nghiem@ut-emx.UUCP (Alex Nghiem) (03/08/90)
I may have an opportunity to run benchmarks on a Risc/6000 Friday morning 3-7-90. I would like to locate the benchmarks people are interested in and have them ready for the demonstration. Please send e-mail regarding what benchmarks I should run, where I can get it from the internet, and what caveats I have to avoid so that the tests will be valid. I am interested in version 2.0 of the drystones test, an SCSI performance benchmark if one is avaliable, etc. If you have the source and want to send it to me, please send it to nghiem@ix2.cc.utexas.edu. If unsuccessful, try any of the addresses in my signature. Also, I assume the new systems will DOS read text from a standard 3.5" floppy. If I am wrong, please tell me so. __________________________________________________________________________ This article is posted for factual information only. Any misrepresentation, if any, is purely unintentional. Any opinion expressed or implicit in these remarks are solely my own. nghiem@emx.utexas.edu !cs.utexas.edu!ut-emx!nghiem nghiem@walt.cc.utexas.edu !cs.utexas.edu!ut-emx!walt!nghiem nghiem@utxvm.bitnet !cs.utexas.edu!ut-emx!ix2!nghiem
shawn@jdyx.UUCP (Shawn Hayes) (03/08/90)
I'm interested in seeing your results. I've run both the Dhrystone
benchmarks and the Byte Unix benchmarks on a RS6000/320. The benchmarks
were run on the 8943Q version of the OS. We just got a copy of the
9005 version today. Once I get the Byte benchmarks working fully I'll
post the results. The first impression that I've gotten is that the
compiler has been improved since the Dhrystone benchmark improved by
about 50% over the 8943Q version.
Shawn Hayesnghiem@ut-emx.UUCP (Alex Nghiem) (03/10/90)
In article <25695@ut-emx.UUCP> nghiem@ut-emx.UUCP (Alex Nghiem) writes: >I may have an opportunity to run benchmarks on a Risc/6000 Friday >morning 3-7-90. I would like to locate the benchmarks people >are interested in and have them ready for the demonstration. I did not have a machine to test at the meeting today. We have scheduled a hands on demo for Wednesday, March 14. Our rep will prepare a system with the latest operating system for this demo. Apparently the version of AIX currently installed is a demo version only. It operates slower than the release version will because of debugging traps, etc. >Please send e-mail regarding what benchmarks I should run, where >I can get it from the internet, and what caveats I have to avoid >so that the tests will be valid. I am interested in version 2.0 of >the drystones test, an SCSI performance benchmark if one is avaliable, >etc. I have received the drystone 2.1 demo and the xfroot demo. Did not have a chance to try the x benchmark yet. I tried the drystone on an RT 6150 AIX 2.1.2 (~2100 drystones), a Sun 3 [150?] 4.0.3 (~2800 drystones), a Vax [11/750?] BSD 4.3 (~1600 drystones), and Macintosh II A/UX 1.01 (~2900 drystones). The RISC/6000 is advertised to run ~45K drystones. >If you have the source and want to send it to me, please send it >to nghiem@ix2.cc.utexas.edu. If unsuccessful, try any of the addresses >in my signature. Please send additional benchmarks or tell me where I can find them. I've heard about the Byte Unix benchmarks. I would like to find some kind of SCSI and some ethernet benchmark as well. Also, IBM's results are in Mips and Mflops. Does anybody have the calculations to convert the drystones to this figure or have an actual benchmark? >Also, I assume the new systems will DOS read text from a standard >3.5" floppy. If I am wrong, please tell me so. The RISC/6000 will dosread a 720K or 1.44m 3.5" floppy __________________________________________________________________________ This article is posted for factual information only. Any misrepresentation, if any, is purely unintentional. Any opinion expressed or implicit in these remarks are solely my own. nghiem@emx.utexas.edu !cs.utexas.edu!ut-emx!nghiem nghiem@walt.cc.utexas.edu !cs.utexas.edu!ut-emx!walt!nghiem
nghiem@ut-emx.UUCP (Alex Nghiem) (03/11/90)
