[QFJ-103] Performance Benchmarking? Created: 09/Nov/06  Updated: 08/Jan/07  Resolved: 08/Jan/07

Status: Closed
Project: QuickFIX/J
Component/s: Engine, Networking
Affects Version/s: None
Fix Version/s: None

Type: Other Priority: Default
Reporter: Ashish Sharma Assignee: Steve Bate
Resolution: Fixed Votes: 0
Labels: None

Issue Links:
Duplicate
duplicates QFJ-94 Performance Closed

 Description   

Are there any benchmarking results for quickfixj engine available? If not, can benchmarking be performed on following parameters :

1. Maximum number of simultaneous sessions to fix engine.
2. Maximum number of orders per second per session.
3. Maximum number of requests (New orders/ Replace/ Cancel) per second per session.
4. Memory, I/O, CPU requirments per session.
5. Network bandwidth required per session.

Optionally, Benchmarking of Broadcast data (market rates) (latency, bandwidth).

If any such work has been done already please let me know. If not what kind of framework can be used to systematically test quickfixj over above mentioned parameters.

Cheers
Ashish

 Comments   
Comment by Steve Bate [ 08/Jan/07 ]

The best plan is to test performance in the context of your business application. Typically, the QFJ session will add neglible overhead compared to the message store (file, JDBC) or application processing.

Generated at Sat Nov 23 07:54:42 UTC 2024 using JIRA 7.5.2#75007-sha1:9f5725bb824792b3230a5d8716f0c13e296a3cae.