Re: [Condor-users] Request_memory and dynamic slots

[Ian Cottam <Ian.Cottam@xxxxxxxxxxxxxxxx>]

Yes, of course, I was confusing CPUs and Memory somewhat (have edited post
below).
Question still stands though :-) -Ian
On 23/05/2012 09:38, "Mark Calleja" <mc321@xxxxxxxxx> wrote:

>Hi Ian,
>
>Actually, request_memory defaults to the ImageSize or JobVMemory (in
>MB). It's request_cpus that defaults to 1. See:
>
>http://docs.redhat.com/docs/en-US/Red_Hat_Enterprise_MRG/1.3/html/Grid_Use
>r_Guide/chap-Grid_User_Guide-Dynamic_slots.html
>
>Best,
>Mark
>
>On 23/05/2012 09:08, Ian Cottam wrote:
>> Scenario: we have a Linux/64 pool with all clients using dynamic slots.
>> We have 7.4/7.6 (but if the answer to the question below is different
>>for
>> 7.8 I would be interested in that too).
>>
>> We are about to tell our users not to use (Memory>=n) in Requirements,
>>but
>> use
>> Request_memory = n
>> as this appears to work much better with dynamic slots.
>>
>> Say a given job will grow to use 3000MB of memory; what is the
>>difference
>> between the following cases:
>>
>> a) User does not supply a Request_memory line.
>> I believe the default is to ImageSize or JobVMemory (how is that
>>computed?)
>>
>> b) User supplies:
>> Request_memory = 1
>> Is this a special case in any sense?
>>
>> c) User supplies:
>> Request_memory = 1000
>>
>> d) User supplies:
>> Request_memory = 3000
>>
>>
>> Thanks v much!
>> -Ian
>
>-- 
>Mark Calleja - Scientific Computing Group
>University of Cambridge Computing Service
>New Museums Site, Pembroke Street
>Cambridge CB2 3QH, UK
>Tel (+44/0) 1223 761254
>
>