IDC Reports Latest Supercomputer Rankings Based On the IDC Balanced Rating Test.Business Editors FRAMINGHAM Framingham (frā`mĭnghăm'), town (1990 pop. 64,994), Middlesex co., E Mass., on the Sudbury River between Worcester and Boston; settled 1650, inc. 1700. , Mass.--(BUSINESS WIRE)--May 16, 2002 Department of Defense Plans To Fund Applications-based Performance Modeling Tool IDC today reported the newest rankings of supercomputer supercomputer, a state-of-the-art, extremely powerful computer capable of manipulating massive amounts of data in a relatively short time. Supercomputers are very expensive and are employed for specialized scientific and engineering applications that must handle very performance based on its comprehensive standard test, called the IDC Balanced Rating. The test, developed with extensive input from the high-performance computing High-speed computing, which typically refers to supercomputers used in scientific research. (HPC (Handheld PC) A palmtop computer that weighs less than one pound and runs specialized versions of popular applications. Microsoft coined the term for its Windows CE operating system, which is an abbreviated version of Windows. See Pocket PC. ) community, is important because industrial competitiveness and scientific progress depend on this most-powerful class of computers. "Despite the strategic and economic importance of supercomputers and HPC systems, there have been no generally available databases or tools for analyzing and comparing these complex technologies along the multiple dimensions which highlight overall system performance," said Debra Goldfarb Goldfarb is German for golden color and is the surname of:
adj. 1. Of or relating to people of the highest office or rank. 2. Of or relating to the highest office or rank: a top-level job. tools for sorting through various aspects of the technologies to better understand and explain how different computer architectures match their requirements. The IDC Balanced Rating provides one such top-level analysis and ranking." The latest rankings list 893 currently installed HPC systems in four market segments. "Our goal is to document as many of the world's installed HPC systems as possible, and make the data available for free to the entire HPC community," said Earl Joseph II, IDC Research Director, Worldwide Systems and Servers. The four market segments are: 1. Technical Capability Computers: Systems purchased to solve the most demanding problems. 2. Technical Enterprise Computers: Capacity systems that sell for $1 million or more. 3. Technical Divisional Computers: Capacity systems sold from $250,000 to $999,000. 4. Technical Departmental Computers: Capacity systems sold for less than $250,000. The IDC Balanced Rating focuses on three broad performance areas: -- Processor performance. SPECfp_rate_base2000, SPECint_rate_base2000 and Linpack Rmax results are used. In the future, additional processor metrics will be incorporated. -- Memory system capability. Actual performance is dependent on how much data can be moved into and out of the processors in a given time period. For the IDC Balanced Rating, memory bandwidth and the system level STREAM TRIAD benchmarks are used. -- Scaling capability. Total processor count and total system memory and interconnect bandwidth are used. In the future, latency and other metrics will be integrated into the benchmark based on the availability of consistent data on HPC computers. The attached partial tables show the top 50 HPC systems in each market segment, based on data submitted to date by HPC vendors and users. A more detailed report and the full list of 893 computers are available without charge at: www.idc.com/hpc. The same website includes a new tool that allows site visitors to custom-rank the 893 listed HPC systems according to according to prep. 1. As stated or indicated by; on the authority of: according to historians. 2. In keeping with: according to instructions. 3. their own computing computing - computer priorities. "Predicting the performance characteristics of a computer for your specific application codes is a very complex process and buyers often narrow the field of choices by first deciding on the basic computer architecture type, e.g., a single computer vs. a cluster cluster, in astronomy: see star cluster; galaxy. (1) Two or more systems working together. See clustering. (2) Also called an "allocation unit" or "file allocation unit," it is some number of disk sectors that are treated as a unit. , large nodes vs. small nodes, RISC RISC in full Reduced Instruction Set Computing Computer architecture that uses a limited number of instructions. RISC became popular in microprocessors in the 1980s. vs. vectors Vectors Something used to transport genetic information to a cell. Mentioned in: Gene Therapy , etc. Any broad ranking list like this one is often more useful in the acquisition of lower priced computers since you can't afford to spent a million dollars to evaluate buying a $200,000 system," said Joseph. "Data for some tests within the IDC Balanced Rating are readily available, such as the Linpack A package of FORTRAN programs for numerical linear algebra that is commonly used to create benchmark programs for testing a computer's floating point performance. See benchmark. 1. LINPACK - A package of linear algebra routines. 