Page History

...

The recommended tools for exporting the data offsite are bbcp and Globus Online. The former, bbcp, is slightly simpler to setup. On the other hand Globus Online is more efficient when transferring large amount of data because it babysits the overall process by, for example, automatically restarting a failed or stalled transfer. The performance of the two tools is very similar.

...

1. Online monitoring nodes - These nodes get detector data over the network and place it in shared memory on each node.  The detector data are received over the network by snooping the multicast traffic between the DAQ readout nodes and the data cache layer. There is a set of monitoring nodes allocated for each instrument.
2. Fast feedback nodes - These nodes read the detector data from dedicated disks that each experimental hutch has. It is possible to read the data in quasi-real time as they are written to the fast feedback (FFB) storage layer by the DAQ without waiting for the run to end. There is a one farm of FFB nodes for each experimental hall.
3. Offline nodes - These nodes read the detector data from disk. These nodes include both interactive nodes (psana pool) and batch farms (psanaq queue, see below).  These nodes are the main resource for doing data analysis and are shared across all LCLS instruments.
   
   

...

The psana pool is currently made of 5 16 servers with the following specifications:

...

LCLS provides a few Matlab licenses and one IDL license. Instructions on how to use these tools are in the Matlab and IDL pages.

Two of the Each psana interactive nodes (psanagpu101 and 102) have GPUs. The other three (psanaphi101, 102 and 103) have Xeon Phi coprocessorshas a GPU accelerator.

Batch Farms

There are a number of batch farms (i.e. collections of compute nodes) located in the central computing building and in the NEH and FEH experimental halls. Instructions describing how to submit jobs can be found on the Submitting Batch Job page.

...