Hi
I have been speaking to Simone about T2D and the requirements are
that the site has good network connectivity to all Tier 1s and that
is has a certain level of support which is why the sites with the
ATLAS experts based there were put forward first. Obviously T2D are
still being set up but I am looking into getting RALPP considered as
a T2D (there shouldn't be network problems and I can hopefully
provide Chris Brew/Rob Harper with additional support). Simone
didn't seem to have an issue with the fact that RALPP isn't going to
be a very big ATLAS site because the point with T2D is that they can
be removed.
At the moment all that being a T2D gets you is some extra channels in
FTS however T2D will be the ones that can be mutli-cloud and there is
a proposal that if you want to have group disk at your site, you will
need to be a T2D too. Its clear that if you are a T2D you will get a
much larger chunk of ATLAS work sent your way.
PD2P is a bit of a distraction. The datadisk at all Tier 2s is now a
"cache" and the data you get depends on what users are requesting.
At the moment Tier 2s will get it all sent from RAL. It may well
soon be the case that T2D will receive it from other Tier 1s but I
don't think that is currently possible.
I don't see why we can't get significant number of T2Ds. I don't
think we should worry about not having a large number of T2D in the
first round but it is certainly worthwhile to keep doing the work
that has been going on (by many people) to improve the network issues.
Alastair
On 6 Apr 2011, at 16:41, John Gordon wrote:
> T2D = a T2 with sufficient network and storage bandwidth to be
> allowed to transfer directly to/from other T1 and T2D sites
>
> -----Original Message-----
> From: Testbed Support for GridPP member institutes [mailto:TB-
> [log in to unmask]] On Behalf Of Ewan MacMahon
> Sent: 06 April 2011 16:28
> To: [log in to unmask]
> Subject: Re: Sonar test summary
>
>> -----Original Message-----
>> From: Sam Skipsey [mailto:[log in to unmask]]
>> Sent: 06 April 2011 16:22
>>
>>> What exactly does being a T2D imply, and require? Does this just
>>> mean
>>> being used in the caching mode that Graeme was describing at Sussex?
>>>
>>
>> Essentially, IIRC... but that also has concomitant effects on the
>> amount
>> of work a site would expect (plus, it can only be good for a
>> region to
>> have T2Ds, since they may end up caching copies of important data
>> for the
>> cloud).
>>
>
> OK; I've just been having a look at the March GDB atlas presentation,
> and if I'm reading it correctly the T2D idea seems to be the
> associating
> a Tier 2 with multiple Tier 1s/clouds. based on Graeme's Sussex
> talk the
> 'tier 2 as cache' thing is PD2P.
>
> So; to unpick this a bit - am I right in thinking that some sites will
> be T2D sites (and if so, what exactly are the requirements?) and all
> Tier 2 space will be used according to the PD2P 'caching' mode, and
> none
> by planned data placement?
>
> Ewan
|