tag:blogger.com,1999:blog-31421954.post8974082837638323630..comments2023-10-30T08:23:12.960-07:00Comments on mySQL DBA, Architecture, Dev, Scale, HA, Code : What do you think about adding ZLIB to memcache storageDathan Pattishallhttp://www.blogger.com/profile/00356367514107959723noreply@blogger.comBlogger8125tag:blogger.com,1999:blog-31421954.post-74511055793712107302009-04-01T01:10:00.000-07:002009-04-01T01:10:00.000-07:00I think server-side compression would be a great t...I think server-side compression would be a great thing in memcached server, especially when client is already under heavy CPU load and cannot afford to compress. I'd very much like to see this functionality in memcached.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-31421954.post-57345583390306826862009-03-31T14:15:00.000-07:002009-03-31T14:15:00.000-07:00@Nathan what client does this, I thought it was co...@Nathan what client does this, I thought it was compression for net bytes only. For instance the PECL memcache lib doesn't indicate the data is stored compress.Dathan Pattishallhttps://www.blogger.com/profile/00356367514107959723noreply@blogger.comtag:blogger.com,1999:blog-31421954.post-6498411646517865262009-03-31T14:10:00.000-07:002009-03-31T14:10:00.000-07:00@topbit what client does this already? It handles ...@topbit what client does this already? It handles zlib compress automatically?<BR/><BR/>@krow: yes (I think you mean values)<BR/><BR/>@mark: on the client I'm running out of CPU cycles.Dathan Pattishallhttps://www.blogger.com/profile/00356367514107959723noreply@blogger.comtag:blogger.com,1999:blog-31421954.post-7583853709898926872009-03-31T13:58:00.000-07:002009-03-31T13:58:00.000-07:00Just do it in the client - it's more scalable ther...Just do it in the client - it's more scalable there. I'm assuming you generally have more clients than servers ( this is not necessarily true in every configuration)Mark Robsonhttps://www.blogger.com/profile/15864507044869250062noreply@blogger.comtag:blogger.com,1999:blog-31421954.post-7564759940730518372009-03-31T12:41:00.001-07:002009-03-31T12:41:00.001-07:00It's already an option in a number of clients, gen...It's already an option in a number of clients, generally triggered on the size of the value being put into cache.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-31421954.post-29397752910168961702009-03-31T12:41:00.000-07:002009-03-31T12:41:00.000-07:00You mean extending libmemcached/etc to do compress...You mean extending libmemcached/etc to do compression on keys before sending?<BR/><BR/>I'd take a patch on this.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-31421954.post-14939076869292141092009-03-31T12:40:00.000-07:002009-03-31T12:40:00.000-07:00Compress on the client side, that's what the flags...Compress on the client side, that's what the flags byte is for and that's what happens already in most client libs I've seen. Client-side cuts down on network traffic as well.Nathanhttps://www.blogger.com/profile/03653814933920551829noreply@blogger.comtag:blogger.com,1999:blog-31421954.post-65303927847477621962009-03-31T12:01:00.000-07:002009-03-31T12:01:00.000-07:00Sounds reasonable, as an option. It would not work...Sounds reasonable, as an option. It would not work well for all workloads.rsynnotthttps://www.blogger.com/profile/12125935382858758107noreply@blogger.com