Eighth Workshop on Computer Architecture Evaluation using Commercial Workloads (CAECW-8)


San Francisco, CA, USA

Saturday, February 12, 2005

 

Immediately precedes

the 11th International Symposium on High Performance Computer Architecture (HPCA-11)

Workshop webpage:  http://tesla.hpl.hp.com/caecw05/

 



The function of this workshop is the discussion of work-in-progress that utilizes commercial workloads for the evaluation of computer architectures.  By discussing this ongoing research, the workshop will expose participants to the characteristics of commercial workload behavior, provide an understanding of how commercial workloads exercise computer systems and potentially help establish methodologies for measuring, modeling and analyzing the execution time characteristics of these workloads.

The tentative final program for the workshop is listed below. The program includes refereed paper presentations, invited talks (including a keynote), and plenty of time for audience participation. Hard copies of the proceedings will be provided for registered attendees at the workshop; an electronic version of the full proceedings will also be made available as a pdf file here.

 

 Final Program

8:00 - 8:30

Registration

8:30 - 9:30

Session 1: Keynote

 

Serving Web Searches on the Cheap: Application and Platform Co-adaptation

W.-D. Weber (Google)

9:30 - 10:00

Session 2

 

Clock-Gating Efficiency: Commercial vs. Scientific Applications (Invited talk)

H. Jacobson, P. Bose (IBM TJ Watson Research Center)

 

10:00 - 10:30

Coffee Break

10:3 0 - 12:00

Session 3: Workload Characterization

A Comparison of SPECjAppServer2002 and SPECjAppServer2004

L. Su, K. Chow, K. Shiv and A. Jha (Intel)

 

Performance Workloads Characterization on POWER5 with Simultaneous Multithreading Support

O.F. Herescu, B. Olszewski and H. Hua (IBM)

 

Data Center Workload Monitoring, Analysis, and Emulation (Invited talk) (talk slides)

J. Moore and J. Chase (Duke University) and K. Farkas and P. Ranganathan (HP Labs)

12:00 - 13:00

Lunch

13:00 - 14:30

Session 4: Storage and Network I/O

Analyzing NIC Overheads in Network-Intensive Workloads

N. L. Binkert, L.R. Hsu, A.G. Saidi, R.G. Dreslinski, A.L. Schultz and S.K. Reinhardt (University of Michigan)

 

Storage over IP -- A Workload Analysis with iSCSI

M. Deval, R. Illikkal and D. Newell (Intel)

 

Is Traditional Power Management + Prefetching == DRPM for Server Disks?

V. Natarajan, S. Gurumurthi and A. Sivasubramaniam (Penn State University)

14:30 - 15:00

Coffee Break

15:00 - 17:00

Session 5: Potpourri

Accurate TPC-C Predictions:  How Many Random Samples are Enough? (Invited talk)

M. Adler (Intel)

 

Analyses of Embedded Processors for Streaming Media Applications

A.R. Iranpour and K. Kuchcinski (Lund University)

 

Virtual Timers: Using Hardware Physical Timers for Profiling Kernel Code Paths (talk slides)

D. Xinidis, M.D. Flouris and A. Bilas (ICS, FORTH)

 

Workload-Driven Analysis of File Systems in Shared Multi-Tier Data Centers over Infiniband (talk slides)

K. Vaidyanathan, P. Balaji, H.-W. Jin and D. K. Panda (Ohio State University)

17:00

Participant Feedback

Closing Remarks


Workshop organizers

Past workshops