The Catia viewset used in SPECviewperf 11, could play a role in SPEC’s forthcoming workstation benchmark
The Standard Performance Evaluation Corp (SPEC) is seeking help to develop a new workstation performance benchmark.
The non-profit corporation’s Graphics and Workstation Performance Group (GWPG) is looking for applications, algorithms and workloads that are representative of those used in Computer Aided Engineering (CAE), digital media and entertainment and other areas. The tests should be scalable (presumably so they can show benefits of multi-core processors), preferably solve large problems, support multiple architectures and be freely available to the public.
The non-profit corporation hopes to augment tests from its widely used SPECviewperf benchmark.
Like SPECviewperf, the key benefit of the proposed workstation benchmark is that it won’t require the full application and associated licensing to be installed on the system under test. This means anyone can use it, and because results are publicly available, it’s easy to compare to other workstations without having to test them all yourself.
The downside is that all engineering datasets can behave differently and this one-size fits all approach doesn’t always yield the most useful results. Manufacturers of workstation hardware have also been known to optimize their technology for such benchmarks, which can skew real-world results. This is true for all standard computer benchmarks.
Early participants in the workstation working group include AMD, Dell, Fujitsu, HP, Intel and Nvidia.
