People | Publications | Tech Reports | Theses | Internal
We use experimentation and analysis to understand how systems interact, how the world is changing, and how file systems and storage systems should evolve.
Current work includes:
- Evaluation of design and protocol trade-offs in distributed network storage.
- Develoment of tools for the modeling and simulation of network storage.
- Provenance-aware storage for scientific applications.
Our previous work (papers below) has aimed to answer the following questions [abstracts hyperlinked]:
-
The Self-Organizing Storage (SOS) project: how can storage systems tune themselves to their workload? [ICAC'04, FAST'03, LISA'03, FREENIX'03]
-
How can distributed file systems take advantage of zero-copy (RDMA) architectures? [Kostas' thesis, Salimah's thesis, FAST'03, USENIX'02, BSDCon'02]
-
How should the performance of file systems be evaluated? [Keith's thesis]
-
How do soft updates and journaling differ in performance and semantics? [USENIX'01, USENIX'00]
-
How do the different FFS allocation algorithms compare? [USENIX'96]
-
How do clustering and file system logging compare? [USENIX'95]
This research is part of the SYRAH group.
People
Current
- Professor Margo Seltzer
- Elaine Angelino
- Uri Braun
- David Holland
- Peter Macko
- Daniel Margo
- Nicholas Murphy
- Robin Smogor
Alumni
- Kiran-Kumar Muniswamy-Reddy. PhD 2010, Foundations for Provenance-Aware Systems
(Now at Amazon). - Lex Stein. PhD 2007, Adaptive Parallel Computation for Heterogeneous Processors
(Now at Microsoft Research Asia). - Daniel J. Ellard. PhD 2004, Trace-Based Analyses and Optimizations for Network Storage Servers
(Now at Network Appliance, Advanced Technology Group). - Kostas Magoutis. PhD 2003, Exploiting Direct-Access Networking in Network Attached Storage Systems
(Now at IBM Research, TJ Watson). - Keith A. Smith. PhD 2000, Workload-Specific File System Benchmarks
(Now at Network Appliance, Advanced Technology Group). - Salimah Addetia. ME 2002, Caching in DAFS
(Now at Network Appliance).
Publications
- Lex Stein, David Holland, Margo Seltzer, and Zheng Zhang.
Can a file system virtualize processors? (PDF)
In Proceedings of the 1st EuroSys Workshop on Virtualization for HPC,
Lisbon, Portugal, March 2007. - Lex Stein.
Stupid file systems are better. (PDF, HTML)
In Proceedings of the 10th ACM Workshop on Hot Topics in Operating Systems (HotOS-X),
Santa Fe, New Mexico, May 2005 - Jonathan Ledlie, Chaki Ng, David A. Holland, Kiran-Kumar Muniswamy-Reddy, Uri Braun, and Margo Seltzer.
Provenance-Aware Sensor Data Storage. (PDF)
In Proceedings of NetDB 2005, Tokyo, Japan, April 2005. - Michael Mesnier, Eno Thereska, Daniel Ellard, Gregory R. Ganger and Margo Seltzer.
File classification in self-* storage systems. (PDF, Postscript)
Carnegie Mellon University Technical Report CMU-PDL-04-101, January 2004.
Proceedings of the International Conference on Autonomic Computing (ICAC-04). - Daniel Ellard and Margo Seltzer.
New NFS Tracing Tools and Techniques for System Analysis (PDF, Postscript)
Appears in Proceedings of the 17th Large Installation System Administration Conference (LISA'03), pages 73-85, San Diego, CA. October 2003. (note that the page numbers in the on-line version are not identical to the printed proceedings)- Source code for the tools described in the paper.
- Slides from the presentation (PDF, HTML).
- Daniel Ellard and Margo Seltzer.
NFS Tricks and Benchmarking Traps (PDF, HTML)
Appears in Proceedings of the 2003 USENIX Annual Technical Conference, FREENIX Track, pages 101-114, San Antonio, Texas. June 2003.- Source code for the changes described in the paper.
- Slides from the presentation (PDF, HTML).
- Errata: there is a typo in Figure 8 of the published paper. The caption should read ``... scsi1 runs 60-70% faster for all tests when cursors are enabled.ide1 is only 50% faster for the 2-stride test...'' This on-line version of the paper is correct.
- Kostas Magoutis, Salimah Addetia, Alexandra Fedorova, Margo I. Seltzer.
Making the Most out of Direct Access Network-Attached Storage, (PDF)
In Proceedings of Second USENIX Conference on File and Storage Technologies (FAST'03), San Francisco, CA, March 31-April 2, 2003.
- Daniel Ellard, Jonathan Ledlie, Pia Malkani and Margo Seltzer.
Passive NFS Tracing of Email and Research Workloads (PDF, HTML)
Appears in Proceedings of the Second USENIX File and Storage Technologies Conference (FAST'03), pages 203-216, San Francisco, California. March, 2003. - Lex Stein, Michael J. Tucker, and Margo I. Seltzer.
Building a Reliable Mutable File System on Peer-to-peer Storage, (Postscript)
In Proceedings of the International Workshop on Reliable Peer-to-peer Distributed Systems, Osaka, Japan, October, 2002.
- Kostas Magoutis, Salimah Addetia, Alexandra Fedorova, Margo I. Seltzer, Jeffrey S. Chase, Andrew J. Gallatin, Richard Kisley, Rajiv G. Wickremesinghe, Eran Gabber.
