On Wed, Mar 09, 2005 at 05:51:43PM -0800, Chris Travers wrote:
> Jim C. Nasby wrote:
> Ok--- I will admit to a anti-Windows bias. But at least my bias is
> informed. In addition to my former employment at Microsoft, I have
> studies both types of OS's in detail. Here are some specific comments I
> would make:
>
> 1) I do not expect PostgreSQL to *ever* perform as well on Windows as
> it does on Linux. This is primarily due to the fundamentally different
> emphasis in kernel architecture between UNIX-style and VMS-style
> operating systems. Windows server applications which are process-based
> are always likely to underperform. Windows applications ported to Linux
> are similarly likely to underperform.
This is akin to saying that an application written to use MySQL will
never perform well on PostgreSQL. It depends on *how* the code is
written. If your SQL is tuned to one database, it will likely have
performance issues on other databases. Likewise, a process-based server
will perform poorly on Windows, while a threaded server will not. This
is an implimentation choice. There's no reason why PostgreSQL on windows
*has* to be process based (though of course there would be serious
technical issues with changing it).
Of course, by simply hand waving and saying "it can never be better", it
never will be better.
--
Jim C. Nasby, Database Consultant decibel@xxxxxxxxxxx
Give your computer some brain candy! www.distributed.net Team #1828
Windows: "Where do you want to go today?"
Linux: "Where do you want to go tomorrow?"
FreeBSD: "Are you guys coming, or what?"
---------------------------(end of broadcast)---------------------------
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match
