[DRBD-user] Performance with DRBD + iSCSI

Eduardo Grosclaude eduardo.grosclaude at gmail.com
Thu Feb 22 11:44:21 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 2/21/07, Lars Ellenberg <Lars.Ellenberg at linbit.com> wrote:
> / 2007-02-21 17:31:22 -0300
> \ Eduardo Grosclaude:
> > >> > Remember that for DRBD Prot C, synchronous io, that io operations
> > >> > will not return (thus the next operation will not execute),
> > >> > until the write operation has been committed to disk on both
> > >> > sides which means that performance will be network bound.
>
> > Is/will be TIPC networking considered in DRBD's roadmap somehow?
>
> and how exactly would TIPC help DRBD performance?
Er... that was my next question :)
Wouldn't single roundtrips benefit from a lower latency under the C
protocol? Even when your bottleneck will (always?) be your disk I/O
for operation in-the-large, maybe some races against failure events at
the primary could be won by shortening disk-to-disk roundtrips.
-- 
Eduardo Grosclaude
Universidad Nacional del Comahue
Neuquen, Argentina




More information about the drbd-user mailing list