Re: Dell PERC H700/H800 - Mailing list pgsql-performance

From Greg Smith
Subject Re: Dell PERC H700/H800
Date
Msg-id 4B764F45.3040503@2ndquadrant.com
Whole thread Raw
In response to Re: Dell PERC H700/H800  (Dave Crooke <dcrooke@gmail.com>)
Responses Re: Dell PERC H700/H800  (Scott Marlowe <scott.marlowe@gmail.com>)
Re: Dell PERC H700/H800  (Bruce Momjian <bruce@momjian.us>)
List pgsql-performance
I've been full-on vocally anti-Dell ever since they started releasing
PCs with the non-standard ATX power supply pinout; that was my final
straw with their terrible quality decisions.  But after doing two tuning
exercises with PERC6 controllers and getting quite good results this
year, just a few weeks ago I begrudgingly added them to my "known good
hardware" list as a viable candidate to suggest to people.  They finally
took a good LSI card and didn't screw anything up in their version.

I am somehow relieved that sanity has returned to my view of the world
now, with Dell right back onto the shit list again.  If they want a HCL
and to warn people they're in an unsupported configuration when they
violate it, which happens on some of their equipment, fine.  This move
is just going to kill sales of their servers into the low-end of the
market, which relied heavily on buying the base system from them and
then dropping their own drives in rather than pay the full "enterprise
drive" markup for non-critical systems.

I do not as a rule ever do business with a vendor who tries to lock me
into being their sole supplier, particularly for consumable replacement
parts--certainly a category hard drives fall into.  Probably the best
place to complain and suggest others do the same at is
http://www.ideastorm.com/ideaView?id=087700000000dwTAAQ

--
Greg Smith    2ndQuadrant   Baltimore, MD
PostgreSQL Training, Services and Support
greg@2ndQuadrant.com  www.2ndQuadrant.com


pgsql-performance by date:

Previous
From: Greg Smith
Date:
Subject: Re: moving pg_xlog -- yeah, it's worth it!
Next
From: Bryce Nesbitt
Date:
Subject: Re: 512,600ms query becomes 7500ms... but why? Postgres 8.3 query planner quirk?