File: control

package info (click to toggle)
lam 6.1-11
  • links: PTS
  • area: main
  • in suites: slink
  • size: 7,764 kB
  • ctags: 5,834
  • sloc: ansic: 55,525; makefile: 1,449; fortran: 184; sh: 39; asm: 26; csh: 12
file content (19 lines) | stat: -rw-r--r-- 658 bytes parent folder | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Source: lam
Section: unknown
Priority: optional
Maintainer: Camm Maguire <camm@enhanced.com>
Standards-Version: 2.4.1

Package: lam
Architecture: any
Provides: mpi
Depends: ${shlibs:Depends}
Description: Enables parallel processing across multiple processors
 The major enhancements in LAM 6.1 are: 
 1) client-to-client MPI communication that utilizes shared memory
 in conjunction with TCP/IP
 2) an implementation of MPI-2 dynamic processes
 3) improved UNIX standard I/O capability from all nodes 
 LAM can also be much faster than MPICH when run in -c2c mode.
 Man pages for the standard MPI function calls are currently inlcuded
 in the mpich package.