Hi, I use systemd timer to run the archiver. I currently use `/usr/bin/php /srv/matomo/console core:archive --concurrent-archivers=4 --concurrent-requests-per-website=1 --url=”…”` (we run 4 systemd timers, increased it to 4 today to see if it’ll help process faster without causing too much cpu/load usage). I switched to using --concurrent-requests-per-website=1 to try and control the load. Since we have one vm that ones matomo, has 8 cores and 8gb of ram. The DB is in its own vm with 6 cores.
What’s the recommended way to run the archiver for thousands of sites? (we seem to have a million hits atm/d). It’s currently taking 6 hours to run. I’d like to make it sustainable for more sites to be added, for it to run faster and for the load to not outstrip what the cpu can handle in the vm.