2. test that measures aspects of processor performance," said Joseph. "Some vendors have not yet run all the other tests that are important for assessing a system's overall capabilities, or don't don't 1. Contraction of do not. 2. Nonstandard Contraction of does not. n. A statement of what should not be done: a list of the dos and don'ts. have results yet for all of their listed systems. The IDC Balanced Rating is a major step forward today and will become increasingly valuable over time, as the data picture is expanded and as the approach evolves." Joseph said that draft results of the latest IDC Balanced Rating were circulated several times within the HPC User Forum, a group of leading U.S. high-performance computing users from government, industry and academia, and to others within the global HPC community. "Any new measurement scheme like this needs time to mature. For this second edition of the IDC Balanced Rating, as with the first edition, we made a concerted effort to involve a large number of reviewers and incorporate their constructive (mathematics) constructive - A proof that something exists is "constructive" if it provides a method for actually constructing it. Cantor's proof that the real numbers are uncountable can be thought of as a *non-constructive* proof that irrational numbers exist. feedback. We invite suggestions at all times via email to hpc@idc.com. To quickly provide the community with updates to the list, based on the addition of new computers, adding missing data items, or the correction CORRECTION,punishment. Chastisement by one having authority of a person who has committed some offence, for the purpose of bringing him to legal subjection. 2. It is chiefly exercised in a parental manner, by parents, or those who are placed in loco parentis. of errors, we will continually con·tin·u·al adj. 1. Recurring regularly or frequently: the continual need to pay the mortgage. 2. update the full database at the web site." US Department of Defense and HPC User Forum Members Plan To Fund Applications-based Performance Modeling Tools Goldfarb also announced that the DoD High Performance Computing Modernization modernization Transformation of a society from a rural and agrarian condition to a secular, urban, and industrial one. It is closely linked with industrialization. As societies modernize, the individual becomes increasingly important, gradually replacing the family, Program (HPCMP HPCMP High Performance Computing Modernization Program (DoD) HPCMP High-Performance Computer Monitoring Project ), along with the HPC User Forum, plans to fund a 12-month effort at approximately ap·prox·i·mate adj. 1. Almost exact or correct: the approximate time of the accident. 2. $250,000 to develop an integrated set of tools and methodologies for performance modeling and performance prediction In computer science, performance prediction means to estimate the execution time or other performance factors (such as cache misses) of a program on a given computer. It is being widely used for computer architects to evaluate new computer designs, for compiler writers to explore of large applications on HPC systems. "We are very pleased that the HPC User Forum has established an environment that facilitates the creation of joint initiatives to investigate these important performance issues and provides a forum for the discussion of these issues," said Larry Davis Larry Davis can refer to several different people:
Davis said that Allan Allan can refer to:
The San Diego Supercomputer Center (SDSC) is an organized research unit of the University of California, San Diego (UCSD). will lead the development effort. Data from this effort will be used to further strengthen and support the scientific underpinnings of the IDC Balanced Ratings, as well as the HPC User Forum-sponsored user benchmark A performance test of hardware and/or software. There are various programs that very accurately test the raw power of a single machine, the interaction in a single client/server system (one server/multiple clients) and the transactions per second in a transaction processing system. tool that will allow each user to set his or her own weightings for various system attributes that can then be used to compared potential HPC systems for a given workload The term workload can refer to a number of different yet related entities. An amount of labor While a precise definition of a workload is elusive, a commonly accepted definition is the hypothetical relationship between a group or individual human operator and task demands. . About IDC IDC is the world's leading provider of information technology (IT) industry analysis, market data and insight, and strategic and tactical guidance to builders, providers and users of information technology. IDC provides global research with local content through more than 600 analysts in 43 countries worldwide. IDC is a division of IDG IDG International Data Group IDG Integrated Drive Generator IDG Installation Design Guide IDG Internet Discussion Group IDG Inset Dielectric Guide IDG International Dangerous Goods (mail, shipping) , the world's leading IT media, research, and exposition exposition or exhibition, term frequently applied to an organized public fair or display of industrial and artistic productions, designed usually to promote trade and to reflect cultural progress. company.
