-
Posts
-
We migrated multiple of our client’s sites over to our new AWS servers using a different OS/Kernel Alma9/RHEL9. The cpu/resources are all the same but it seems like w3tc is slowing down the sites. It’s difficult to pinpoint what configuration is better since multiple sites using the w3tc plugin are on that server\vying for resources and are showing very inconsistent results.
Does anybody have experience with configuring it on Alma9/RHEL9 that they might know what is traditionally a performant configuration (or storage engines to avoid)?
Hello Matt,
Thank you for reaching out and I am happy to help!
Were those websites migrated with the existing W3 Total Cache configuration or was the configuration done after the migration?
Can you please share an example of the website and let me know if the problem is at the frontend or backend and what caching methods are used?
Thanks!The website was configured before migration. here’s an example of old/new server (with fairly similar theme/assets pulled onto the page):
New Server – https://new.learnh2o.com/my-account/ (1-3s server response for page)
Old Server – https://www.prwa.com/my-account (<0.5s server response for page)
It has to be the back-end. Even when running wp-cli it takes 3x longer than the old server. When trying to reconfigure using the wizard I get very sporadic times:
Test 1 for object cache:
None – 3.92s
Disk – 4.04s
Redis – 2.78s
Memcached – 2.94sTest 2 for object cache:
None – 19.68s
Disk – 15.16s
Redis – 9.12s
Memcached – 18.36sTest 3 for object cache:
None – 3.83s
Disk 3.07s
Redis – 3.93s
Memcached – 2.99sBoth sites are using the following caches:
- Page Cache – Disk:Enhanced
- Opcode Cache – Opcode:Zend Opcache
- Database Cache – Disk
- Object Cache – Disk
- Browser Cache
Viewing 3 posts - 1 through 3 (of 3 total)
Viewing 3 posts - 1 through 3 (of 3 total)
