CS 7934 — Computer Systems Seminar, Spring 2010

Fridays, 2:00–3:00 PM, 3485 MEB

Organizer: Eric Eide

Overview

The spring 2010 offering of CS 7934 will cover a variety of systems topics, but with an eye toward three goals.

The first is to increase participants' understanding of testbed-like infrastructures for cloud computing and scientific computing. We will study publications at the intersections of these fields, toward the goal of better understanding how these topics are connected and relate to one another. How are testbeds like Emulab, PlanetLab, and GENI related to the emerging notions of cloud computing? How should testbeds and clouds be designed so that they support experiments not only in computer science, but also in computational science?

The second is to be a venue for student presentations. Every student participating in the seminar will be required to give at least one “formal” research presentation during the semester. Ideally these will be presentations of students' current work, but other topics are also possible.

The third is to stay abreast of papers from recent or imminent top-tier systems conferences: e.g., SOSP, OSDI, NSDI, SIGCOMM, and so on. Papers will be selected for their relevance to participants' research or upcoming Utah visitors.

CS 7934 is often called “the CSL seminar.” The name CSL is historic.

Mailing list

To get on the class mailing list, use Mailman to subscribe to csl-sem.

Credit

Although the course is listed as “variable credit,” the course is only available for one (1) credit in most circumstances. If you want to take the course for more than one credit, you will need to get approval from the instructor.

Those taking the course for credit must read all of the papers, submit a short summary of each paper prior to class (PDF, Postscript, LaTeX), participate in each discussion, and make at least one research presentation. We urge students to sign up for one credit if you're going to be attending anyway.

Schedule

(You can check out what we did last semester here.)

Week Date Topic(s) Facilitator Paper(s)
1 Eide no meeting — organizational email
2 1/22 cloud computing background Eide Above the Clouds: A Berkeley View of Cloud Computing. Armbrust et al. Technical Report UCB/EECS-2009-28.
3 1/29 Nimbus Eide Sky Computing. Katarzyna Keahey et al. IEEE Internet Computing, 13(5):43–51, September/October 2009.

Science Clouds: Early Experiences in Cloud Computing for Scientific Applications. Katarzyna Keahey et al. In CCA '08, October 2008.
4 2/5 EC2 et al. Ricci An Evaluation of Amazon's Grid Computing Services: EC2, S3 and SQS. Simson L. Garfinkel. Harvard School for Engineering and Applied Sciences Technical Report TR–08–07. July 2007.

Amazon S3 for Science Grids: A Viable Solution? Mayur R. Palankar et al. In DADC '08, June 2008.
5 2/12 OpenNebula Hibler Virtual Infrastructure Management in Private and Hybrid Clouds. Borja Sotomayor et al. IEEE Internet Computing, 13(5):14–22, September/October 2009.

Elastic Management of Cluster-Based Services in the Cloud. Rafael Moreno-Vozmediano et al. In ACDC '09, June 2009.
6 2/19 VM I/O Burtsev Achieving 10Gb/s using Safe and Transparent Network Interface Virtualization. Ram et al. In VEE 2009, March 2009.

IBMon: Monitoring VMM-Bypass Capable InfiniBand Devices using Memory Introspection. Ranadive et al. In HPCVirt 2009, March 2009.
7 2/26 Adaptive Computing

and

components for data storage
Eide In 3147 MEB, 11:50 AM–12:40 PM:
Industry Forum presentation by Brady Kimball, Director of Graphical Interfaces, Adaptive Computing

At the usual time and place:
Generalized File System Dependencies. Frost et al. In SOSP '07, October 2007.
Modular Data Storage with Anvil. Mammarella et al. In SOSP '09, October 2009.
8 3/5 no meeting — student research posters
9 3/12 context-based filesystems Pullakandam quFiles: The Right File at the Right Time. Veeraraghavan et al. In FAST '10, February 2010.
10 3/19 MapReduce Wong Improving MapReduce Performance in Heterogeneous Environments. Zaharia et al. In OSDI '08, December 2008.

MapReduce and Parallel DBMSs: Friends or Foes? Stonebraker et al. CACM, 53(1):64–71, January 2010.
11 3/26 no meeting — University spring break
12 4/2 OS virtualization Lin Container-Based Operating System Virtualization: A Scalable, High-Performance Alternative to Hypervisors. Soltesz et al. In EuroSys '07, March 2007.
13 4/9 fault-tolerant filesystems Sun Membrane: Operating System Support for Restartable File Systems. Sundararaman et al. In FAST '10, February 2010.
14 4/16 request partitioning Duerig Centrifuge: Integrated Lease Management and Partitioning for Cloud Services. Adya et al. In NSDI '10, April 2010. To appear.
15 4/23 autonomic computing

and

scientific applications
Ricci In 3760 WEB, 2:00 PM–3:00 PM:
Distinguished SCI Seminar: “Addressing Complexity in Emerging Cyber-Ecosystems — Exploring the Role of Autonomics in Emerging Computational Applications” by Manish Parashar, NSF and Rutgers University

Also read:
High-Performance Cloud Computing: A View of Scientific Applications. Vecchiola et al. In ISPAN 2009, December 2009.

Scientific Computing in the Cloud. Rehr et al. Computing in Science and Engineering, 2010. To appear.
16 4/30 wrap-up Eide At 11:30 AM:
Cloud Computing and Grid Computing 360-Degree Compared. Foster et al. In Grid Computing Environments Workshop (GCE) '08, November 2008.

Toward a Cloud Computing Research Agenda. Birman et al. SIGACT News, 40(2):68–80, June 2009.

Science-Cloud-Related Papers, Whitepapers, Etc.

Science-Cloud-Related Events

Science-Cloud-Related Projects and Software

Science-Cloud-Related Testbeds

Other Potential Papers

For general systems topics, papers from upcoming/recent OSDI, SIGCOMM, SOSP, NSDI, SIGMETRICS, SenSys, and similar conference proceedings are good sources of papers for discussion.