Re: Replication on the backend - Mailing list pgsql-hackers

From Luke Lonergan
Subject Re: Replication on the backend
Date
Msg-id 3E37B936B592014B978C4415F90D662D0C42B2@MI8NYCMAIL06.Mi8.com
Whole thread Raw
In response to Re: Replication on the backend  ("Mario Weilguni" <mario.weilguni@icomedias.com>)
List pgsql-hackers
Andrew,
> And if postgres could actually use an infiniband fabric for
> clustering a single database instance across Opteron servers, that
> would be very impressive...

That's what we do with Bizgres MPP.  We've implemented an interconnect to do the data shuffling underneath the
optimizer/executorand we currently use TCP/IP, though we could haul out SDP over Infiniband should we need it. 
However, our optimizer effectively minimizes traffic over the interconnect now and that works well for all of the plans
we'verun so far.  It would be nice to characterize the improvements we could get from moving to 3x infiniband. 
Regarding a direct Hypertransport to Infiniband bridge, have you looked at Pathscale? http://www.pathscale.com/  I know
thefellow behind the scenes who designed it, and I think it's probably well thought out.  We were gunning for less than
1usRTT through the adapter and switch once, and I bet they are close. 
- Luke



pgsql-hackers by date:

Previous
From: "J. Andrew Rogers"
Date:
Subject: Re: Replication on the backend
Next
From: Martijn van Oosterhout
Date:
Subject: Re: Feature Request: Multi-octet raw