Article View: comp.sys.sequent
Article #598Re: I/O bottleneck?
From: Gene Spafford
Date: Thu, 06 Jun 1991 13:38
Date: Thu, 06 Jun 1991 13:38
20 lines
1423 bytes
1423 bytes
Vincent notes that sometimes with 200+ people on, the CPU is 30%-40% idle, and keyboard response is slow. He concludes it is an I/O problem. That is very possibly correct, but i may not be the Ethernet as he suggests. Memory-starved machines behave the same way. If the CPU is spending a lot of time moving user pages in and out of memory, processes will be blocked while awaiting their pages, and will not appear to be using much of the CPU. Vincent's attempt to type something makes his job active and causes it to fight for the page pool, too, thus giving the "jerky" behavior he notices. Using "vmstat" to observe page faults would be a good idea. It will reveal if this is a problem or not. The cure is not faster I/O, but more memory. -- Gene Spafford NSF/Purdue/U of Florida Software Engineering Research Center, Dept. of Computer Sciences, Purdue University, W. Lafayette IN 47907-1398 Internet: spaf@cs.purdue.edu phone: (317) 494-7825
Technical Details
Message-ID:
<14947@ector.cs.purdue.edu>Path:
arc1-ca.pugleaf.net!utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!swrinde!elroy.jpl.nasa.gov!ncar!gatech!purdue!spafReferences:
<30721@hydra.gatech.EDU>Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!swrinde!elroy.jpl.nasa.gov!ncar!gatech!purdue!spaf From: spaf@cs.purdue.EDU (Gene Spafford) Newsgroups: comp.sys.sequent Subject: Re: I/O bottleneck? Message-ID: <14947@ector.cs.purdue.edu> Date: 6 Jun 91 13:38:28 GMT References: <30721@hydra.gatech.EDU> Sender: news@cs.purdue.EDU Organization: Department of Computer Science, Purdue University Lines: 20