Am 17.04.2013 16:39, schrieb Dietmar Maurer:
3000 VMs all with 32 nics seems unrealistic to me.
BTW, we have several users with >500VMs, so I guess we can easily reach that
limit.
Sure but those won't have 32 NICs. So if we go the way one file per NIC.
We might have 3800 files for 3000 VMs. (800 VMs with 2 Nics so they have
1600 files and the rest with just one nic)
Updating those works pretty good even on IDE and even without rrdcached.
It generates nearly no load at all just 10% I/O Wait. But running so many VMs
shouldn't be done on IDE anyway.
I am confused now. If we have 500 VMs, we have at least 500 rrd files
(currently). If you
add one additional file per VM we have 1000 rrd files. If we assume a VM has
2nics on average
we are at 1500 files...
Sorry i was just testing the netin / netout part. A main point seems to
be the rows in file - so my tests don't cover this scenario i was just
testing 32 in/out values vs. 1 file per nic for in/out.
But right now i would prefer the API query SNMP like way anyway.
Stefan
_______________________________________________
pve-devel mailing list
pve-devel@pve.proxmox.com
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
