Author dino.viehland
Recipients dino.viehland, eric.snow, methane
Date 2019-05-30.18:03:30
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1559239410.96.0.187067751517.issue36839@roundup.psfhosted.org>
In-reply-to
Content
In the Instagram case there's about 20mb of byte code total and there are 3-4 dozen worker processes running per server.  The byte code also represents the second largest section of memory as far as serialized code objects are concerned, the only larger one is strings (which are about 25mb).  
 
Anyway, that leads to around 1gb of memory savings per server, which is the reason why this was an interesting optimization to pursue when it's applied to many servers.
History
Date User Action Args
2019-05-30 18:03:30dino.viehlandsetrecipients: + dino.viehland, methane, eric.snow
2019-05-30 18:03:30dino.viehlandsetmessageid: <1559239410.96.0.187067751517.issue36839@roundup.psfhosted.org>
2019-05-30 18:03:30dino.viehlandlinkissue36839 messages
2019-05-30 18:03:30dino.viehlandcreate