RAMDisk and Memcached or only one of it?

[Dennis1993]

Hello,

I would like to speed up our Zammad instance a bit, as we are now growing from 15 agents to around 35 agents, as another department would like to use the system with us. For this I have already doubled the hardware (now 4 CPU, 16 GB RAM) and activated RamDisk for the directory /opt/zammad/tmp. The directory is always only a few MB in size for me (currently 30 MB).

I read in the documentation that memcached can be used. Can I use this for the RamDisk or do I have to/should I deactivate it for this? Does that bring more power afterwards than a RamDisk?

I would be happy about a short info.
I have already set WEB_CONCURRENCY to 4 and ZAMMAD_PROCESS_SCHEDULED_JOBS_WORKERS to 1.

Do you have any other tips for better performance?
We noticed last week that tickets with so many users take 2-3 seconds to open.
I can also pack more hardware underneath, since the system is a virtual machine if needed.

Thanks in advance!

We use: Zammad 5.2.1 (update from yesterday, Ubuntu 20.04.3 with latest postgres and elastic version)

[dvanzuijlekom]

Putting that directory in RAM or using memcached might help. If you run your Zammad instance on spinning disks, this will probably have some sort of positive impact on performance regarding disk I/O. We use SSD’s in our setups, so for us there’s little performance gain of having that data either in RAM or on disk.

You may want to check out some of the troubleshooting tips mentioned here (and read the warnings on blindly following advice):

Performance troubleshooting 5.2.1

Regarding the database deadlocks, you might need to look into PostgreSQL tuning. I have written a bit about this in earlier posts, as we were facing the same sort of issues. This was on a Zammad 4.x system though, but PostgreSQL database tuning and best practices don’t change too much in 5.x. We are currently running 5.1.x and will soon be upgrading to 5.2.x with the same tuning in place. Please do read the comments by Zammad Core Team members about not just blindly copying parameters/advice and…

You might also look into tuning PostgreSQL, if you haven’t already. You can compare some configurations here:

Performance troubleshooting 5.2.1

Re-setting the WEB_CONCURRENCY and ZAMMAD_SESSION_JOBS_CONCURRENT to the values you’ve previously set would probably help you a lot already and might be the solution to most of the sluggishness. The values of 8 sounds reasonable to me based on your hardware specs and my own trial & error experiences (tune a setting, monitor performance, tune a setting, monitor performance… etc). For comparison, here are some stats about a few or our setups. Other than the concurrency settings, you might also wa…