The worker nodes are for running production jobs (both batch and non-batch style). These nodes are "node1", "node2", etc. For non-batch use, connect to workers via ssh from hexfm1.rutgers.edu. For batch use, see below.
The farm is part of the "CAF" distributed computing facility of CDF. Hexcaf.rutgers.edu is a portal node and is the head node for the "rutgers" CAF. CAF jobs can be submitted from anywhere in CDF. The underlying batch facility for this CAF is fbsng and jobs can be submitted directly to fbsng. (This is useful for non-CDF users.)
Hexsam.rutgers.edu is the SAM station cdf-rutgers and is thus used for import/export of data. Hexfs1.rutgers.edu is an alias for hexsam.rutgers.edu.
Hexfm1 has 2GB of memory, all other nodes have 1GB of memory. Hexfm1, hexsam, hexcaf, and node2 through node7 have dual 1GHz Pentim III cpu's. Nodes 8 - 15 have dual 1.5 GHz AMD cpu's. Others have dual 2.0 GHz AMD cpu's.
Data disks total about 10 TB with 2/3 managed by the Disk Inventory Manager (DIM) or by SAM, and the rest in scratch disks (or static data disks) /scr*. Usage of disk space is monitored from these web pages. Most DIM or SAM disk is on two servers hexserver1.rutgers.edu and hexserver2.rutgers.edu. Scratch disk is often on some worker nodes as well as on the file servers. All disks are NFS mounted on all nodes, at the moment.
There are two tape drives on hexfm1:
