CS594 Advanced Operating Systems (+fault tolerance, distributed systems)

Jim Plank --- Summer, 1998

This class meets on thursdays at 1:45 in the CS reading room.

Background

If you haven't done so, you should read the following two papers, which provide a good introduction to checkpointing systems:

An Overview of Checkpointing in Uniprocessor and Distributed Systems, Focusing on Implementation and Performance, James S. Plank. Technical Report UT-CS-97-372, University of Tennessee, July 1997.
A Survey of Rollback-Recovery Protocols in Message Passing Systems, E. N. Elnozahy, D. B. Johnson and Y. M. Wang. Technical Report CMU-CS-96-181, Department of Computer Science, Carnegie Mellon University, September 1996.
Yes, this is a long one, but it's an excellent survey of the field (and the margins are big).

Papers

June 11, 1998: Remote I/O: Fast Access to Distant Storage, I. Foster and D. Kohr and R. Krishnaiyer and J. Mogill, Proceedings of I/O in Parallel and Distributed Systems, 1997.
The Globus Home Page.
June 16, 1998: WebOS: Operating System Services for Wide Area Applications, Amin Vahdat, Paul Eastham, Chad Yoshokawa, Eshwar Belani, Thomas Anderson, David Culler and Michael Dahlin, Seventh Symposium on High Performance Distributed Computing, July, 1998.
Also available as: Tech Report CSD-97-938, University of California, Berkeley, 1997.
The WebOS Home Page.
June 18, 1998: I will be giving practice talks for my FTCS papers:
Experimental Assessment of Workstation Failures and Their Impact on Checkpointing Systems, James S. Plank and Wael R. Elwasif, 28th International Symposium on Fault-tolerant Computing, June, 1998, pp. 48-57.
Netsolve: An Environment for Deploying Fault-Tolerant Computing, James S. Plank, Henri Casanova, Jack J. Dongarra, Terry Moore, FastAbstracts Session, 28th International Symposium on Fault-tolerant Computing, June, 1998.
The NetSolve Home Page.
July 9, 1998: ``PREACHES -- Portable Recovery and Checkpointing in Heterogeneous Systems'', K-F. Ssu and W. K. Fuchs, 28th International Symposium on Fault-tolerant Computing, June, 1998, pp. 38-47.
The paper is in Rachel's box -- please make yourself a copy.
July 14, 1998: ``The Design and Architecture of the Microsoft Cluster Service - A Practical Approach to High-Availability and Scalability'', W.Vogels, D.Dumitriu, K.Birman, R.Gamache, M.Massa, R.Short, J.Vert, J.Barrera, J.Gray. 28th International Symposium on Fault-tolerant Computing, June, 1998, pp. 422-431.
If you can't get it online, the paper is in Rachel's box -- please make yourself a copy.
July 21, 1998: ``Automated Robustness Testing of Off-The-Shelf Software Components'', N.P. Kropp, P.J. Koopman and D.P.Siewiorek, 28th International Symposium on Fault-tolerant Computing, June, 1998, pp. 230-239.
The paper is in Rachel's box -- please make yourself a copy.
July 28, 1998: `Just-in-time Transparent Resource Management in Distributed Systems,'' Arash Baratloo, Ayal Itzkovitz, Zvi M. Kedem and Yuanyuan Zhao, Technical Report TR1998-762, Department of Computer Science, New York University, March, 1998.
The Milan Home Page
July 30, 1998: ``Active Networking for Storage Exploiting Active Networks for Network-Attached Storage'', D. Nagle, G. Ganger, G. Gibson and H. Zhang, White paper, March 13, 1998.
Home page of the Parallel Data Lab at CMU.
August 6, 1998: Jini white papers. Yes, this is a lot of text (30 some pages), but a lot of it is skippable garbage, so it's not bad.
- ``What is Jini,'' a small summary from Sun. The link above is pdf. If you can't deal with pdf, here is a postscript copy that I'll delete in a week or two.
- ``Jini Architecture Overview,'' Jim Waldo. The link above is pdf. If you can't deal with pdf, here is a postscript copy that I'll delete in a week or two.
- ``Why Jini Now,'' some sales prose from Sun. The link above is pdf. If you can't deal with pdf, here is a postscript copy that I'll delete in a week or two.
The Jini home page.
August 13, 1998: ``Discount Checking: Transparent, Low-Overhead Recovery for General Applications,'' P. M. Chen, D. E. Lowell and G. W. Dunlap. I assume this is a submission to some conference. 1998.