----------------------------------------------------------------------
IDC Balanced Rating: All Capability Computers
----------------------------------------------------------------------
Source: IDC, 2002
----------------------------------------------------------------------
Number Number 1 Computer
of of or IDC
Rank Vendor Model CPUs Nodes Cluster Rating
----------------------------------------------------------------------
1 NEC Earth Simulator 5120 640 C 40478.22
----------------------------------------------------------------------
2 IBM ASCI White,SP Power3 8192 512 C 4867.41
----------------------------------------------------------------------
3 HP LANL AlphaServerSC45 P-2 4096 1,024 C 3920.67
----------------------------------------------------------------------
4 HP PSC AlphaServer SC45 3016 754 C 2914.15
----------------------------------------------------------------------
5 HP CEA DAM AlphaSvr SC45 2560 640 C 2485.04
----------------------------------------------------------------------
6 SGI ASCI Blue Mountain 6144 48 C 2369.20
----------------------------------------------------------------------
7 IBM SP Power3 375 MHz 16 way 3328 208 C 1946.19
----------------------------------------------------------------------
8 IBM eServer pSeries p690 Turbo 928 29 C 1683.80
----------------------------------------------------------------------
9 NEC SX-6/192M24 192 24 C 1527.20
----------------------------------------------------------------------
10 HP LANL AlphaServerSC45 P-1 1536 384 C 1515.87
----------------------------------------------------------------------
11 IBM NERSC SP Power3 375 NH2 2528 158 C 1504.69
----------------------------------------------------------------------
12 IBM eServer pSeries p690 Turbo 800 25 C 1451.55
----------------------------------------------------------------------
13 IBM eServer pSeries p690 Turbo 768 24 C 1393.49
----------------------------------------------------------------------
14 IBM eServer pSeries p690 Turbo 768 24 C 1393.49
----------------------------------------------------------------------
15 Fujitsu VPP5000/100 100 100 C 1377.96
----------------------------------------------------------------------
16 NEC SX-5/128M8 128 8 C 1344.98
----------------------------------------------------------------------
17 IBM eServer pSeries p690 Turbo 704 22 C 1277.36
----------------------------------------------------------------------
18 Intel ASCI Red 9632 9,632 C 1248.72
----------------------------------------------------------------------
19 IBM eServer pSeries p690 512 32 C 1042.18
----------------------------------------------------------------------
20 NEC SX-6/128M16 128 16 C 1018.14
----------------------------------------------------------------------
21 IBM Power3 H 375 MHz 1920 120 C 961.60
----------------------------------------------------------------------
22 Cray T3E-600 1904 1 1 845.57
----------------------------------------------------------------------
23 SGI Origin 3000 400 MHz 1408 12 C 821.29
----------------------------------------------------------------------
24 IBM ASCI Blue-Pacific SST,604e 5808 1,464 C 816.58
----------------------------------------------------------------------
25 IBM eServer pSeries p690 Turbo 448 14 C 812.87
----------------------------------------------------------------------
26 IBM SP Power3 375 MHz 8 way 1152 144 C 774.80
----------------------------------------------------------------------
27 SGI ORIGIN 2000 250 2048 16 C 773.34
----------------------------------------------------------------------
28 Fujitsu VPP5000/56 56 56 C 771.66
----------------------------------------------------------------------
29 SGI Origin 3000 600 MHz 1024 1 1 771.63
----------------------------------------------------------------------
30 NEC SX-6/96M12 96 12 C 763.60
----------------------------------------------------------------------
31 IBM SP Power3 375 MHz 16 way 1280 80 C 748.53
----------------------------------------------------------------------
32 IBM SP Power3 375 MHz 16 way 1260 79 C 736.84
----------------------------------------------------------------------
33 HP Sandia 1024 ASDS10LMyrinet 1024 1,024 C 734.01
----------------------------------------------------------------------
34 IBM eServer pSeries p690 Turbo 400 13 C 725.78
----------------------------------------------------------------------
35 HP JAERI AlphaServer SC40 908 227 C 708.35
----------------------------------------------------------------------
36 IBM eServer pSeries p690 Turbo 384 12 C 696.74
----------------------------------------------------------------------
37 IBM eServer pSeries p690 Turbo 384 12 C 696.74
----------------------------------------------------------------------
38 IBM eServer pSeries p690 Turbo 384 12 C 696.74
