LAM (Local Area Multicomputer) is an Message-Passing Interface (MPI)
programming environment and development system for heterogeneous
computers on a network. With LAM/MPI, a dedicated cluster or an
existing network computing infrastructure can act as one parallel
computer to solve one problem. LAM/MPI is considered to be "cluster
friendly" because it offers daemon-based process startup/control as
well as fast client-to-client message passing protocols. LAM/MPI can
use TCP/IP and/or shared memory for message passing (different RPMs
are supplied for this -- see the main LAM website at
http://www.mpi.nd.edu/lam/ for details).<
LAM features a full implementation of MPI version 1 (with the
exception that LAM does not support cancelling of sends), and much of
version 2. Compliant applications are source code portable between LAM
and any other implementation of MPI. In addition to meeting the
standard, LAM/MPI offers extensive monitoring capabilities to support
debugging. Monitoring happens on two levels: On one level, LAM/MPI has
the hooks to allow a snapshot of a process and message status to be
taken at any time during an application run. The status includes all
aspects of synchronization plus datatype map/signature, communicator
group membership and message contents (see the XMPI application on the
main LAM website). On the second level, the MPI library can produce a
cumulative record of communication, which can be visualized either at
runtime or post-mortem.