Re: Initial prefetch performance testing - Mailing list pgsql-hackers

From Ron Mayer
Subject Re: Initial prefetch performance testing
Date
Msg-id 48D7E0DB.4040109@cheapcomplexdevices.com
Whole thread Raw
In response to Re: Initial prefetch performance testing  (Gregory Stark <stark@enterprisedb.com>)
Responses Re: Initial prefetch performance testing  (Gregory Stark <stark@enterprisedb.com>)
Re: Initial prefetch performance testing  (Bruce Momjian <bruce@momjian.us>)
List pgsql-hackers
Gregory Stark wrote:
> Simon Riggs <simon@2ndQuadrant.com> writes:
>> I'm not in favour of introducing the concept of spindles....
> 
> In principle I quite strongly disagree with this....
> Number of blocks to prefetch is an internal implementation detail that the DBA
> has absolutely no way to know what the correct value is. 

Even more often on systems I see these days, "spindles"
is an implementation detail that the DBA has no way to know
what the correct value is.

For example, on our sites hosted with Amazon's compute cloud (a great
place to host web sites), I know nothing about spindles, but know
about Amazon Elastic Block Store[2]'s and Instance Store's[1].   I
have some specs and are able to run benchmarks on them; but couldn't
guess how many spindles my X% of the N-disk device that corresponds
to.  For another example, some of our salesguys with SSD drives
have 0 spindles on their demo machines.

I'd rather a parameter that expressed things more in terms of
measurable quantities -- perhaps seeks/second?  perhaps
random-access/sequential-access times?



[1] http://www.amazon.com/gp/browse.html?node=201590011
[2] http://www.amazon.com/b/ref=sc_fe_c_0_201590011_1?ie=UTF8&node=689343011&no=201590011&me=A36L942TSJ2AJA



pgsql-hackers by date:

Previous
From: Simon Riggs
Date:
Subject: Re: Initial prefetch performance testing
Next
From: Simon Riggs
Date:
Subject: Re: get_relation_stats_hook()