Noticed newer versions of vbulletin

[CMullen]

Being used on the DTR and my jeep forum, wondering when TDR will upgrade, peformance on the newer release seems to be better on DTR.



Just wondering Steve ??

 

[Steve St.Laurent]

Actually the DTR runs an older version of vB than we do (don't know about the Jeep site you are on) - they are running 3. 0. 8 and we are running 3. 0. 9 (with the security updates from 3. 0. 10, 3. 0. 11, and 3. 0. 12 installed). You really shouldn't see much difference performance wise on either site except when you are loading threads here that have pictures linked to other servers because I am reformatting those pictures so that they don't screw up the formatting on the page - because of the way that's done if there is a problem with the other server it can significantly slow down the page loads (5 second timeout per picture). The alternative to that is if people link to pictures that are wider than your screen then you'll have to scroll left and right in order to read the entire thread. I personally don't see any difference in speed between the two sites from my location. You could be closer to their server (have fewer hops on the backbone getting to them than to us) which could make a difference. Keep in mind too that there is 50% more traffic here.



3. 0. 10-3. 0. 12 have been bug fix releases - mostly for security reasons (which I implemented manually). I am currently working on upgrading to 3. 5. 3 which is the next major release - I've been working on that since early December. vBulletin released version 3. 5 back in September and I've been monitoring it since to see how it's working for people with large sites. Large sites such as ours always wait for a while to see how new major releases are working before implementing them. There was a major change in how the code works and we have a lot of hacks installed here for functionality that I'm having to convert over so it's taking some time.



Although there are currently no issues with hardware performance we will be upgrading hardware later this year. I monitor the server performance on a daily basis and they very rarely are over 50% CPU usage (at most for 20 minutes each day). Currently we are running on two high end servers - one for the web traffic and one for the database. We will be moving to a load balancer running 2-3 servers just for the web pages and the existing 2 servers on the back end for the database.

 

[CMullen]

Thanks for the info steve, i stand corrected