Kimberly Keeton, David A. Patterson, Yong Qiang He, Roger C. Raphael and Walter E. Baker

EECS Department
University of California, Berkeley
Technical Report No. UCB/CSD-98-1001
April 1998

http://www2.eecs.berkeley.edu/Pubs/TechRpts/1998/CSD-98-1001.pdf

Commercial applications are an important, yet often overlooked, workload with significantly different characteristics from technical workloads. The potential impact of these differences is that computers optimized for technical workloads may not provide good perfomance for commercial applications, and these applications may not fully exploit advances in processor design. To evaluate these issues, we use hardware counters to measure architectural features of a four-processor Pentium Pro-based server running a TPC-C-like workload on an Informix database. We examine the effectiveness of out-of-order execution, branch prediction, speculative execution, superscalar issue and retire, caching and multiprocessor scaling. We find that out-of-order execution, superscalar issue and retire, and branch prediction are not as effective for database workloads as they are for technical workloads, such as SPEC. We find that caches are effective at reducing processor traffic to memory; even larger caches would be helpful to satisfy more data requests. Multiprocessor scaling of this workload is good, but even modest bus utilization degrades application memory latency, limiting database throughput.

BibTeX citation:

@techreport{Keeton:CSD-98-1001,
    Author = {Keeton, Kimberly and Patterson, David A. and He, Yong Qiang and Raphael, Roger C. and Baker, Walter E.},
    Title = {Performance Characterization of the Quad Pentium Pro SMP Using OLTP Workloads},
    Institution = {EECS Department, University of California, Berkeley},
    Year = {1998},
    Month = {Apr},
    URL = {http://www2.eecs.berkeley.edu/Pubs/TechRpts/1998/6414.html},
    Number = {UCB/CSD-98-1001},
    Abstract = {Commercial applications are an important, yet often overlooked, workload with significantly different characteristics from technical workloads. The potential impact of these differences is that computers optimized for technical workloads may not provide good perfomance for commercial applications, and these applications may not fully exploit advances in processor design. To evaluate these issues, we use hardware counters to measure architectural features of a four-processor Pentium Pro-based server running a TPC-C-like workload on an Informix database. We examine the effectiveness of out-of-order execution, branch prediction, speculative execution, superscalar issue and retire, caching and multiprocessor scaling. We find that out-of-order execution, superscalar issue and retire, and branch prediction are not as effective for database workloads as they are for technical workloads, such as SPEC. We find that caches are effective at reducing processor traffic to memory; even larger caches would be helpful to satisfy more data requests. Multiprocessor scaling of this workload is good, but even modest bus utilization degrades application memory latency, limiting database throughput.}
}

EndNote citation:

%0 Report
%A Keeton, Kimberly
%A Patterson, David A.
%A He, Yong Qiang
%A Raphael, Roger C.
%A Baker, Walter E.
%T Performance Characterization of the Quad Pentium Pro SMP Using OLTP Workloads
%I EECS Department, University of California, Berkeley
%D 1998
%@ UCB/CSD-98-1001
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/1998/6414.html
%F Keeton:CSD-98-1001