In article <25834@ut-emx.UUCP> nghiem@emx.UUCP (Alex Nghiem) writes: >I have received the drystone 2.1 demo and the xfroot demo. Did not >have a chance to try the x benchmark yet. I tried the drystone >on an RT 6150 AIX 2.1.2 (~2100 drystones), a Sun 3 [150?] 4.0.3 >(~2800 drystones), a Vax [11/750?] BSD 4.3 (~1600 drystones), >and Macintosh II A/UX 1.01 (~2900 drystones). The RISC/6000 is >advertised to run ~45K drystones. Our RT is a 6150 Model 25 with the original processor board running under AIX 2.1.2 . In all tests, the standard c compiler was used. Late last night, I downloaded some Unix Benchmarks from directory PD2:<UNIX-C.BENCHMARKS> @SIMTEL20.ARPA . However, the makefile bombed under BSD4.3 on the VAX. __________________________________________________________________________ This article is posted for factual information only. Any misrepresentation, if any, is purely unintentional. Any opinion expressed or implicit in these remarks are solely my own. nghiem@emx.utexas.edu !cs.utexas.edu!ut-emx!nghiem nghiem@walt.cc.utexas.edu !cs.utexas.edu!ut-emx!walt!nghiem
clarke@acheron.uucp (Ed Clarke/10240000) (03/11/90)
From article <25834@ut-emx.UUCP>, by nghiem@ut-emx.UUCP (Alex Nghiem): > and Macintosh II A/UX 1.01 (~2900 drystones). The RISC/6000 is > advertised to run ~45K drystones. Ok, I tried a copy that Dan Ehrlich sent to me from shire.cs.psu.edu using the xlc compiler with "-DTIMES -DHZ=100 -O" and no loader flags. I got 39920 dhrystones and it didn't make any significant difference whether or not I used register variables. The benchmark says "C/2.0" in the source, while my OS build level is 8951(O). This is a floor standing machine, but not the real fast one ( strangely enough, all the faster machines went to the people who put them together ... ). My code wanted a number input through stdio - I used 1000000 to be sure things evened out. Open mouth insert foot time ... not exactly about the RS-6000 - I'd be happier if this benchmark were slightly modified. Take a small driver program that runs twenty copies of Dhrystone 2.0 in parallel. Give me the real elapsed time between the start of the first copy and the death of the last ... use a standard loop count of 1000000. Benchmark should be run on a freshly re-booted system with all the normal ( sendmail, named, NFS, gettys, X11 ) tasks running but only one user ( not the same as 'single user maint mode' ). I'm an "End User", and I don't really care how fast you can calculate floating point numbers and move strings. Gime a real number - how long am I going to sit here waiting for my job to finish? The numbers right now just give user process time. That's ok, but what about paging and task switching and spooling and and and ... -- Ed Clarke | Happiness, n. An agreeable sensation arising from acheron!clarke | contemplating the misery of another. - Ambrose Bierce
nghiem@ut-emx.UUCP (Alex Nghiem) (03/11/90)
In article <25834@ut-emx.UUCP> nghiem@emx.UUCP (Alex Nghiem) writes: >I have received the drystone 2.1 demo and the xfroot demo. Did not >have a chance to try the x benchmark yet. I tried the drystone >on an RT 6150 [Model 25] AIX 2.1.2 (~2100 drystones), a Sun 3 [150?] 4.0.3 >(~2800 drystones), a Vax [11/750?] BSD 4.3 (~1600 drystones), >and Macintosh II A/UX 1.01 (~2900 drystones). The RISC/6000 is >advertised to run ~45K drystones. Add to this list: Next 1.0 ~5900 drystones __________________________________________________________________________ This article is posted for factual information only. Any misrepresentation, if any, is purely unintentional. Any opinion expressed or implicit in these remarks are solely my own. nghiem@emx.utexas.edu !cs.utexas.edu!ut-emx!nghiem nghiem@walt.cc.utexas.edu !cs.utexas.edu!ut-emx!walt!nghiem
nghiem@ut-emx.UUCP (Alex Nghiem) (03/12/90)
In article <1990Mar10.211340.13479@acheron.uucp> clarke@acheron.uucp (Ed Clarke/10240000) writes: >of 1000000. Benchmark should be run on a freshly re-booted system >with all the normal ( sendmail, named, NFS, gettys, X11 ) tasks running >but only one user ( not the same as 'single user maint mode' ). The drystone 2.1 figures I posted earlier for the Sun 3, the NeXT, and the Vax were run in multiuser mode. These machines are NFS'ed together. The tests on the RT 6150 Model 25 and Mac II A/UX were run in multiuser mode, but without NFS. When I ran the tests, the Mac had two users, the Vax had nine users, and the Sun 3 had about five users. I was the only user on the RT and the NeXT. The figure I got on the Next 1.0 machine (~5900 drystones )matches the figure that was posted in the March, 1990 Byte article on Unix Benchmarks. Does anybody know where I can ftp the latest Byte benchmarks? __________________________________________________________________________ This article is posted for factual information only. Any misrepresentation, if any, is purely unintentional. Any opinion expressed or implicit in these remarks are solely my own. nghiem@emx.utexas.edu !cs.utexas.edu!ut-emx!nghiem nghiem@walt.cc.utexas.edu !cs.utexas.edu!ut-emx!walt!nghiem