|
|
|
|||||
|
IceStorm throughput
I rewrote the demo/IceStorm/clock example in Java to conduct some throughput tests. Running a single subscriber and a single publisher on dual 2.8 GHz atholon/xeon computers on a gigabit ethernet, the number of events received by the subscriber ranges from 5000 to over 17000 events per second. The average is roughly 6000 events/second. Do these results seem reasonable or consistent with what you would expect from IceStorm?
Thanks!
__________________
Janet Tvedt |
![]() |
| Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
| Thread Tools | |
| Display Modes | Rate This Thread |
|
|
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| Throughput Performance in Java | jae | Help Center | 2 | 12-05-2006 02:47 PM |
| Server-side throughput performance measures? | SteveWampler | Help Center | 3 | 08-31-2004 07:53 PM |
| Ice vs. JNI throughput performance? | brian | Help Center | 4 | 06-11-2004 02:17 AM |
| AMD/AMI throughput | roelof | Help Center | 10 | 10-23-2003 10:20 AM |
| Making C++ throughput demo oneway? | SteveWampler | Help Center | 4 | 09-02-2003 06:14 PM |