Structure and Performance of the Direct Access File System, (PDF)
In Proceedings of 2002 USENIX Annual Technical Conference, Monterey, CA, June 9-14, 2002.
- Kostas Magoutis.
Design and Implementation of a Direct Access File System (DAFS) Kernel Server for FreeBSD (Postscript, PDF),
Appears in Proceedings of USENIX BSDCon 2002 Conference, San Franscisco, CA, February 11-14, 2002.
- Kostas Magoutis.
The Optimistic Direct Access File System: Design and Network Interface Support (Postscript, PDF)
Appears in Proceedings of Workshop on Novel Uses of System Area Networks (Held in conjunction with HPCA-8), Cambridge, MA, February 2, 2002.
- Daniel Ellard.
Slides from the Birds-of-a-Feather session on File System Tracing (PDF, Postscript)
First USENIX File and Storage Technologies Conference (FAST'02), Monterey, California. January, 2002. - Christopher A. Stein, John H. Howard, and Margo I. Seltzer.
Unifying File System Protection (PDF, Postscript),
Appears in Proceedings of 2001 USENIX Annual Technical Conference, Boston, MA, 2001. - Elizabeth Shriver, Eran Gabber, Lan Huang, and Christopher A. Stein.
Storage Management for Web Proxies (PDF, Postscript, HTML),
Appears in Proceedings of the 2001 USENIX Annual Technical Conference, Boston, MA, 2001. - Margo I. Seltzer, Gregory R. Ganger, M. Kirk McKusick, Keith A. Smith, Craig A.N. Soules, and Christopher A. Stein.
Journaling versus Soft Updates: Asynchronous Meta-data Protection in File Systems (Postscript, PDF),
Appears in Proceedings of the 2000 USENIX Annual Technical Conference, San Diego, CA, June, 2000. - Keith A. Smith and Margo I. Seltzer.
File System Aging - Increasing the Relevance of File System Benchmarks (Postscript),
Appears in Proceedings of the 1997 ACM SIGMETRICS International Conference on Measurement and Modeling of Computer Systems, June 1997, Seattle, WA. - Keith A. Smith and Margo Seltzer.
A Comparison of FFS Disk Allocation Algorithms (Postscript),
Appears in Proceedings of the 1996 USENIX Annual Technical Conference, January 1996, San Diego, CA. - Seltzer, M., Smith, K., Balakrishnan, H., Chang, J., McMains, S., and Padmanabhan, V.
File System Logging versus Clustering: A Performance Comparison (Postscript),
Appears in Proceedings of the 1995 USENIX Annual Technical Conference, January 1995, New Orleans, LA, pp. 249--264.
Technical Reports
- Daniel Ellard, Michael Mesnier, Eno Thereska, Gregory R. Ganger, and Margo Seltzer.
Attribute-Based Prediction of File Properties. (PDF, Postscript),
Harvard Computer Science Technical Report TR-14-03, December, 2003. - Ningning Zhu, Jiawu Chen, Tzi-cker Chiueh, and Daniel Ellard.
An NFS Trace Player for File System Evaluation. (PDF, Postscript),
Harvard Computer Science Technical Report TR-16-03, December, 2003. - Daniel Ellard, Jonathan Ledlie and Margo Seltzer.
The Utility of File Names (PDF, Postscript),
Harvard Computer Science Technical Report TR-05-03, March, 2003. - Daniel Ellard, Jonathan Ledlie, Pia Malkani and Margo Seltzer.
Everything You Always Wanted to Know About NFS Trace Analysis, But Were Afraid to Ask (PDF, Postscript),
Harvard Computer Science Technical Report TR-06-02, June, 2002. - Salimah Addetia.
"User-level Client-side Caching for DAFS", Harvard University TR-14-01, (Postscript, PDF). - Keith A. Smith and Margo Seltzer.
File Layout and File System Performance (Postscript) [Overview],
Harvard Computer Science Technical Report TR-35-94.
Theses
- Lex Stein.
Adaptive Parallel Computation for Heterogeneous Processors (Abstract),
A dissertation presented by Lex Stein to the School of Engineering and Applied Sciences in partial fulfillment of the requirements for the Doctor of Philosophy in the subject of Computer Science, Harvard University, Cambridge, MA, May, 2007 - Daniel J. Ellard.
Trace-Based Analyses and Optimizations for Network Storage Servers (Abstract),
A dissertation presented by Daniel J. Ellard to The School of Engineering and Applied Sciences in partial fulfillment of the requirements for the Doctor of Philosophy in the subject of Computer Science, Harvard University, Cambridge, MA, May, 2004 - Kostas Magoutis.
Exploiting Direct-Access Networking in Network Attached Storage Systems (Abstract),
A dissertation presented by Kostas Magoutis to The School of Engineering and Applied Sciences in partial fulfillment of the requirements for the Doctor of Philosophy in the subject of Computer Science, Harvard University, Cambridge, MA, May, 2003 - Keith Arnold Smith.
Workload-Specific File System Benchmarks (Abstract),
A dissertation presented by Keith Arnold Smith to The School of Engineering and Applied Sciences in partial fulfillment of the requirements for the Doctor of Philosophy in the subject of Computer Science, Harvard University, Cambridge, MA, January, 2001 - Salimah Addetia.
CacheDAFS: User Level Client-side Caching for the Direct Access File System (DAFS) (Abstract),
A thesis presented by Salimah Addetia to The School of Engineering and Applied Sciences in partial fulfillment of the requirements for the Master of Engineering in the subject of Computer Science, Harvard University, Cambridge, MA, 2001