Re: Some mmap observations compared to Linux 2.6/OpenBSD

From: Vivek Pai (vivek_at_CS.Princeton.EDU)
Date: 11/03/03

  • Next message: Doug Barton: "Re: I need help fixing the agp_nvidia.c driver" 
    Date: Mon, 03 Nov 2003 00:27:28 -0500
To: David Schultz <das@FreeBSD.ORG>

    David Schultz wrote:
    >
    > Okay, I guess SpecWeb99 is ``real world'' enough for me to justify
    > the assertion that there is an mmap() performance problem. Just
    > out of curiosity, how many regions did SpecWeb99 map?
    > (i.e. what does 'dd if=/proc/$pid/map bs=64k count=1 | wc -l' give?)

    The SpecWeb99 benchmark is self-scaling - the higher request rate you
    try to achieve, the more files are included in the data set. It's
    entirely up to the server to decide how to go about dealing with the
    data, and the Flash server used to keep a cache of mmap'd regions.
    Files smaller than 64KB were mapped as a single region, while larger
    files were broken into 64KB pieces, basically to avoid exhausting the
    address space.

    For the kinds of throughputs that we achieve, we're dealing with
    about 13,000 files. We probably had a number of regions somewhere in
    that ballpark, but I don't have a free machine at the moment to give
    you an exact number.

    Our new server basically gave up on using mapped files entirely due
    to the problems with scaling to larger memory sizes - that's why I'm
    pushing for the sendfile changes I mentioned in a different mail.

    -Vivek

    _______________________________________________
    freebsd-hackers@freebsd.org mailing list
    http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
    To unsubscribe, send any mail to "freebsd-hackers-unsubscribe@freebsd.org"

  • Next message: Doug Barton: "Re: I need help fixing the agp_nvidia.c driver"

    Relevant Pages

    • ODBC / mssql very slow
      ... In my current project I have a performance problem. ... That same queryX takes 2 seconds to complete when executed in SQL Server Management Studio. ... I have the impression my slow performance is caused by the fact that ODBC fetches each record one by one. ... "There are two ways of constructing a software design: One way is to make it so simple that there are obviously no deficiencies, and the other way is to make it so complicated that there are no obvious deficiencies. ...
      (comp.databases.ms-sqlserver)
    • RE: performance problem with dbd oracle
      ... I am running all these tests on the same server - does the network latency ... oracle Solaris and we didn't see the performance problem with that combo. ... Shouldn't network latency have affect DBD::mysql also? ... >> The performance problem is actually worse with the real app which I have ...
      (perl.dbi.users)
    • ODBC (mssql) fetches results really slow
      ... In my current project I have a performance problem. ... That same queryX takes 2 seconds to complete when executed in SQL Server Management Studio. ... I have the impression my slow performance is caused by the fact that ODBC fetches each record one by one. ... Is there some way to make ODBC fetch all records faster? ...
      (microsoft.public.sqlserver.odbc)
    • hp dl-380 and freebsd 6.4,7.1 problem
      ... I have two hp dl-380 server with 6.4 and 7.1 freebsd, but all of them have a poor performance. ... Reference machine (AMD 2500 Athlon, pata disk, 6.4 bsd with generic kernel): ... # time tar xvzf php-5.1.4.tar.gz ... The performance problem is come forward again while executiing a php script, its 10x slower then running same script on the reference machine. ...
      (freebsd-hackers)