FYI, here are somewhat related papers of performance-oriented colleagues:
Dynamic Monitoring of High-Performance Distributed Applications, D. Gunter,
B. Tierney, K. Jackson, J. Lee, M. Stoufer, Proceedings of the 11th IEEE
Symposium on High Performance Distributed Computing, HPDC-11, July 2002,
http://www-didc.lbl.gov/papers/HPDC02-HP-monitoring.pdf
Monitoring Data Archives for Grid Environments, J. Lee, D. Gunter, M.
Stoufer, B. Tierney, Proceeding of IEEE Supercomputing 2002 Conference,
Nov. 2002
http://www-didc.lbl.gov/papers/Monitoring-archive-SC02-extended-abstract.pdf
Wolfgang.
At 15:49 24/6/2002 +0200, Wolfgang Hoschek wrote:
>>In early 2002, a user (I can't find the name since Bugzilla is down)
>>filed a bug report claiming that on a heavily used server machine
>>adding buffered IO to FileAppender gave a perceptible boost to logging
>>performance. As a response we added buffered IO to FileAppender.
>
>Perhaps you are referring to my message which was about the size of buffering.
>In some environments varying buffersize can help (try 64 bytes vs. 8KB or
>64KB).
>There is also a bugzilla ref to someone elses experience with async logging...
>
>http://marc.theaimsgroup.com/?l=log4j-dev&m=99771529009797&w=2
>
>Wolfgang.
>
>
>
>--
>To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]>
>For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
>
-----------------------------------------------------------------------
Wolfgang Hoschek | e-mail: [EMAIL PROTECTED]
CERN IT Division | phone: +41 (22) 767 8089
European Organization | fax: +41 (22) 767 7155
for Nuclear Research | home: http://cern.ch/hoschek
-----------------------------------------------------------------------
--
To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
