kruger@16bits.dec.com (I've got 50nS memory. What did you say?) (02/27/88)
I'd like to see the dhrystones benchmark, preferably implemented in C. If you have it, please mail it to me. Other popular benchmarks also accepted -- ie the Byte test suite and other such nonsense. Thanks! dov
eugene@pioneer.arpa (Eugene N. Miya) (02/28/88)
If you want benchmarks, use NBSLIB.
nbslib@icst-cmr.arpa
Body of message shoud be like:
send index
will show available directories. Do this to see what you get.
send index from stones
Will show what stones are available.
send dhryc from stones
Gets you the version 1 dhrystones.
In recent days, I've promised to help them update and maintain this
thing. I've "refer'ed" their benchmark bibliography and added our
references (Alan Smith's and mine) the NBS benchmark record
will be updated shortly. The NBS needs our support in this.
You want to add something? Fine.
From the Rock of Ages Home for Retired Hackers:
--eugene miya, NASA Ames Research Center, eugene@ames-aurora.ARPA
"You trust the `reply' command with all those different mailers out there?"
"Send mail, avoid follow-ups. If enough, I'll summarize."
{uunet,hplabs,hao,ihnp4,decwrl,allegra,tektronix}!ames!aurora!eugenefritz@vlsi.Caltech.Edu (fritz nordby) (05/31/90)
Hello. I'm looking for all the benchmarks I can find. I know about SPEC, dhrystone, whetstone, linpack, the Livermore kernels, and a few others. What I'm looking for are (1) what people have used as benchmarks, and (2) where I can get the code they ran. The types of benchmarks I'm primarily interested in are CPU/memory benchmarks, but please, don't hold back, if you know of something, send it to me. I'm particularly interested (at the moment) in a group of small CPU benchmarks: the Sieve of Eratosthenes, Ackermann's function, Towers of Hanoi, Baskett's puzzle, quick-sort, the 8-queens problem. These benchmarks were popular around the time of the original RISC papers, and in some cases I have been able to find the algorithm, but I'm searching for some of the algorithms and most of the input data sets. If you know anything about these benchmarks, I'd very much appreciate knowing what you can tell me. Again, this is a rather peculiar request or rather particular interest. Please respond via e-mail. If there is a general interest, I will summarize what I find. Thank you very much for your time, and for any help you can give. Fritz Nordby. fritz@vlsi.caltech.edu cit-vax!cit-vlsi!fritz
jan@ivory.SanDiego.NCR.COM (Jan Stubbs) (06/02/90)
In article <15049@cit-vax.Caltech.Edu> fritz@vlsi.caltech.edu (fritz nordby) writes: > >Hello. > >I'm looking for all the benchmarks I can find. >I know about SPEC, dhrystone, whetstone, linpack, >the Livermore kernels, and a few others. What I'm >looking for are (1) what people have used as benchmarks, >and (2) where I can get the code they ran. > IOCALL RESULTS 10,001 Iteration Version, 9/27/87 SYSTEM UNIX VERSION SYSTEM TIME SECONDS ----------- ---------------- ------------------- Dec Pro-380 2.9 BSD 184 MicroVax I Ultrix V1.1 180 Altos 68000 SIII, Altos v2.0a 178.6 Dec Rainbow Venix/86 v2.0 177.5 DEC Rainbow100 w/NECV20 Venix/86 148 *d Onyx C8002s Z8000 SIII 137 Onyx C8002 Z8000 v7 130 Symmetric 375 10MHz NS32016 4.2BSD 128.7 TIL NS32016 9MHz No Wait states Local Port 122 VAX 11/750 BSD 4.2 114.7 PC/AT w/Sritek M68000 SV/68 Rel2 V1.0 114 Sequent Balance 8000 6 CPU Dynix V2.1 113.8 *b Sequent Balance 21000 6 CPU 110.5 *c Symmetric 375 101.4 PC Limited 286-6 6MHz 80286 SCO Xenix SV/286 V2.1.3 100.5 Tandy 6000 8Mhz M68000 Xenix 3.0 109 ATT 3b2/300 SV 103 VAX 11/750 4.2 BSD 100 VAX 11/750 Ultrix V1.2 8 MHz 80286, 1 wait state Microport SV/AT286 2.2u 98.5 PLexus P35 12.5 MHz M68000 SIII 98 ICM-32016 10 MHz NS 32016 SV.2 98 PDP 11/44 ISR 2.9 BSD 95 Motorola S2000 10MHz M68000 SV/68 Rel1 94 Concurrent XF/200 (PE7350A) ? 