Meeting minutes - 13th September

Present: David Giaretta, Patricia Herterich, Karlheinz Schmitt, Heiko Tjalsma, Sharon McMeekin, Kirnn Kaur

The following tables are to be updated on the WP wiki pages :

  • Table of models
  • Analysis work carried out
ACTION: ALL

VCoE to provide suggestions as to which cost model is useful for a specific archive or sector

Cost data collection *Communication to be carried out in two stages

    • the first email will be sent to consortium partners and will include; an introduction to the WP, request for input to the cost data collection exercise, data confidentiality and anonymisation issues, data protection issues
    • the second email will be sent to those consortium partners who have expressed an interest/willingness to participate and will include details of cost parameters required, possible examples, overview of the cost models being analysed and any other information which will assist those contributing (e.g. guidance notes)
ACTION: KK TO DRAFT, ALL TO INPUT

DG stated that with the large repositories e.g. STFC and Helmholtz, we could expect several inputs (in terms of cost data) from the larger consortium partners.

For the model owners who are particpate in this WP i.e. DANS and DNB a request to re-use the cost data already used to test their models was made. HK and KS agreed to send over the test cost data after the data has been standardised. This data may be relevant in the testing of other models, although this exercise may require some further data manipulation around clear assumptions ACTION: HK & KS TO PROVIDE TEST DATA

DANS model is primarily for research data archives and is based on the OAIS reference model. This model could be adapted to become a universal model. HK to add these details to the table of models as an option we could review later. ACTION HK

Looking ahead to the final deliverable for this WP:

  • models predict costs in different ways e.g some focus on annual costs and others on specific activities e.g. storage, ingest
  • look at different scenarios:
    • Testing actual costs against the predictions made using historical data by the models (raised previously in relation to the LIFE 3 model), ESA and DANS could also be tested in this way.
    • Testing models using cost data already used to test models (manipulate the data for other models)
    • Test models using new data collected - need to liaise closely with model owners
  • expectations of the reviewer if we cannot test the models using the data collected - we will need to put questions to model owners and repository owners. also look at the various scenarios

Post meeting note - see which models can be tested in which ways - some are at different stages of development e.g. CMDP requires further testing using scientific repository data, some can be uesd to test historical predicted costs vs actual. ACTION - KK to raise at next meeting

-- KirnnKaur - 2012-09-13

Edit | Attach | Watch | Print version | History: r2 < r1 | Backlinks | Raw View | Raw edit | More topic actions
Topic revision: r2 - 2012-09-17 - KirnnKaur
 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback