Re: summary of mini mock data challenge

From: Konstantin Olchanski (olchansk@ux1.phy.bnl.gov)
Date: Thu Feb 03 2000 - 18:45:44 EST

  • Next message: videbaek: "Cygnus & win95"

    >   f.. Data were only transferred for a short while from the BRAHMS DAQ to RCF. (Monday evening). HPSS returned errors. 
    >   g.. Since Phobos did not take part in this MDC Bruce will like to repeat this likely by end of February. 
    >   h.. Star achieved DAQ -> HPSS sustained rates of 18Mb/sec over the 2 days. 
    > ------------------------------
    
    
    The failure in the DAQ to HPSS transfers was traced to an error in my
    script. I was misinterpreting the HPSS error and was not creating
    the right directory before trying to "pput" a file into it.
    
    After the error was corrected, the data started moving again,
    at peak transfers rates of 12-16 Mbytes/sec.
    
    This is a bit low (the disk spool on opus should sustain up to 20 Mbytes/sec),
    but consistent with the STAR numbers (they use MTU=9000 "jumbo" frames,
    we use MTU=1500, so their data ought to move faster).
    
    While running the sending scripts I see a lot of HPSS "hickups",
    for example, sometimes HPSS logins "hang" and some HPSS
    commands sometimes take a long time to complete.
    
    Also, Tom Throwe's scripts that report free HPSS disk cache space
    no longer work, and I now can easily (and happily did) use up all the
    cache space. Tom is working on fixing this problem.
    
    
    -- 
    Konstantin Olchanski
    Physics Department, Brookhaven National Laboratory, Long Island, New York
    olchansk@bnl.gov
    



    This archive was generated by hypermail 2b29 : Thu Feb 03 2000 - 18:47:55 EST