[Halld-offline] computing model, alpha release

[David Lawrence]

Hi Graham.

     Certainly we should be able to put forward a set of requirements. 
I'd have to point out though that we are already developing expertise 
using certain GRID technologies between at least 4 of our universities 
(if I'm counting right). The thing that has me confused though is that 
this idea of it taking so much manpower to support it. I know of a 
certain collaborator in GlueX who, in addition to maintaining a full 
teaching schedule, maintaining his own computing farm, and heading a 
group responsible for one of our major detector systems in GlueX, has 
still managed to get his farm on the GRID and use it as well as other 
GRID resources for simulation studies. I can't imagine him spending more 
that 15-20% of his time on this. Why do we expect it to take a JLab IT 
person (with a CS degree) to have to spend so much more time? I must be 
missing something.

Regards,
-David

On 2/15/11 3:09 PM, Heyes Graham wrote:
> Physics division does maintain quite a few packages but it is done by 
> gentleman's agreement, i.e. nobody is explicitly paid to support ROOT 
> etc. The issue with GRID is that it is big and ugly enough that you 
> will have to pay out someone (probably more than one) to support GRID 
> at the JLab end if we are anything other than a client on someone 
> else's GRID implementation.
>
> Without putting words in his mouth I think that is what Chip was 
> talking about at the end of last week. We should write down what 
> resources we (Physics in 12 GeV era) expect to have available to us 
> offsite, what rates we need to those sites and other related 
> parameters. Again, implementation neutral so not explicitly GRID. If 
> it turns out that GRID is the best way to go then so be it. The 
> perennial argument is that it isn't worthwhile to the lab to pay 
> salaries and overhead for a couple of people to support something 
> unless it gives the lab access to resources that are at least greater 
> than you could buy with the same funds. So at least that argument 
> needs to be fleshed out. What exactly is on the non-JLab end of such a 
> "GIRD"?
>
> There'll probably be much discussion of this...
>
> Graham
>
> On Feb 15, 2011, at 2:53 pm, David Lawrence wrote:
>
>> Hi Mark,
>>
>>     This is a great start. I have one quick question: have we 
>> verified that the reconstruction rate for the 10^8 data is the same 
>> as the 10^7? Even though the event rate to disk will be the same, I 
>> can imagine the 10^8 data being busier and therefore, take longer to 
>> process. They are hopefully comparable, but we should verify it if it 
>> hasn't been already.
>>
>>     I'd also like to follow up a little on the discussion started a 
>> while back on the how the GRID is to be included in the computing 
>> plan. It seems clear that our collaborators feel this will be an 
>> important part of the GlueX experiment's operation. Chip has 
>> indicated though that supporting it from the JLab side has 
>> significant resource costs and must be clearly justified. Has there 
>> been any discussion as to what exactly these costs are? I thought 
>> from earlier conversations that this was a manpower cost on the JLab 
>> side, but at this point, I don't know what is required. The Physics 
>> Division maintains several software packages already (ROOT, CERNLIB, 
>> Xerces,...) are there additional resources besides software 
>> maintenance? I know the our original CDR called for something like an 
>> OC24 connection due to the anticipated high bandwidth usage. Are 
>> there other hardware costs involved?
>>
>>     We'll need to start gathering cost information as we develop the 
>> benefit argument in order to make a valid cost-benefit analysis.
>>
>> Regards,
>> -David
>>
>
> _______________________________________________
> Halld-offline mailing list
> Halld-offline at jlab.org
> https://mailman.jlab.org/mailman/listinfo/halld-offline
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/halld-offline/attachments/20110215/fa096fab/attachment-0002.html>