Termination of RAL ADS/VTP Tape Service

Last modified: Fri Sep 12 05:53:47 BST 2008
This page is now obsolete. In the end we decide to copy all the data to CASTOR where it can be found under:- 
  /castor/ads.rl.ac.uk/prod/minos/tape/archive/ADS/soudan2/

!!!! WARNING !!!!

After ~ December 2008 - There will be no Soudan2 tape data at RAL

If this makes you unhappy please mail Nick West (n.west1@physics.ox.ac.uk) ASAP.

Funding by GRIDPP of the old RAL tape datastore will cease in March 2008. Soudan2 have about 2TBs worth of data there. This data all belongs jointly to the Soudan2 and MINOS collaborations.

An overview of the Soudan2 data can be found at:-

http://hepunx.rl.ac.uk/soudan2/doc/data/

with the usual numi password.

Here is list of the data we have on tape that I have managed to find:-

No. of files Total size (GB) Data Set
2217 1935 Raw data PDKDATA
2585 999 Processed PDKDST
967 7.0 ISIS Test Beam Data PDKISIS

We don't necessarily loose the data after that, but we do have to pay for it if we want to keep it. I don't know how much that will be but my guess is that if we ask to have the tapes removed from the robot and stored that the cost won't be high. I have requested an annual cost for this.

However, unless there is a realistic chance that we could recover the data there's no point in paying, no matter how cheap it is. Recovering the data requires locating all the old support libraries, resurrecting the old offline software and recreating the database. The code ran on VAX/VMS and I did do some work on porting it to Linux but cannot now recall the state of that endevour or if/where its fruit now are. Then there is the complication that machines on site at RAL will probably stop supporting the interface to the obsolescent data store and to even get the data back off tape will involve installing that software elsewhere. In short we are talking about at least a couple of months of effort of my time if I had to start today. That estimate will only grow with time. It is time I don't have right now, nor will I as I start working on T2K later this year.

So my own opinion is that we let this data go, but it's not my decision to take. Do any other sites still have this data that could recovered if we did ever need it? Recall that the primary data store was never RAL but Soudan. Does anyone know what state that is in? Perhaps interested parties could briefly discuss this at the Week in the Woods?