Sharing information with the collaboration

Documents

• Use TDS -- See the TDS manual to create a TDS entry
• This wiki allows to attach documents and notes to a wiki page. This is not a good practise as the attached document is not version-controlled and can be changed by anyone. TDS is the preferred solution as it is more robust and sustainable.

Software and computer codes

• Use Git -- See the Git usage instructions to create a new project

Data and analysis results

• For small volume of data, Git is the best option
• For large volume of data or large number of files (> 1000) it is possible to publish them via the scientists.virgo-gw.eu web site
  • They must be located in a corresponding area in the storage space at /data/prod/web/ from the linux Cascina hosts. The directory name should reflect the name of the Group or the name of the Project in charge of those data.
  • Procedure to activate the area
    • Send an e-mail to EGO IT at service@ego-gw.it and Cc: the Group/Project Coordinator:
      • Brief description and purpose of the data
      • Name of the Group and/or of the Project that will manage the area and a suggestion on the directory name
      • Indicate the Linux group/users that need the write permissions
      • Explain whether the data are produced:
        • by a permanent "on-line" process (meaning attached to DAQ/hoft continuous production, via Cm or shared memory) with the aim to be used during science runs
        • by a permanent "in-time" process (meaning attached to DAQ/other_processes data via disk access) with the aim to be used during science runs
        • by an out-of-line process (meaning an interactive or batch process run by a user on a non-permanent schedule , or permanently but not needed by the science runs operation )
      • Explain whether the writing of the data is critical for the purpose in case of permanent writing (i.e. whether time "holes" can be tolerated or filled later starting from the input data)
      • Explain whether the availability of the data via web is critical for low-latency MMA, ITF operation, etc
      • Provide the estimated amount of data and and flux over a time period (GB and number of files per time period and duration of the production)
      • Provide the desired history: the data will cover a period of X months
      • Precise the desired retention period: the data must stay published for at least X months after first publication
    • Once EGO IT has made the area available you can fill with the data
  • Procedure to share data though scientists.virgo-gw.eu
    • Go to `/opt/w3/XXX` where XXX is the relevant topical area (for instance, XXX = DataAnalysis)
    • Create a symbolic link pointing to /data/prod/web/.
    • Your data will be accessible via the URL https://scientists.virgo-gw.eu/XXX/.
  • Important notes
    • These storage areas are directly connected to the internal/ITF domain networks. Their access via web from outside Cascina is subject to the pre-authentication via the firewall with the usual EGO Active Directory account as described at https://landing.ego-gw.it
    • The area /data/prod/web is meant for web content or images should be stored. Actual data should go to /data/prod/_project_name_ areas, especially if belonging to a more complex on-line data analysis workflow. This applies specifically when retention time of the web data is longer than the generation time.