93 ICM32332 Beta version 90.2 VAX 11/750 4.3 BSD 90 Sun-2 10MHz 68010 4.2 BSD Rel 2.0 90 Sun-2 4.2BSD Rel 3.2 88 VAX 11/750 SV.2 88 Sun-2 10MHz 68010 4.2 BSD Rel 3.0 87 Plexus P60 M68000 SIII 87 PE 3220 V7 Workbench 85 *a ATT 3b2/400 SV.2 83 VAX 11/750 research version 8 81 HP9000-530 2 CPU's 75.0 *c HP9000-320 16MHz 68020 73.4 Apple Mac II 16MHz M68020 A/UX 5.2 r1 (SV.2) 73.1 *n NCR PC-8 8MHz 80286 SCO Xenix-286 SV R2.0.4 72.7 VAX 11/750 4.1 BSD 72 Radio Shack 16A Xenix (v7) 72 Altos 886 80286 Xenix 3.2fs3 71.1 Sperry IT 8MHz 80286 Xenix 5.0 71 ICL DRS300 8MHz 80286 DRS/NX (SV.2) 70.7 VAX 11/750 4.1BSD (lightly hacked) 70 PC/AT Venix 5.2 68 Encore Supermax NS32032 SV.2 Version 4 68.0 *c VAXStation 2000 4.3BSD+NFS (wisconsin) 67.5 Arete 1100 M680?0 SV.2 65 *c ATT7300 Unix PC 10MHz 68010 SV.2 64 IBM PC/RT 170Ns 4.2BSD 64 Concurrent 3230 Xelos Rel R01 (SV) 64 MicroVAXII 4.3BSD+NFS (wisconsin) 63.8 Convergent MiniFrame CTIX 3.20 63.3 Gould PN6080 UTX 1.1C 62 MicroVaxII Mach4.3 61.7 Pyramid 90x w/cache OSx2.5 61 Apollo DN300 10 MHz M68000 Domain/IX 60 *e IBM PC/RT 170MHz ? 60 Pyramid 90x w/cache OSx2.3 58 ATT 3B15 SV.2 56.1 Plessey Mantra 12.5Mhz 68000 Uniplus SV Release 0 55 MicroVax II Ultrix/32-m V1.2 53.4 VAX 11/780 4.2 BSD 53 Concurrent 3250XP Xelos Rel R01 (SV) 52 MicroVax II Ultrix 1.1 52 HP9000-550 3cpu's HP-UX 5.01 51 *c PC/AT 7.5 Mhz Venix286 SV.2 51 Sun 3/50 SunOS 3.2 50.9 Sun 3/52 16MHz 68020 50.1 VAX 11/780 SV.2 50 Convex C-1 4.2 BSD 46 AT&T 3b20S SV.2 44.4 Alliant FX/8 2 IPs, 4 CEs Concentrix 2.0 (4.2BSD) 43.3 *c IBM PC/RT 4.2A (4.2BSD) 42.8 IBM 4341 II UTS 2.4(V7 on VM) 42 PC/AT 9MHz 80286 Venix SV.2 40.9 Gould PN 6040 UTX/32 1.2 39.7 Plexus P/60 12.5 MHz 68020 Rel 1.5 of SV.2 38.3 VAX 11/785 4.3 BSD 36 Sun-3/75 16.67Mhz 68020 4.2 BSD 36 Sun-3/160M-4 16.67Mhz 68020 4.2 BSD Rel 3.0 Alpha 36 Altos 3068 16.67 MHz SV.2 34.7 DG MV10000 DG/UX 3.00 33 CT MightyFrame S/320 68020 CTIX (SV.2) 32.1 Compaq 386-130 16Mhz 80386 SCO Xenix-386 2.2.1 SV3 31.2 HP9000/350 25MHz M68020 HP-UX 5.22 (SV) 30.2 Apollo Dn330 12Mhz M68020 Domain/IX 30 *e VAX 11/785 SV.2 30 Celerity 1230 Accel/32 (NCR/32) 4.2 BSD 28.6 Masscomp 5600 16Mhz 68020 RTU 3.1 (SV) 28.5 Gould Concept/97 UTX/32 1.2 (4.3BSD/SV2) 28 GEC 63/40 S 5.1 27 Gould PN9080 UTX 1.2 (4.3BSD) 25 ICL Clan 4 Model245 12Mhz 68020 Uniplus+ V.2 24.5 Diab DS 90-20 16MHz 68020 D-NIX 5.2.1.2 (SV) 23.9 Sun 3/260 SunOS 3.2 20.2 Gould PN9080 UTX 2.0 (4.3BSD) 19.4 Sperry 7000/40 aka CCI 6/32 4.2 BSD 19 ConvergentServer/PC 20MHz 80386 SV.3 19.0 MIPS M/500, 8Mhz R2000 UMIPS-BSD 2.0 (4.3+NFS) 19.0 VAX 11/785 Ultrix 2.0 18.9 Sun 3/280 25MHz M68020 SunOS 3.3 18.5 HP9000-840 (RISC) HP-UX (4.2BSD) 17.2 Edge I/Model 1100 130NS. GSX 3.1 (SV2) 14.9 Harris HCX-7 HCX-UX Vers. 2.2 (SV+) 14.5 *n MIPS M/500, 8MHz R2000 UMIPS-V 1.1 (SVR3) 13.7 VAX 8600 4.3 BSD 12 VAX 8600 Ultrix 1.2-1 11 Sun 4/260 Sunrise SunOS 11.0 IBM 3083 UTS SV 10 Cray 2, 4NS Clock, 4 CPU Unicos 10.4 *c, *d Cray 2, 4NS Clock, 1 CPU Unicos 9.4 *d Amdahl 470/V8 UTS/V (SV Rel 2,3)V1.1+ 9 MIPS M/800, 12.5Mhz R2000 UMIPS-BSD 2.0 (4.3+NFS) 8.4 MIPS M/1000, 15MHz R2000 UMIPS-BSD 2.01 7.1 VAX 8550 Ultrix 2.0 3.8 Cray X/MP-24 SysV (Pre release 8) 3.8 Amdahl 5890 UTS SV.2 3.63 Notes: *c Multi-cpu system. IOCALL was run single thread, which probably did not utilize all cpu's. This system probably has considerably more power than is reflected by the result. A better measurement of this system's capability is to run as many copies of IOCALL as there are