*** J1m has quit IRC | 00:43 | |
*** J1m has joined #zodb | 01:24 | |
*** J1m has quit IRC | 01:34 | |
*** J1m has joined #zodb | 03:01 | |
*** J1m has quit IRC | 04:12 | |
*** JaRoel|4d has quit IRC | 09:25 | |
*** JaRoel|4d has joined #zodb | 10:07 | |
*** JaRoel|4d has quit IRC | 10:09 | |
*** JaRoel|4d has joined #zodb | 10:09 | |
*** saily has joined #zodb | 10:46 | |
saily | hi all, i've want to do some relstorage related discussions here. I've debugged the code and discovered the memcached-cache is shared over zope threads. This could save a lot of memory on my large installation with 16 instances. So if i'm right, this could bypass the disadvantage of having multple caches on multiple zope threads and can reduce or completly disable the zodb-cache-size. is this a good or bad idea? | 10:51 |
---|---|---|
*** fdrake has quit IRC | 14:05 | |
*** J1m has joined #zodb | 15:58 | |
*** fdrake has joined #zodb | 15:59 | |
*** J1m_ has quit IRC | 16:01 | |
*** J1m has quit IRC | 16:16 | |
*** J1m has joined #zodb | 16:17 | |
*** JaRoel|4d has quit IRC | 18:26 | |
davisagli | saily: but memcache only stores pickles, so will be slower than the ZODB cache | 19:33 |
J1m | davisagli, I'm not sure he understand what you're assuming :) | 20:01 |
J1m | saily, memcached in relstorage is an alternative to ZEO's client disk cache, not to the ZODB in-memory cache. | 20:02 |
*** JaRoel|4d has joined #zodb | 20:32 | |
*** JaRoel|4d has quit IRC | 20:32 | |
*** JaRoel|4d has joined #zodb | 20:32 |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!