Date: Mon, 9 Dec 2002 09:21:00 -0500 From: Mikhail Teterin <mi+mx@aldan.algebra.com> To: Michael.Schuster@sun.com, freebsd-sparc@freebsd.org Subject: Re: why would Sparc be soo sloow? Message-ID: <200212090921.00890.mi%2Bmx@aldan.algebra.com> In-Reply-To: <3DF45D27.AC42361B@sun.com> References: <3DF45D27.AC42361B@sun.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Monday 09 December 2002 04:06 am, Michael Schuster wrote: = Mikhail, = = (I read -digest, so this may already be answered) = = > I'm puzzled by the poor performance of our SparcIII @900MHz. = = You need to give us some more information to even give an educated guess, = things like = - memory installed = - no# of CPUs installed Two SparcIIIs on the SunFire, one Pentium4 on the Dell. But the program is single threaded/processed. = - other HW config aspects, like storage in use, etc for both (!) = machines, as well as characteristic of the application, such = as = - memory footprint = - I/O behaviour, etc. Very little -- all the program is doing is computations -- the CPU is the bottleneck on both machines. It calls the same financial formula over and over -- for the same set of numbers. The numbers -- 8 double values per call -- are read using scanf(3) from a file, and the result -- 7 double values per call are written with a printf(3) to /dev/null. There are no file being opened/closed by the program at all... = Is anything else happening while you're testing, or do you have = exclusive access to the machines? Yes, there is. But not very much. And, once again, I'm measuring the CPU time, not the total time. = - iostat = - vmstat * = - mpstat * = - prstat = For obvious reasons, I cannot offer to analyse any data for you, I = hope though that this will get you going. I don't believe those utilities are going to help much -- this is a very simple program, with virtually no IO. The simple top(1) shows, that it is only limited by the CPU, and I remain puzzled by the speed difference, as well as by the high "system" time component on Solaris: mi@attila:~/fxpr/test (70) time ./fxprt < ../tasks.txt > /dev/null 819.11u 819.41s 28:35.00 95.5% mi@attila:~/fxpr/test (71) /usr/bin/time ./fxprt < ../tasks.txt > /dev/null real 28:27.5 user 13:35.6 sys 13:39.1 Very suprising, for, according even to truss(1), the program makes no system calls at all after the few at startup -- except for the very occasional read(2) and write(2) (to /dev/null). Indeed, on FreeBSD the "sys" part is negligeable. Is there a known problem with sys-time accounting on these systems: mi@attila:~/fxpr/test (76) uname -a SunOS attila 5.8 Generic_108528-13 sun4u sparc SUNW,Sun-Fire-280R mi@attila:~/fxpr/test (77) uname -X System = SunOS Node = attila Release = 5.8 KernelID = Generic_108528-13 Machine = sun4u BusType = <unknown> Serial = <unknown> Users = <unknown> OEM# = 0 Origin# = 1 NumCPU = 2 Yours, -mi To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-sparc" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200212090921.00890.mi%2Bmx>