----------------------------------------------------------------------
39 IBM eServer pSeries p690 Turbo 384 12 C 696.74
----------------------------------------------------------------------
40 IBM eServer pSeries p690 Turbo 384 12 C 696.74
----------------------------------------------------------------------
41 IBM SP Power3 375 MHz 1336 334 C 694.16
----------------------------------------------------------------------
42 SGI Origin 3000 500 MHz 1024 2 C 692.87
----------------------------------------------------------------------
43 NEC SX-6/80M10 80 10 C 636.33
----------------------------------------------------------------------
44 IBM SP Power3 375 MHz 16 way 1088 68 C 636.25
----------------------------------------------------------------------
45 SGI Origin 3800-400 1024 1 1 618.00
----------------------------------------------------------------------
46 IBM eServer pSeries p690 Turbo 320 10 C 580.62
----------------------------------------------------------------------
47 IBM SP Power3 375 MHz 1104 276 C 573.62
----------------------------------------------------------------------
48 IBM SP Power3 375 MHz 1104 276 C 573.62
----------------------------------------------------------------------
49 Cray T3E-900 1324 1 1 565.54
----------------------------------------------------------------------
50 SGI Origin 3000 500 MHz 768 2 C 525.72
----------------------------------------------------------------------
----------------------------------------------------------------------
IDC Balanced Rating: All Enterprise Computers
----------------------------------------------------------------------
Source: IDC, 2002
----------------------------------------------------------------------
Number Number 1 Computer
of of or IDC
Rank Vendor Model CPUs Nodes Cluster Rating
----------------------------------------------------------------------
1 IBM eServer pSeries p690 256 16 C 521.09
----------------------------------------------------------------------
2 SGI Origin 2000 300 800 25 C 344.47
----------------------------------------------------------------------
3 IBM eServer pSeries p690 160 10 C 325.68
----------------------------------------------------------------------
4 IBM eServer pSeries p690 122 8 C 248.33
----------------------------------------------------------------------
5 HP SuperDome (PA-8700)@750 256 4 C 218.87
----------------------------------------------------------------------
6 SGI Origin 3800-500 320 1 1 197.03
----------------------------------------------------------------------
7 HP SuperDome (PA-8600) 256 4 C 176.21
----------------------------------------------------------------------
8 SGI Origin 2000 300 384 6 C 165.47
----------------------------------------------------------------------
9 SGI Origin 2000 300 384 6 C 165.47
----------------------------------------------------------------------
10 Sun HPC 10000 400 MHz Cluster 192 96 C 163.53
----------------------------------------------------------------------
11 IBM p690-1.1GHz 96 3 C 151.24
----------------------------------------------------------------------
12 Sun 8 x SF6800, 900 Mhz 192 8 C 146.00
----------------------------------------------------------------------
13 SGI ORIGIN 2000 250 320 10 C 122.22
----------------------------------------------------------------------
14 Sun 64 x SB1000, 750 Mhz 128 64 C 112.88
----------------------------------------------------------------------
15 HP SuperDome (PA-8700)@750 128 2 C 112.27
----------------------------------------------------------------------
16 Sun HPC 10000 400 MHz Cluster 128 64 C 109.17
----------------------------------------------------------------------
17 Sun HPC 10000 333 MHz Cluster 128 64 C 107.70
----------------------------------------------------------------------
18 IBM p690-1.1GHz 64 2 C 100.83
----------------------------------------------------------------------
19 Sun HPC 10000 400 MHz Cluster 120 60 C 97.02
----------------------------------------------------------------------
20 Sun HPC 10000 400 MHz Cluster 120 60 C 96.92
----------------------------------------------------------------------
21 HP SuperDome (PA-8600) 128 2 C 91.42
----------------------------------------------------------------------
22 Cray SV1ex-4 128 4 C 88.66
----------------------------------------------------------------------
23 Cray SV1ex-4 128 4 C 88.66
----------------------------------------------------------------------
24 Sun HPC 10000 400 MHz Cluster 112 56 C 85.49
----------------------------------------------------------------------
25 Sun HPC 10000 400 MHz Cluster 144 72 C 73.57
----------------------------------------------------------------------
26 Sun 4 x SF6800, 900 Mhz 96 4 C 73.00
----------------------------------------------------------------------
