This content has been marked as final.
Show 4 replies
-
1. Re: With distribute mode, asynchronous adding key/value to cache is ONLY two times faster than synchronous adding.
mircea.markus Sep 5, 2012 7:10 PM (in response to ywang4362a)Agreed.
I don't think 10k is a big enogh data set, do you get the same results with 100k?
-
2. Re: With distribute mode, asynchronous adding key/value to cache is ONLY two times faster than synchronous adding.
ywang4362a Sep 5, 2012 7:55 PM (in response to mircea.markus)100K key/value: 34 seconds for syn. 22 seconds for asyn.
-
3. Re: With distribute mode, asynchronous adding key/value to cache is ONLY two times faster than synchronous adding.
mircea.markus Sep 5, 2012 8:15 PM (in response to ywang4362a)1 of 1 people found this helpfulDoesn't sound right.
Can you please post the cache configuration(and jgroups file if present) and the test?
-
4. Re: With distribute mode, asynchronous adding key/value to cache is ONLY two times faster than synchronous adding.
ywang4362a Sep 6, 2012 11:13 AM (in response to mircea.markus)The whole test package is attached.
In asserts/config, asserts/config1, asserts/config2, you will find the infinispan (xmatter_infinispan.xml) and jgroups-tcp.xml. They are modified accroding to the sample.
The test source code is in src
The eclipse project file and class file are attached too.
Thanks for your help.
-
test.zip 34.2 KB
-