This is a discussion on How to calculate MIPS on AIX? within the AIX Operating System forums, part of the Unix Operating Systems category; --> Hello, I find shell scripts to test the power (MIPS by ex.) of different AIX systems. Thanks for your ...
| |||||||
| FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
| |||
| PASDEPUB <PASDEPUB@tixis.arcelor.com> wrote: P> Hello, P> I find shell scripts to test the power (MIPS by ex.) of different AIX P> systems. "Shell scripts" and "MIPS" are odd partners in that sentence. Do you mean simply that you are looking for CPU benchmarks that you can compile (yes, compile) and run on a machine that runs AIX? Regards, Nicholas -- "Why shouldn't I top-post?" http://www.aglami.com/tpfaq.html "Meanings are another story." http://www.ifas.org/wa/glossolalia.html |
| |||
| In article <3810FD4EAD8AD611A50D00805F0DA7FA045D0999@proxyusi nor.usinor.com>, PASDEPUB <PASDEPUB@TIXIS.ARCELOR.COM> wrote: >I find shell scripts to test the power (MIPS by ex.) of different AIX >systems. If you want to measure potential performance of each machine, then you really want to benchmark each machine. This is not simple and is best approached by running the application(s) of direct interest to you. The age-old "MIPS" or measurement of "millions of instructions per second" is of little practical value in determining potential machine performance on applications of interest. It is often snidely referred to as "Meaningless Index of Performance". If you want an idea of the hardware on a given system, then "hinv -c processor" will tell you the basic processor models and clock speeds in the machine. If you use "hinv -v -c processor" you can also get the size and speed of the Scache on each processor. -- Daniel Packman NCAR/ACD pack@ucar.edu |
| |||
| I have 8 different AIX systems on my platform. I want to evaluate their differences as MIPS and TPM. I don't know if it's exist some free program to do that. But, if I execute a same program on my different systems , I can compare them. Thank's, Christophe LEGAL "Nicholas Dronen" <ndronen@io.frii.com> a écrit dans le message de news: 3fd4afe1$0$200$75868355@news.frii.net... > PASDEPUB <PASDEPUB@tixis.arcelor.com> wrote: > P> Hello, > > P> I find shell scripts to test the power (MIPS by ex.) of different AIX > P> systems. > > "Shell scripts" and "MIPS" are odd partners in that sentence. Do > you mean simply that you are looking for CPU benchmarks that you > can compile (yes, compile) and run on a machine that runs AIX? > > Regards, > > Nicholas > > -- > "Why shouldn't I top-post?" http://www.aglami.com/tpfaq.html > "Meanings are another story." http://www.ifas.org/wa/glossolalia.html |
| |||
| In article <br2d67$kk7$1@news.ucar.edu>, Daniel Packman <pack@eos.ucar.edu> wrote: >In article <3810FD4EAD8AD611A50D00805F0DA7FA045D0999@proxyusi nor.usinor.com>, >PASDEPUB <PASDEPUB@TIXIS.ARCELOR.COM> wrote: ...... >If you want an idea of the hardware on a given system, then ..... Oops. Wrong group. That is for IRIX. For aix, "lsdev -C -l proc0" shows little. "lsattr -E -l proc0 -F'description value' " shows the processor type such as power3 or power4 and gives processor speed, "lsattr -E -l sys0' sometimes gives cache sizes. -- Daniel Packman NCAR/ACD pack@ucar.edu |
| ||||
| pack@eos.ucar.edu (Daniel Packman) wrote in news:br2d67$kk7$1@news.ucar.edu: > [...] The age-old "MIPS" or measurement of "millions of instructions > per second" is of little practical value in determining potential > machine performance on applications of interest. It is often snidely > referred to as "Meaningless Index of Performance". Miserable Indicator of Processing Speed. -- Doing AIX support was the most monty-pythonesque activity available at the time. Eagerly awaiting my thin chocolat mint. |