weird memory consumption vs machine spec
Bizarre scenario, but we had an issue where the apps I run (40 sites running the same code) - were hitting out of memory exceptions since moving to a new bigger server (46 cores, 128GB ram) and the apps were consuming much more memory than they did on our old 24 core server with less ram.
To resolve temporarily I put them on a 32 core, 256GB ram server. (CPU count only changed because its what AWS offered) and the memory usage of each app has dropped back to how they used to behave.
We are currently using about 98GB (same as before we moved to the new 46 core server), so it's as if it needed the ram when only 128GB was available, we upgraded, now it doesnt need it anymore. The app usage appears to be consistent throughout.
Any ideas why this would happen?
To resolve temporarily I put them on a 32 core, 256GB ram server. (CPU count only changed because its what AWS offered) and the memory usage of each app has dropped back to how they used to behave.
We are currently using about 98GB (same as before we moved to the new 46 core server), so it's as if it needed the ram when only 128GB was available, we upgraded, now it doesnt need it anymore. The app usage appears to be consistent throughout.
Any ideas why this would happen?
