Performance Problem

I tested GigaSpaces over one month, great functionality, great product . But the performance of GS is still a problem for me.

I used GS as a distributed partitioned in-memory cache . In my prototype, I store over 2 millions POJOs in space, and I need aggregate 2 millions values in POJOs each time, which means about 2 millions read Operation , and 0.2 millions write operation each time.

As I have to use remote space for partitioned topologies , I got 1.8K Operations/Sec. For totally 2 millions operations, this is far more from expected, even if I improved performance 4 times by using Externalizable Support. And from your documents, [ http://www.gigaspaces.com/wiki/displa... ] all the operation to remote space is serialized each time.

So for my Read/Write heavy application, is GS the wrong solution for me , or still I miss some points of GS?

{quote}This thread was imported from the previous forum. For your reference, the original is [available here|http://forum.openspaces.org/thread.jspa?threadID=1948]{quote}

asked 2008-03-19 05:35:47 -0500

peng0 gravatar image

updated 2013-08-08 09:52:00 -0500

jaissefsfex gravatar image
edit retag flag offensive close merge delete