[[asc:workbook| < /atlasfs/atlas/local/ You will see your name (likely). You can also have data directly deposited in a the same file system from a grid job by using the "--destSE=ANL-ATLAS-GRIDFTP1" for prun or pathena. It will appear in the directory "/atlasfs/atlas/dq2/user/" . The subdirectory structure is detemined by the output dataset name. You'll have to grope around a bit to find where the output went under this directory (some of the "."'s in the datset get transformed to "/"). To access files using root via xrootd on the worker nodes [this is true on ascinode but is it on 16 and 18?] you can use "xprep". Let's say you ran a grid job and your data is now available in the following area: /atlasfs/atlas/dq2/user//user..xxx/yyy.root You can prepare the files for use with root in the following way You run the following xprep command to register it (note that "/atlasfs" is not in the name): xprep -s aschead01 /atlas/dq2/user//user..xxx/yyy.root Instead of doing this file by file you may provide a list of files with the "-f" option. In root you can open the file as "root://aschead01///atlas/dq2/user//user..xxx/yyy.root". To check that the files have been registered you can do the following command: xrd aschead01 locatesingle /atlas/dq2/user//user..xxx/yyy.root