processors, in a script with unique file names, and report the longest of the resulting system times divided by the number of cpus, which should be about the same on each processor, but longer than the single copy time. *d This time was run on a busy system, so it probably is not the best time that is attainable on a idle system. A busy VAX time is about 20% worse than the idle time, but whether this applies to other machines is unknown. *e Real time reported because system time appeared to be unreasonable. Some implementations of Unix kernel don't charge for CPU time to do IO properly. *m This time uses a modified version of IOCALL for a mapped file under the Mach variant of Unix. The time is much better because the mapped file interface doesn't use system calls, and is ideal for this repetitive and unlikely program. Send your results to me directly. The benchmark is a "C" program which measures Unix kernel performance. To run it put the source below in iocall.c, then: cc iocall.c -o iocall time iocall Send all 3 times (user, system, real), but I am reporting the system time only. The user time for this benchmark should be insignificant. The real time should be about equal to system time plus user time, if not you aren't running a real Unix, or your Unix has a bug. (Some people have reported finding a bug in their port after running IOCALL). On BSD systems, which report the number of IO's from TIME command, the number of IO's should be 2 or 3 for the open, and a few paging io's to read in the program. Please also send: 1)Type of machine and model #. 2) Brand, model and clock rate of Microprocessor if any. 3) Version and name of OS, and its ancestry (e.g. SV2 or BSD 4.2) The opinions expressed herein are those of the author. Your mileage may vary. The times herein are obtained from unreliable sources, rely on them at your peril. Benchmark should be run on an otherwise idle machine. If you can please run them so, it does improve the timings. COMMENTARY: What does this benchmark measure? It attempts to simulate a typical mix of reading, writing and seeking. The cpu time used in the Unix kernel is reported by the kernel. It exercises the system call interface in a way less trivial than the getpid benchmark. It does not measure and is independent of, your IO hardware, and drivers. It does seem to show differences in Unix kernel efficiency on the same hardware. It will exercise heavily your caches, and perhaps your block move bandwidth. NO BENCHMARK IS PERFECT, (except your application), but this one shows what a very IO intensive workload with good buffer cache hit rates runs like on your cpu. Many synthetic benchmarks are criticized for giving unrealistic results when run through optimizers that may throw out stuff that does nothing useful. This is NOT a problem with IOCALL. If your compiler finds something in the UNIX kernel that does nothing useful and throws it out, MORE POWER TO IT! -------cut----cut------cut------------------------------- /*This benchmark tests speed of Unix system call interface and speed of cpu doing common Unix io system calls. */ char buf[512]; int fd,count,i,j; main() { fd = creat("/tmp/testfile",0777); close(fd); fd = open("/tmp/testfile",2); unlink("/tmp/testfile"); for (i=0;i<=10000;i++) { /*do seek, write, seek, read, read, read. */ lseek(fd,0L,0); count = write(fd,buf,500); lseek(fd,0L,0); for (j=0;j<=3;j++) count = read(fd,buf,100); } } -----cut---cut---cut---cut----------------------------------------- "There are lies, damn lies, and benchmarks." Jan Stubbs ....sdcsvax!ncr-sd!stubbs 619 485-3052 NCR Corporation Advanced Development 16550 W. Bernardo Drive MS4010 San Diego, CA. 92127