27 NEC NEC_SX_5-4C 4 1 1 70.48
----------------------------------------------------------------------
28 Sun 13X Sun Fire V880, 750 Mhz 104 13 C 65.43
----------------------------------------------------------------------
29 NEC SX-6/8A 8 1 1 63.63
----------------------------------------------------------------------
30 Cray SV1-4 128 4 C 61.15
----------------------------------------------------------------------
31 IBM p690-1.3GHz 32 1 1 59.18
----------------------------------------------------------------------
32 HP SuperDome (PA-8700)@750 64 1 1 56.84
----------------------------------------------------------------------
33 NEC SX-5/4B 4 1 1 56.51
----------------------------------------------------------------------
34 NEC SX-5/4C 4 1 1 56.51
----------------------------------------------------------------------
35 IBM p690-1.1GHz 32 1 1 50.41
----------------------------------------------------------------------
36 SGI Origin 3800-600 64 1 1 49.95
----------------------------------------------------------------------
37 Cray SV1-4/64-128 64 4 C 47.11
----------------------------------------------------------------------
38 Cray SV1-4/64-128 64 4 C 47.11
----------------------------------------------------------------------
39 Sun HPC 10000 400 MHz Cluster 80 40 C 47.09
----------------------------------------------------------------------
40 Sun HPC 10000 400 MHz Cluster 110 55 C 46.74
----------------------------------------------------------------------
41 HP SuperDome (PA-8600) 64 1 1 46.74
----------------------------------------------------------------------
42 SGI Origin 3800-500 64 1 1 44.92
----------------------------------------------------------------------
43 Cray SV1ex-2 64 2 C 44.37
----------------------------------------------------------------------
44 HP V2600/HyperPlex 256 8 C 44.03
----------------------------------------------------------------------
45 Sun Sun Fire 15K 72 1 1 42.98
----------------------------------------------------------------------
46 HP SuperDome (PA-8700)@750 48 1 1 42.69
----------------------------------------------------------------------
47 Sun HPC 10000 400 MHz Cluster 104 52 C 42.47
----------------------------------------------------------------------
48 SGI Origin 3800-400 64 1 1 40.04
----------------------------------------------------------------------
49 IBM p690-1.1GHz 24 1 1 38.16
----------------------------------------------------------------------
50 SGI Origin 2000 500 64 1 1 37.39
----------------------------------------------------------------------
----------------------------------------------------------------------
IDC Balanced Rating: All Divisional Computers
----------------------------------------------------------------------
Source: IDC, 2002
----------------------------------------------------------------------
Number Number 1 Computer
of of or IDC
Rank Vendor Model CPUs Nodes Cluster Rating
----------------------------------------------------------------------
1 HP SuperDome (PA-8700)@750 32 1 1 28.89
----------------------------------------------------------------------
2 IBM p670 16 1 1 25.94
----------------------------------------------------------------------
3 IBM p690-1.1GHz 16 1 1 25.94
----------------------------------------------------------------------
4 HP AlphaServer HPC320 6/667 32 8 C 24.28
----------------------------------------------------------------------
5 HP SuperDome (PA-8600) 32 1 1 23.88
----------------------------------------------------------------------
6 IBM p690-HPC-1.3GHz 8 1 1 15.54
----------------------------------------------------------------------
7 IBM p690-1.3GHz 8 1 1 15.20
----------------------------------------------------------------------
8 NEC Express5800/1160Xa 16 1 1 15.09
----------------------------------------------------------------------
9 HP SuperDome (PA-8700)@750 16 1 1 14.78
----------------------------------------------------------------------
10 HP rp8400 (PA-8700)@750MHz 16 1 1 14.70
----------------------------------------------------------------------
11 HP AlphaServer HPC160 6/833 16 4 C 14.48
----------------------------------------------------------------------
12 HP AlphaServer HPC320 6/833 16 4 C 14.48
----------------------------------------------------------------------
13 HP rp8400 (PA-8700)@650MHz 16 1 1 13.95
----------------------------------------------------------------------
14 IBM p670 8 1 1 12.97
----------------------------------------------------------------------
15 IBM p690-1.1GHz 8 1 1 12.95
----------------------------------------------------------------------
16 SGI Origin 3400-600 16 1 1 12.67
----------------------------------------------------------------------
17 SGI Origin 3800-600 16 1 1 12.67
