Re: why postgresql over other RDBMS

[Stefan Kaltenbrunner <stefan(at)kaltenbrunner(dot)cc>]

Tom Lane wrote:
> Stefan Kaltenbrunner <stefan(at)kaltenbrunner(dot)cc> writes:
>> Tom Lane wrote:
>>> A more interesting question is what sort of hardware you need for that
>>> actually to be a win, though.  Loading a few tables in parallel sounds
>>> like an ideal recipe for oversaturating your disk bandwidth...
> 
>> you don't actually need that much of disk bandwidth both COPY and CREATE
>> INDEX are CPU bottlenecked on modern boxes and reasonable disk
>> subsystems - spreading their work over multiple cores/processes can give
>> big benefits.
> 
> Hmm ... I wonder if that's true for COPY BINARY ...

not sure on that - I was simply trying to say that even a simple
parallel dump & restore capability could result in a serious improvement
for people running large databases(especially considering that one can
now buy 1U boxes with 8+ cores or 2U boxes that can hold 14disks+)  ;-)


Stefan