Re: LRU writes

On Thu, 2 May 1996, Mickey Mestel wrote: > brian, > > > We have never had anything BUT chunk writes, and are trying to make > > better use of our BUFFERS. Should we be trying to increase the number of > > queues, maintaining the number of buffers, until we just start seeing LRU > > writes? > > it depends on your enviornment. you might be just fine having all > chunk writes, remember, they are the most efficient type of write. if you > are in a situation where your users don't have any problem with system > response time, that is pretty much the end goal. Our response times suck! > is your response time what you want it to be? i have my system > set up 2:1 lru writes:chunk writes. that keeps good response time for my > system overall. I've finally got some LRU writes coming but they are fractions of the chunk writes. > about making use of your buffers, if your %read cached is in the > high 90's, and your %write cached is in the low 90's or above, you are making > good use of your buffers. i don't know if you are trying to achieve > anything more specific than that. Our % read cache is 99.9x% and writes are 98.x We have 10240 buffers distributed in 20 LRUs. LRUMIN is 5, LRUMAX is 10. Would it make sense to increase the number of LRUS, thus having more queues to spread the buffering across? Or does queue management start to become an issue. Checkpoints, every 30 minutes, are 10 secs. B. ------------------------------------------------------------------------- |Brian P. Hampson Internet : brian@asl-labs.bc.ca | |System Administrator, Air(2m/70cm): VE7NNW | |Analytical Service Labs Fidonet : Brian Hampson 1:153/733| |Vancouver, BC | |+604-253-4188 | | Specialists in Environmental Chemistry | | | ------------------http://www.asl-labs.bc.ca/-----------------------------