NESC9717 SCWEB.
SCWEB, Scientific Workstation Evaluation Benchmark
NAME OR DESIGNATION OF PROGRAM, COMPUTER, DESCRIPTION OF PROGRAM OR FUNCTION, METHOD OF SOLUTION, RESTRICTIONS ON THE COMPLEXITY OF THE PROBLEM, TYPICAL RUNNING TIME, UNUSUAL FEATURES OF THE PROGRAM, RELATED AND AUXILIARY PROGRAMS, STATUS, REFERENCES, MACHINE REQUIREMENTS, LANGUAGE, OPERATING SYSTEM UNDER WHICH PROGRAM IS EXECUTED, OTHER PROGRAMMING OR OPERATING INFORMATION OR RESTRICTIONS, NAME AND ESTABLISHMENT OF AUTHORS, MATERIAL, CATEGORIES
[ top ]
[ top ]
To submit a request, click below on the link of the version you wish to order.
Only liaison officers are authorised to submit online requests. Rules for requesters are
available here.
| Program name | Package id | Status | Status date |
|---|---|---|---|
| SCWEB | NESC9717/01 | Tested | 16-JUN-1988 |
Machines used:
| Package ID | Orig. computer | Test computer |
|---|---|---|
| NESC9717/01 | Many Computers | DEC VAX 11/780 |
[ top ]
3. DESCRIPTION OF PROGRAM OR FUNCTION
The SCWEB (SCIENTIFIC WORKSTATION EVALUATION BENCHMARK) software includes 16 programs which are executed in a well-defined scenario to measure the following performance capabilities of a scientific workstation: implementation of FORTRAN77, processor speed, memory management, disk I/O, monitor (or display) output, scheduling of processing (multiprocessing), and scheduling of print tasks (spooling).
The SCWEB (SCIENTIFIC WORKSTATION EVALUATION BENCHMARK) software includes 16 programs which are executed in a well-defined scenario to measure the following performance capabilities of a scientific workstation: implementation of FORTRAN77, processor speed, memory management, disk I/O, monitor (or display) output, scheduling of processing (multiprocessing), and scheduling of print tasks (spooling).
[ top ]
4. METHOD OF SOLUTION
The benchmark programs are: DK1, DK2, and DK3, which do Fourier series fitting based on spline techniques; JC1, which checks the FORTRAN function routines which produce numerical results; JD1 and JD2, which solve dense systems of linear equations in double- and single-precision, respectively; JD3 and JD4, which perform matrix multiplication in single- and double-precision, respectively; RB1, RB2, and RB3, which perform substantial amounts of I/O processing on files other than the input and output files; RR1, which does intense single-precision floating-point multiplication in a tight loop, RR2, which initializes a 512x512 integer matrix in a manner which skips around in the address space rather than initializing each consecutive memory cell in turn; RR3, which writes alternating text buffers to the output file; RR4, which evaluates the timer routines and demonstrates that they conform to the specification; and RR5, which determines whether the workstation is capable of executing a 4-megabyte program.
The benchmark programs are: DK1, DK2, and DK3, which do Fourier series fitting based on spline techniques; JC1, which checks the FORTRAN function routines which produce numerical results; JD1 and JD2, which solve dense systems of linear equations in double- and single-precision, respectively; JD3 and JD4, which perform matrix multiplication in single- and double-precision, respectively; RB1, RB2, and RB3, which perform substantial amounts of I/O processing on files other than the input and output files; RR1, which does intense single-precision floating-point multiplication in a tight loop, RR2, which initializes a 512x512 integer matrix in a manner which skips around in the address space rather than initializing each consecutive memory cell in turn; RR3, which writes alternating text buffers to the output file; RR4, which evaluates the timer routines and demonstrates that they conform to the specification; and RR5, which determines whether the workstation is capable of executing a 4-megabyte program.
[ top ]
[ top ]
NESC9717/01
NEA-DB ran the test cases included in this package on a VAX-11/780 computer. CPU times varied between 0.8 and 130 seconds.[ top ]
[ top ]
[ top ]
NESC9717/01, included references:
- R.C. Raffenetti :Scientific Workstation Comparisons Using a Benchmark
ANL/TM-415 (April 1984)
[ top ]
| Package ID | Computer language |
|---|---|
| NESC9717/01 | FORTRAN-77 + ASSEMBLER |
[ top ]
[ top ]
[ top ]
NESC9717/01
| File name | File description | Records |
|---|---|---|
| NESC9717_01.001 | Information file | 98 |
| NESC9717_01.002 | SCWEB DK1 Source Fortran | 1393 |
| NESC9717_01.003 | SCWEB DK2 Source Fortran | 1390 |
| NESC9717_01.004 | SCWEB DK3 Source Fortran | 1389 |
| NESC9717_01.005 | SCWEB JC1 Source Fortran | 732 |
| NESC9717_01.006 | SCWEB JD1 Source Fortran | 612 |
| NESC9717_01.007 | SCWEB JD2 Source Fortran | 611 |
| NESC9717_01.008 | SCWEB JD3 Source Fortran | 563 |
| NESC9717_01.009 | SCWEB JD4 Source Fortran | 564 |
| NESC9717_01.010 | SCWEB RB1 Source Fortran | 204 |
| NESC9717_01.011 | SCWEB RB2 Source Fortran | 202 |
| NESC9717_01.012 | SCWEB RB3 Source Fortran | 204 |
| NESC9717_01.013 | SCWEB RR1 Source Fortran | 50 |
| NESC9717_01.014 | SCWEB RR1 Source Fortran | 50 |
| NESC9717_01.015 | SCWEB RR2 Source Fortran | 61 |
| NESC9717_01.016 | SCWEB RR3 Source Fortran | 56 |
| NESC9717_01.017 | SCWEB RR4 Source Fortran | 74 |
| NESC9717_01.018 | SCWEB RR5 Source Fortran | 52 |
| NESC9717_01.019 | SCWEB DK1 Printed sample output | 108 |
| NESC9717_01.020 | SCWEB DK2 Printed sample output | 108 |
| NESC9717_01.021 | SCWEB DK3 Printed sample output | 108 |
| NESC9717_01.022 | SCWEB JC1 Printed sample output | 345 |
| NESC9717_01.023 | SCWEB JD1 Printed sample output | 30 |
| NESC9717_01.024 | SCWEB JD2 Printed sample output | 26 |
| NESC9717_01.025 | SCWEB JD3 Printed sample output | 22 |
| NESC9717_01.026 | SCWEB JD4 Printed sample output | 22 |
| NESC9717_01.027 | SCWEB DK1 Input data | 181 |
| NESC9717_01.028 | SCWEB DK2 Input data | 181 |
| NESC9717_01.029 | SCWEB DK3 Input data | 181 |
Keywords: FORTRAN, computers, performance testing.