...
Each Fermi data run corresponds to a single top-level stream of this task.
The task consists of seven job steps. One top-level stream reprocesses one Fermi run. Because each Fermi run may contain millions of events, the reprocessing is spread across multiple CPUs by breaking the run into fixed-length 'clumps'. Each clump is reprocessed independently; the primary output (typically ROOT files) is finally merged. Secondary data products (e.g., FITS files) are created after the merging of ROOT files.
The task structure is fairly simple, with various short bookkeeping steps, and two 'heavy lifters': processClump.py runs Gleam; 'mergeClumps.py' merges run fragments and produces any final data products. Depending on the task configuration, these two job steps may be substantial (xlong queue) or trivial (short queue).
...