[rds-devel] The meaning of MR invalidation

Or Gerlitz ogerlitz at voltaire.com
Tue Feb 19 04:38:11 PST 2008

Olaf Kirch wrote:
> On Monday 18 February 2008 08:17, Or Gerlitz wrote:

> Ouch, that's the rub indeed. One could argue that this is a bug in the FMR
> implementation/specification. The map count should *not* be reset when
> we call invalidate; it should be allowed to increment until we hit max_remaps,
> at which point the FMR needs to be destroyed.

If you don't want the driver to reset the map count, why were you 
calling fmr unmap in the first place, just to make SYNC_TPT being 
called? Also when the count hits max_remaps, what would destroying the 
FMR buys you? if you allocate new FMR, you might get the same rkey, no?

>> So how many IOs (=FMR mappings) per second are needed to serve this load?

> With 4K messages and 64K RDMAs, I hit 10,000 ops/second. With smaller messages
> even more.

I see, so for production you require RDS to be able to serve order ten-K 


More information about the rds-devel mailing list