Re: Perfomance Tuning

Martha Stewart called it a Good Thing whengsw(at)globexplorer(dot)com ("Gregory S. Williamson")wrote:
> FWIW, Informix can be run using a "cooked" (Unix) file for storing
> data or it uses "raw" disk space and bypasses the ordinary (high
> level) UNIX controllers and does its own reads/writes. About 10
> times faster and safer. Of course, itmay have taken a lot of
> programmer time to make that solid. But the performance gains are
> significant.

Are you _certain_ that's still true? Have you a metric that shows
Informix being 10x faster on a modern system? That would be quite
surprising...

It may have been true on '80s style UFS implementations, but a couple
of decades have passed, and pretty much any Unix system has new
selections of filesystems that probably aren't so much slower.

It could conceivably be an interesting idea to implement a
block-oriented filesystem where the granularity of files was 8K (or
some such number :-)).

Oracle seems to have done something vaguely like this...
http://otn.oracle.com/tech/linux/open_source.html

But long and short is that the guys implementing OSes have been
putting a LOT of effort into making the potential performance gains of
using "raw" partitions less and less.
--
select 'cbbrowne' || '@' || 'acm.org';
http://www.ntlug.org/~cbbrowne/sap.html
(eq? 'truth 'beauty) ; to avoid unassigned-var error, since compiled code
; will pick up previous value to var set!-ed,
; the unassigned object.
-- from BBN-CL's cl-parser.scm