[systemd-devel] We are working on trying to scale up to > 1000 containers.

David Strauss david at davidstrauss.net
Mon Jun 24 09:55:10 PDT 2013


In our case, the issue wasn't kernel process creation; it was the CPU
and I/O overhead of service start-up. At some point, the system gets
dominated by context-switching, and throughput suffers. For example,
you certainly wouldn't want the box to go into swap because of
start-up allocation spikes.


More information about the systemd-devel mailing list