Alert.png The wiki is deprecated and due to be decommissioned by the end of September 2022.
The content is being migrated to other supports, new updates will be ignored and lost.
If needed you can get in touch with EGI SDIS team using operations @ egi.eu.

Difference between revisions of "DCH-RP:PoC 1 Belgium"

From EGIWiki
Jump to navigation Jump to search
Line 34: Line 34:
Combining Scenario 1 and 2 together the following PoC outline may be implemented:
Combining Scenario 1 and 2 together the following PoC outline may be implemented:


First, an all in-house (within Belspo) solution is deployed that to implement quality assurance on data archiving processes and technology. This solution is then integrated into the existing infrastructure at Belspo, and employed in a simulated day-today operating environment, with the help of the CH institutes in Belgium. This simulation then is monitored and described in a typical experiment way, and can serve as a blueprint for how to integrate new technology into an existing infrastructure - fulfilling scenario 2. By comparing this blueprint with other blueprints on the same topic, DCH-RP can develop best practices on how to manage ''technology insertion'' into its own e-Infrastructure. However, in preparation for examining Scenario 1, the second aspect of this study should include using Grid Storage (provided from VUB for Belspo) for the locally archived data. Local support should be provided by VUB for Grid storage access and tools. By accomplishing this, a second blueprint for technology insertion will be produced, though on a different aspect of the DCH e-Infrastructure.
KIK-IRPA, one of the Belgian DCH organisations, has already a local preservation system for their data. They have described their preservation system in a “Best practices” document that is accepted throughout the organisation. However one of their main concerns is to maintain the integrity of their data. Their exist auditing and certification schemes for trustworthy repositories, see: “Trustworthy Repositories – Audit and Certification” http://www.digitalrepositoryauditandcertification.org/pub/Main/ReferenceInputDocuments/trac.pdf and "Risk-analysis for E-depots:DRAMBORA" http://www.repositoryaudit.eu/. Such an audit also equals a risk analysis of the chosen archiving method. How esay this all may sound, real life shows that almost no one ever terminates the whole procedure, hence there is no common “best practices” available.


Once this scenario is accomplished, this then can be extended into looking at scenario 1 in a more broader scope than initially envisaged. This might be accomplished by VUB joining the existing EGI vo for DCH-RP, i.e. vo.dch-tp.eu. Through this, users registered in that VO can access the data stored at VUB for that VO. By providing data in the storage "bucket" provisioned at VUB for the DCH-RP VO, the data is made available for use in external tools employed by the individual researcher. The blueprint information coming out of this experiment will produce lessons learned on how to set up and administer Grid storage for DCH and share data in the community.
Doing an audit in a consequent way requests to use the necessary tools.
In this scenario we want to use existing tools and document the auditing process. We will do this on the local data that is in the KIK-IRPA preservation scheme and on data of other partners if possible.
Such an audit is in fact independent of where the data is stored but it is certainly a “tool” that will be very valuable for preservation done on data stored with e-infrastructures or other storage service providers.
 
Once done it would be useful to execute the procedure on preservation done on grid and cloud. This could be done in the PoC 2 projects as then experience with grid and cloud in the project will be available.
 
Suggested test data: KIK-IRPA will use their existing repository to do the audit


[[Category:Proofs Of Concept 1]]
[[Category:Proofs Of Concept 1]]

Revision as of 16:15, 2 May 2013

WP5: Proofs of Concept Scenarios PoC Phase 1 PoC Phase 2 DCH Glossary
Proof of Concept 1 | Belgium | Estonia | Hungary | Italy | Poland | Sweden


In Belgium, the first Proof of Concept will involve the following CH institutes:

  1. KIK
  2. KMKG
  3. KB
  4. RA

EGI Resources

EGI, as an e-Infrastructure provider, is supporting the DCH-RP project with resources as follows:

Grid storage Resources

Provider: Vrije Universiteit Brussel (VUB) - EGI site name "begrid-vub-ulb"
Storage: Grid SE storage, based on DPM, with SRM interface(?)

Cloud Storage

Cloud storage will be handled through the EGI Federated Clouds Task Force. The exact provider needs to be determined.

Proof of Concept scenarios

The Belgian partners would like to investigate Scenarios 1 and 2 (coming from DCH-RP Deliverable D3.1) as follows in PoC 1:

For Scenario 1, we should like to do the following: check the integrity/audit the data that is archived locally. In fact this equals a risk analysis of the chosen archiving method. We do not have to invent what needs to be done, all this is described in the document: “Trustworthy Repositories – Audit and Certification” http://www.digitalrepositoryauditandcertification.org/pub/Main/ReferenceInputDocuments/trac.pdf More information is also available at: "Risk-analysis for E-depots:DRAMBORA" http://www.repositoryaudit.eu/

Of course we want to do this with other partners.

Not to find in the current scenarios is a grid storage solution. I think this is mentioned in the project with the use of the eCSGW. In Belgium we are very interested to test the possibilities of such a storage solution. In the sense: how fast can data be retrieved from the grid storage, how easy is it to search for the data, also for the general public (in fact for all those that already use the existing local archive today).

Tentative architectural outline

Combining Scenario 1 and 2 together the following PoC outline may be implemented:

KIK-IRPA, one of the Belgian DCH organisations, has already a local preservation system for their data. They have described their preservation system in a “Best practices” document that is accepted throughout the organisation. However one of their main concerns is to maintain the integrity of their data. Their exist auditing and certification schemes for trustworthy repositories, see: “Trustworthy Repositories – Audit and Certification” http://www.digitalrepositoryauditandcertification.org/pub/Main/ReferenceInputDocuments/trac.pdf and "Risk-analysis for E-depots:DRAMBORA" http://www.repositoryaudit.eu/. Such an audit also equals a risk analysis of the chosen archiving method. How esay this all may sound, real life shows that almost no one ever terminates the whole procedure, hence there is no common “best practices” available.

Doing an audit in a consequent way requests to use the necessary tools. In this scenario we want to use existing tools and document the auditing process. We will do this on the local data that is in the KIK-IRPA preservation scheme and on data of other partners if possible. Such an audit is in fact independent of where the data is stored but it is certainly a “tool” that will be very valuable for preservation done on data stored with e-infrastructures or other storage service providers.

Once done it would be useful to execute the procedure on preservation done on grid and cloud. This could be done in the PoC 2 projects as then experience with grid and cloud in the project will be available.

Suggested test data: KIK-IRPA will use their existing repository to do the audit