----------------------------------------------------------------------
18 HP AlphaServer HPC160 6/667 16 4 C 12.45
----------------------------------------------------------------------
19 HP AlphaServer HPC320 6/667 16 4 C 12.45
----------------------------------------------------------------------
20 HP SuperDome (PA-8600) 16 1 1 12.22
----------------------------------------------------------------------
21 SGI Origin 3400-500 16 1 1 11.39
----------------------------------------------------------------------
22 SGI Origin 3800-500 16 1 1 11.39
----------------------------------------------------------------------
23 HP rp8400 (PA-8700)@750MHz 12 1 1 10.99
----------------------------------------------------------------------
24 SGI Origin 3400-400 16 1 1 10.11
----------------------------------------------------------------------
25 SGI Origin 3800-400 16 1 1 10.11
----------------------------------------------------------------------
26 Sun Sun Fire 4800 12 1 1 9.74
----------------------------------------------------------------------
27 SGI Origin 300 500 16 1 1 9.66
----------------------------------------------------------------------
28 SGI Origin 2000 500 16 1 1 9.57
----------------------------------------------------------------------
29 IBM POWER3 H- 375MHz 16 1 1 9.51
----------------------------------------------------------------------
30 HP AlphaServer GS320 6/1000 8 1 1 8.63
----------------------------------------------------------------------
31 HP AlphaServer GS160 6/1000 8 1 1 8.45
----------------------------------------------------------------------
32 IBM POWER3 T 375 MHz 16 4 C 8.31
----------------------------------------------------------------------
33 SGI Origin 2000 400 16 1 1 8.29
----------------------------------------------------------------------
34 HP AlphaServer GS80/1000 8 1 1 8.29
----------------------------------------------------------------------
35 IBM POWER3 H- 375MHz 12 1 1 7.80
----------------------------------------------------------------------
36 Sun Sun Fire 6800 8 1 1 7.45
----------------------------------------------------------------------
37 HP AlphaServer HPC160 6/833 8 2 C 7.44
----------------------------------------------------------------------
38 HP AlphaServer HPC320 6/833 8 2 C 7.44
----------------------------------------------------------------------
39 HP rp8400 (PA-8700)@750MHz 8 1 1 7.41
----------------------------------------------------------------------
40 HP AlphaServer GS160 6/731 8 1 1 6.64
----------------------------------------------------------------------
41 HP AlphaServer GS80/731 8 1 1 6.51
----------------------------------------------------------------------
42 Sun Sun Fire 3800 8 1 1 6.50
----------------------------------------------------------------------
43 Sun Sun Fire 4800 8 1 1 6.50
----------------------------------------------------------------------
44 HP AlphaServer GS320 6/731 8 1 1 6.48
----------------------------------------------------------------------
45 HP AlphaServer HPC160 6/667 8 2 C 6.38
----------------------------------------------------------------------
46 HP AlphaServer HPC320 6/667 8 2 C 6.38
----------------------------------------------------------------------
47 SGI Origin 3400-600 8 1 1 6.38
----------------------------------------------------------------------
48 SGI Origin 3800-600 8 1 1 6.38
----------------------------------------------------------------------
49 IBM POWER3 H- 375MHz 8 1 1 6.03
----------------------------------------------------------------------
50 Sun Sun Fire V880 8 1 1 5.99
----------------------------------------------------------------------
----------------------------------------------------------------------
IDC Balanced Rating: All Departmental Computers
----------------------------------------------------------------------
Source: IDC, 2002
----------------------------------------------------------------------
Number Number 1 Computer
of of or IDC
Rank Vendor Model CPUs Nodes Cluster Rating
----------------------------------------------------------------------
1 HP rp8400 (PA-8700)@650MHz 12 1 1 10.40
----------------------------------------------------------------------
2 HP rp7410(PA-8700)@750MHz 8 1 1 7.64
----------------------------------------------------------------------
3 NEC Express5800/1160Xa 8 1 1 7.54
----------------------------------------------------------------------
4 HP rp8400 (PA-8700)@650MHz 8 1 1 7.09
----------------------------------------------------------------------
5 HP rp7410(PA-8700)@650MHz 8 1 1 6.88
----------------------------------------------------------------------
6 HP rp7400 (PA-8700)@750 8 1 1 6.70
----------------------------------------------------------------------
