Data preservation tasks

These tasks are part of the LS1 tasks list. If you interested in these tasks please contact Marco Cattaneo, Silvia Amerio, Pete Clarke, Tim and Patrick.

Data and software preservation

Lead (tools, or other) group Software/Computing/Physics
Other relevant groups (if any)  
Task name Data and software preservation
Task description Make sure that the LHCb data can still be analyzed years after the data had been recorded. This means defining the model to do this (i.e. is it that the latest version of analysis code can always read the oldest data - or do we freeze analysis code with data?). Then defining and implementing a suitable regression test engine, as well as defining the tests and reference results needed to determine compliance.
Estimated total effort required (FTE) 3
Deadline First milestone: not yet set
People/groups currently involved Several people are involved in high-level discussions re: data access and preservation
  Recently in addition CERN/Padova (Amerio ~ 0.5 FTE), Edinburgh (Cowan, Clarke ~0.2)
New effort required? Yes

More details:

This includes preservation of the data itself, and its access at the various computing centres, but moreover preservation of the HLT/reconstruction/analysis software. Main efforts for LHCb, some examples
  1. preservation of software: make sure that a suite of old data, HLT/reco/stripping/analysis software is routinely tested.
  2. documentation.
  3. preservation of data taking conditions: Would be nice to have for each period of data taking a human readable summary report. For example, a concise summary of all TCKs used together with a human readable comment. Also, any bugs/features of the Hlt/reconstruction software used.
  4. Re-evaluate data persistency model. For example, to extract the lifetime from a (m)DST, a rather sophisticated piece of software (DTF) is required. Making these kind of information persistent, would ease the issue of software preservation. Ideally, would like to have all "cut" parameters used in the stripping to be made persistent.
  5. What to do with MC ? Do we need to maintain all versions of Moore, in order to be able to simulate all used TCKs
Edit | Attach | Watch | Print version | History: r4 < r3 < r2 < r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r4 - 2013-11-06 - PatrickSKoppenburg
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    LHCb All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2022 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback