Windows NT Workstation Performance Evaluation Based on Pro/E 2000i BENCHMARK
A performance evaluation of several computers was necessary, so an evaluation program, or benchmark, was run on each computer to determine maximum possible performance. The program was used to test the Computer Aided Drafting (CAD) ability of each computer by monitoring the speed with which several functions were executed. The main objective of the benchmarking program was to record assembly loading times and image regeneration times and then compile a composite score that could be compared with the same tests on other computers. The three computers that were tested were the Compaq AP550, the SGI 230, and the Hewlett-PackardP750C. The Compaq and SGI computers each had a Pentium III 733mhz processor, while the Hewlett-Packard had a Pentium III 750mhz processor. The size and speed of Random Access Memory (RAM) in each computer varied, as did the type of graphics card. Each computer that was tested was using Windows NT 4.0 and Pro/ENGINEER{trademark} 2000i CAD benchmark software provided by Standard Performance Evaluation Corporation (SPEC). The benchmarking program came with its own assembly, automatically loaded and ran tests on the assembly, then compiled the time each test took to complete. Due to the automation of the tests, any sort of user error affecting test scores was virtually eliminated. After all the tests were completed, scores were then compiled and compared. The Silicon Graphics 230 was by far the overall winner with a composite score of 8.57. The Compaq AP550 was next with a score of 5.19, while the Hewlett-Packard P750C performed dismally, achieving a score of 3.34. Several factors, including motherboard chipset, graphics card, and the size and speed of RAM, were involved in the differing scores of the three machines. Surprisingly the Hewlett-Packard, which had the fastest processor, came back with the lowest score. The above factors most likely contributed to the poor performance of the Hewlett-Packard. Based on the results of the benchmark test, the SGI 230 appears to be the best CAD software solution. The Hewlett-Packard most likely performed poorly due to the fact that it was only running a 100mhz Front Side Bus (FSB), while the SGI machine was running at a 133mhz. The Compaq was using a new type of RAM called RDRAM. While this RAM was at first perceived to be a great performer, various benchmarks, including this one, have found that the computers using RDRAM really only achieve average performance.
- Research Organization:
- Sandia National Lab. (SNL-NM), Albuquerque, NM (United States); Sandia National Lab. (SNL-CA), Livermore, CA (United States)
- Sponsoring Organization:
- US Department of Energy (US)
- DOE Contract Number:
- AC04-94AL85000
- OSTI ID:
- 761843
- Report Number(s):
- SAND2000-1946C; TRN: AH200033%%1
- Resource Relation:
- Conference: Summer Technical Student Internship Program 5th Annual Symposium and Awards Banquet, Albuquerque, NM (US), 08/10/2000; Other Information: PBD: 2 Aug 2000
- Country of Publication:
- United States
- Language:
- English
Similar Records
Experience with mixed MPI/threaded programming models
Experiences with Installing and Benchmarking SCALE 4.0 on Workstations