Greetings Squiders,
I originally sent the following to our local 'cache' group and recieved no
replies, do I am polling the wider audience.
Greetings,
Swinburne is currently experiencing some problems with our proxy. We are
seeing timeouts from different parts of our networks, and the response
times to retrieve pages from the cache are fairly poor.
When we originally built our cache, we used ADVFS (we are running on a
DEC Alpha using Digital Unix 4.0b) and response was pretty good. When the
entire university was forced through the proxy, response times started to
suffer.
At this time, there was discussion on the squid-users list about ADVFS
for the cache partition. We tried rebuilding our volume and disk and allocating
1024 pages - which is for approx 400,000 files.
Now though we are seeing response problems again, partly due to the cache
and partly due to comms.
My question is this : If I am to rebuild the cache disk(s) as UFS (which is
supposed to be faster for this type of application), do all the disks
have to be the same size ?
I have a 4Gb drive now (existing cache) but can scrounge a 2Gb drive to add.
Is this possible ? Is UFS significantly faster ?
Thanks.
Regards,
Andrew
Unix Systems Administrator Phone : 61 +3 9214-8252
Computer and Network Services Fax : 61 +3 9214-8944
Swinburne University of Technology E-Mail: andrew@swin.EDU.AU
Hawthorn, Victoria, Australia 3122 URL: http://opax.swin.edu.au/andrew
Received on Tue Apr 15 1997 - 04:37:01 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:34:59 MST