[DRBD-user] DRBD Transport Latency

H.D. devnull at deleted.on.request
Tue Jan 30 15:21:15 CET 2007

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.

On 30.01.2007 14:08, Lars Ellenberg wrote:
> also note that rtt in LAN is usually orders of magnitude
> less than average seek time/fsync time of storage.
> also in a dedicated fo for over 20km,
> rtt might be in the order of one digit ms, iirc.

OK, thanks for your reply. I just thought we might hit some limit here, 
as we fsync to big battery backed controller caches, so seek times is 
less an issue.

On our workload (a small active set of data), the benefit from using the 
controller cache seems extreme. We can push 3 times as many transactions 
per second than with disabled controller cache.


More information about the drbd-user mailing list