IUBio Biosequences .. Software .. Molbio soft .. Network News .. FTP

performance problems in loading

Dave Matthews matthews at greengenes.cit.cornell.edu
Wed Apr 26 06:51:44 EST 2000

jean said:

: Date: Wed, 19 Apr 2000 13:29:34 +0100 (BST)
: You fly while you fill the cache
: but then it should stay fast
: you describe gioing to 1 record/minute which ius ridiculously slow
: so i suppose you have defines a cachesize much larger than you actual 
: memeory and are swapping
: acedb should not swap
: please check wspec/cachesize.wrm , the size declared should be a good
: proportion of your real ram

Xiaowu Gai reported back that the problem was indeed his cache sizes, but
in the opposite direction.  He originally had 
CACHE1 = 4000
CACHE2 = 4000
on a machine with 0.5 GB memory.  (I'm not sure if that's RAM or total.)
The fix:

> I increased the cache sizes to the same as the ones that you listed [16000,
> 32000] and it solved the problem. I loaded the whole 53,000 sequences in 10
> minutes. I also tried to increase them a little more and it became even
> faster.


More information about the Acedb mailing list

Send comments to us at biosci-help [At] net.bio.net