in Massachussetts %D 1994 %T An Experimental Flow-Controlled Multicast ATM Switch %A Blackwell, T %A Chan, K %A Chang, K %A Charuhas, T %A Karp, B %A Kung, HT %A Lin, D. %A Morris, R. %A Seltzer, M %A M. Smith %A others %B Proceedings of the First Annual Conference on Telecommunications in Massachussetts %G eng %0 Conference Paper %B USENIX Winter %D 1993 %T An Implementation of a Log-Structured File System for UNIX. %A Seltzer, Margo I %A Bostic, Keith %A McKusick, Marshall K %A Staelin, Carl %A others %B USENIX Winter %P 307–326 %G eng %0 Conference Paper %B Data Engineering, 1993. Proceedings. Ninth International Conference on %D 1993 %T Transaction support in a log-structured file system %A Seltzer, Margo I %B Data Engineering, 1993. Proceedings. Ninth International Conference on %I IEEE %P 503–510 %G eng %0 Journal Article %D 1993 %T Evolving line drawings %A Baker, Ellie %A Seltzer, Margo I %G eng %0 Book %D 1992 %T LIBTP: Portable, modular transactions for UNIX %A Seltzer, Margo Ilene %A Olson, Michael Allen %I Electronics Research Laboratory, College of Engineering, University of California %G eng %0 Conference Paper %B ACM SIGPLAN Notices %D 1992 %T Non-volatile memory for fast, reliable file systems %A Baker, Mary %A Asami, Satoshi %A Deprit, Etienne %A Ouseterhout, John %A Seltzer, Margo %B ACM SIGPLAN Notices %I ACM %V 27 %P 10–22 %G eng %N 9 %0 Conference Paper %B USENIX Winter %D 1991 %T A New Hashing Package for UNIX. %A Seltzer, Margo I %A Yigit, Ozan %B USENIX Winter %P 173–184 %G eng %0 Journal Article %J Proceedings of the SeventhInternational Conference ollDataEngineering %D 1991 %T Read optimized file system designs: a performance evahiation %A Seltzer, Margo I %A Stonebraker, Michael %B Proceedings of the SeventhInternational Conference ollDataEngineering %P 602–6l1 %G eng %0 Conference Paper %B Data Engineering, 1991. Proceedings. Seventh International Conference on %D 1991 %T Read Optimized file system designs: A performance evaluation %A Seltzer, Margo %A Stonebraker, Michael %B Data Engineering, 1991. Proceedings. Seventh International Conference on %I IEEE %P 602–611 %G eng %0 Journal Article %J Proceedings of the Winter 1990 USENIX Technical Conference %D 1990 %T Disk scheduling revisited %A Seltzer, Margo %A Chen, Peter %A Ousterhout, John %B Proceedings of the Winter 1990 USENIX Technical Conference %I Washington, DC %P 313–323 %G eng %0 Conference Paper %B VLDB %D 1990 %T Transaction support in read optimized and write optimized file systems %A Seltzer, Margo I %A Stonebraker, Michael %B VLDB %P 174–185 %G eng %0 Book %D 1989 %T Parallelism in XPRS %A Stonebraker, Michael %A Aoki, Paul Masami %A Seltzer, Margo Ilene %A others %I Electronics Research Laboratory, College of Engineering, University of California %G eng