DRBD performance question ?
alanr at unix.sh
Tue Mar 5 14:07:33 MST 2002
Ravi Wijayaratne wrote:
> I was looking at DRBD performance numbers from
>>From Phillip Reisner's paper I gather that protocol A
> is an asynchronous protocol. Therefore protocol A
> should not severely impact the write performance at
> primary server. Is the above assertion correct ?
> If it is so how is it that the performance of protocol
> A on the primary side is seems to be limitted by the
> network B/W ? If protocol A is asynchronous we should
> see a significant difference in throughput between
> protocol A and C. However they seem to be quite close.
> Is this discrepency caused by write ordering or is
> there a hidden bottleneck in the protocol ?
First of all I'm sure that these performance numbers (from a year ago) were
based on 2.2 kernels. In 2.2, DRBD had to suffer the disk I/O scheduling
twice: one at the DRBD level, and one at the real disk layer. So, with 2.4
(where this is avoided), the numbers look a lot different.
As an aside, disks don't write that much compared to their total bandwidth,
and that the smart protocol only rarely has to synchronize.
A dedicated 100mbit connection (which is what he tested with) provides about
8 megabytes/second writes. If you say each write is a 2kbyte block, then
that's nearly 4k block writes per second. Commonly read rates significantly
outweigh write rates. I have measured 10 to 1 ratios on general purpose
development systems. That would mean that it would have to be doing 44k
block I/Os per second, which is a VERY amazing speed for most PC-based I/O
If you have a very fast disk device like a ramdisk, or an expensive RAID
controller with battery-backed up write cache, I would expect to see DRBD
impact performance some - even in 2.4 kernels.
-- Alan Robertson
alanr at unix.sh
More information about the Linux-HA