[nylug-talk] LOWMEM vs. HIGHMEM performance advantage?

[jh]

Ajai Khattri wrote:
> On Thu, 17 Apr 2008, jh wrote:
> 
>> Here's the current readout from free -m:
>>
>>               total       used       free     shared    buffers     cached
>> Mem:          2026       1975         50          0          0       1106
>> -/+ buffers/cache:        869       1156
>> Swap:         3812          0       3812
>>
>> So, swap isn't being touched at all. Plenty of free physical RAM, too.
> 
> What sort of application is it?

Custom app for document processing. Written in Python.

> Does it use files? Do they increase in
> number or size over time? 

It uses files in the sense that it is checking them into a file server. 
It first scans through the collection, and then dispenses them. Memory 
usage actually decreases over time.

> Does it use a database? Any sort of application 
> caching going on?
> 

It does use a database; no caching.

> Without knowing some detail about the app itself its hard to do any 
> theorizing.

The reason why I ask: using this app with smaller jobs uses less memory, 
and hence, I was wondering if the himem/lowmem split could explain 
anything. Otherwise, it is likely a peculiarity of the app itself, and 
the question is better directed to the developer.

jh

>