Skip to content

hpcgroup/damselfly

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Damselfly v1.0

Damselfly is a model-based parallel network simulator. It can simulate communication patterns of High Performance Computing applications on different network topologies. It outputs steady-state network traffic for a communication pattern, which can help in studying network congestion and its impact on performance.

Limitation: can be run on up to 16,384 MPI processes. The primes.h file has a limited number of entries.

Build

mkdir build && cd build
cmake ..
make

Sample cmake command for Blue Gene/Q:

cmake -DCMAKE_C_COMPILER=mpixlc_r -DCMAKE_CXX_COMPILER=mpixlcxx_r -DCMAKE_BUILD_TYPE=Release -DJOB_TRAFFIC=ON ..

Run

mpirun -np 8 ./damselfly conffile map_file out_file

where:

  • conffile - configuration file (format described below)
  • map_file - if non-existent, default is used (Blue Gene style, i.e. coordinates in each line represents a rank; first "n1" ranks assumed for job 1, next "n2" ranks assumed for job 2, etc.)
  • out_file - generated, contains the bytes on links
Format of conffile
#total number of routers
#groups #rows #columns #nodes #cores
filename for intra group connection (binary format)
filename for inter group connection (binary format)
filename #ranks #edges (binary)
filename #ranks #edges (binary)
... as many files as needed

Code for generating sample connection files are in the utils directory.

Reference

Any published work that utilizes this software should include the following reference:

Nikhil Jain, Abhinav Bhatele, Xiang Ni, Nicholas J. Wright, and Laxmikant V.
Kale. Maximizing Throughput on a Dragonfly Network. In Proceedings of the
International Conference for High Performance Computing, Networking, Storage
and Analysis, SC '14, November 2014. LLNL-CONF-653557.

Release

Copyright (c) 2014, Lawrence Livermore National Security, LLC. Produced at the Lawrence Livermore National Laboratory.

Written by:

    Nikhil Jain <[email protected]>
    Abhinav Bhatele <[email protected]>
    Peer-Timo Bremer <[email protected]>

LLNL-CODE-678961. All rights reserved.

This file is part of Damselfly. For details, see: https://github.com/LLNL/damselfly. Please also read the LICENSE file for our notice and the LGPL.