Hey,
so I have some good news. I've re-ran my tests for memory consumption where I store 100MB of data and here's the results for in-VM mode and client-server with HotRod:

Previously, the memory overhead for HotRod was (can be found at http://infinispan.blogspot.cz/2013/01/infinispan-memory-overhead.html):
entry size -> overhead per entry
512B       -> ~174B
1kB        -> ~178B
10kB       -> ~176B
1MB        -> ~245B

Now it is:
512B  -> ~159 (~ -9%)
1kB   -> ~159 (~ -11%)
10kB  -> ~154 (this is perhaps affected by the low number of entries stored in mem.)
1MB   -> ~184 (the same applies here)


For in-VM mode it was:
512B -> ~151B
1kB  -> ~151B
10kB -> ~155B
1MB  -> ~143B

Now it is:
512B  -> ~150
1kB   -> ~150
10kB  -> ~149 (this is perhaps affected by the low number of entries stored in mem.)
1MB   -> ~143 (the same applies here)


Summary: Almost no change for in-VM mode but a nice improvement for the HotRod client.

I can make a blog post once we have this for Memcached and REST. I guess it is not ready yet.


Cheers,
Martin


Dne 6.5.2013 08:57, Martin Gencur napsal(a):
Hi Mircea,
yes, I'll try to run the tests this week.

Martin

Dne 3.5.2013 00:40, Mircea Markus napsal(a):
Hi Martin,

The server memory consumption should improve now that we have ISPN-2281 integrated. would it be possible to run the memory consumption tests to see where we are?

Begin forwarded message:
I've just integrated the implementation for ISPN-2281 - great stuff!


As this is a very intrusive change it is important to know to measure the gain in memory consumption. i.e. would be great to see some figures on memory usage before this optimisation and after it in your blog post.
^ QE guys have some memory tests already, why don't we get them to re-run those tests? :)
Cheers,