7 IBM p670 4 1 1 6.47
----------------------------------------------------------------------
8 SGI Origin 3200-600 8 1 1 6.38
----------------------------------------------------------------------
9 HP rp7400 (PA-8700)@650 8 1 1 6.22
----------------------------------------------------------------------
10 SGI Origin 3200-500 8 1 1 5.72
----------------------------------------------------------------------
11 SGI Origin 3800-500 8 1 1 5.72
----------------------------------------------------------------------
12 HP N4000 (PA-8600) 8 1 1 5.38
----------------------------------------------------------------------
13 Sun Sun Enterprise 4500 14 1 1 5.25
----------------------------------------------------------------------
14 SGI Origin 3200-400 8 1 1 5.09
----------------------------------------------------------------------
15 SGI Origin 3400-400 8 1 1 5.09
--------------------------------------------------------------------
16 SGI Origin 3800-400 8 1 1 5.09
----------------------------------------------------------------------
17 SGI Origin 300 500 8 1 1 5.07
----------------------------------------------------------------------
18 NEC Express5800/1160Xa 4 1 1 4.95
----------------------------------------------------------------------
19 SGI Origin 3200-600 6 1 1 4.92
----------------------------------------------------------------------
20 HP AlphaServer ES45/1000 4 1 1 4.88
----------------------------------------------------------------------
21 SGI Origin 2000 500 8 1 1 4.79
----------------------------------------------------------------------
22 Sun Sun Enterprise 4000 12 1 1 4.53
----------------------------------------------------------------------
23 SGI Origin 3200-500 6 1 1 4.44
----------------------------------------------------------------------
24 HP rp7410(PA-8700)@750MHz 4 1 1 4.25
----------------------------------------------------------------------
25 IBM POWER3 H- 375MHz 4 1 1 4.23
----------------------------------------------------------------------
26 SGI Origin 2000 400 8 1 1 4.14
----------------------------------------------------------------------
27 HP rp7400 (PA-8700)@750 4 1 1 4.13
----------------------------------------------------------------------
28 IBM p660-6H1-750MHz 6 1 1 3.99
----------------------------------------------------------------------
29 IBM p620-6F1-750MHz 6 1 1 3.96
----------------------------------------------------------------------
30 HP rp7410(PA-8700)@650MHz 4 1 1 3.87
----------------------------------------------------------------------
31 HP rp7400 (PA-8700)@650 4 1 1 3.85
----------------------------------------------------------------------
32 HP AlphaServer HPC160 6/833 4 1 1 3.82
----------------------------------------------------------------------
33 HP AlphaServer HPC320 6/833 4 1 1 3.82
----------------------------------------------------------------------
34 Sun Sun Fire V880 4 1 1 3.80
----------------------------------------------------------------------
35 HP rp5470 (PA-8700)@750MHz 4 1 1 3.79
----------------------------------------------------------------------
36 HP RX4610 (Itanium)@800 4 1 1 3.74
----------------------------------------------------------------------
37 HP rp8400 (PA-8700)@750MHz 4 1 1 3.71
----------------------------------------------------------------------
38 IBM p620-6F1-668MHz 6 1 1 3.60
----------------------------------------------------------------------
39 IBM p660-6H1-668MHz 6 1 1 3.60
----------------------------------------------------------------------
40 HP rp8400 (PA-8700)@650MHz 4 1 1 3.54
----------------------------------------------------------------------
41 HP rp5470 (PA-8700)@650MHz 4 1 1 3.51
----------------------------------------------------------------------
42 HP RX4610 (Itanium)@733 4 1 1 3.49
----------------------------------------------------------------------
43 IBM p660-6M1 4 1 1 3.43
----------------------------------------------------------------------
44 HP N4000 (PA-8600) 4 1 1 3.38
----------------------------------------------------------------------
45 IBM POWER3 H- 375MHz 2 1 1 3.31
----------------------------------------------------------------------
46 HP AlphaServer HPC160 6/667 4 1 1 3.27
----------------------------------------------------------------------
47 HP AlphaServer HPC320 6/667 4 1 1 3.27
----------------------------------------------------------------------
48 Sun Sun Fire 3800 4 1 1 3.25
----------------------------------------------------------------------
49 HP AlphaServer GS80/731 4 1 1 3.25
----------------------------------------------------------------------
50 SGI Origin 3200-600 4 1 1 3.16
----------------------------------------------------------------------
|
|
||||||||||||
Printer friendly
Cite/link
Email
Feedback
Reader Opinion