Storage Management TEG: Questionnaire Level 1 - Massimo Lamanna

This twiki is to collect the input of Massimo Lamanna. Please answer the questions below. For more information, please refer to the Storage TEG main twiki.

Question 1

- In your view, what are the 3 main current issues in Storage Management (SM)?

My answer:

• Weak error-resilience of storage services (to failures and to legitimate users overloads)

• Experiments framework not always capable to readapt automatically to SM malfunctioning
• Data location information often duplicated hence (Storage catalogues and "Central" experiment catalogues)

Question 2

- What is the greatest future challenge which would greatly impact the SM sector?

My answer:

• Define a clear, agreed and effective strategy out of a distributed data management system based on to SRM
• (Re)define clear boundaries between SM and DM (especially in the areaa of file catalogues)
  
• (Re)consider commonality especially on the DM layer

Question 3

- What is your site/experiment/middleware currently working on in SM?

My answer:

• Operating AFS (file protocol)
• Operating CASTOR (mainly rfio, xroot, gftp and srm)
• Operating EOS (mainly xroot, gftp and srm)

Question 4

- What are the big developments that you would like to see from your site/experiment/storage system in the next 5 years?

My answer:

I am not sure I understand the question. I think that our developments (on CASTOR and EOS) cover quite well the present horizon (2011-12 data taking). Assuming the 5-year period is after the 2013 shutdown I would assume that the experience of the next year plus the evolution on promising technologies (S3, NFSx.y, ...) will be the ingredient for new decisions.

I do not believe that in 2013 we will be starting (or at least we should start) a multi-year campaign (like the SRM definition/development/etc...): if in 2013 there will be a reasonable mix of techology to deliver the next generation storage servicies we will use it, otherwise we will continue to "tweek" the existing systems (this is valid both for the services provided by computer centres and their integration in the expts frameworks). With "reasonable mix" I mean, for example, a clear set of recipe to set up a Tier2 storage starting from a filesystem, or by buying S3 services from resource providers, etc...

A very interesting question is also to verify if the (global) filesystem approach is the way to go. Solutions ranging from a global filesystem (a` la AFS) to laptop downloading data (a` la Dropbox - no common filesystem whatsover) are on the table and we should find a convincing way to identify the best solution (cost-effective).

Question 5

- In your experience and area of competence, what are the (up to) 3 main successes in SM so far?

My answer:

Operating (on a wide range of sites) a coherent layer of storage services for LHC experiment activities. This includes long-term data custodial (tape), data access fro users (analysis) and data distribution.

Question 6

- In your experience and area of competence, what are the (up to) 3 main failures or things you would like to see changed in SM so far?

My answer:

• Service are provided with too much human effort (alternative formulation: storage services are all in all still too fragile)
• Lots of time lost in reinventing standards compared on developing robust and performant solutions

That's it!

Thanks! Feel free to edit again at any time, until the date of the kick-off meeting.

-- DanieleBonacorsi - November 2011