TraceR is a trace replay tool built upon the ROSS-based CODES simulation framework. TraceR can be used for predicting network performance and understanding network behavior by simulating messaging in High Performance Computing applications on interconnection networks.
TraceR depends on CODES and ROSS. There are several ways to build TraceR:
- Use spack to build TraceR and its dependencies:
spack install tracer
- Build TraceR and its dependencies manually:
- Download and install ROSS and CODES. Set the appropriate paths: ROSS_DIR, and CODES_DIR in tracer/Makefile.common.
- Pick between the two trace formats supported by TraceR: OTF2 or BigSim, and accordingly build the OTF2 or Charm++ library. If using OTF2 traces (default), set SELECT_TRACE = -DTRACER_OTF_TRACES=1, and ensure that otf2-config is in your PATH. If using BigSim traces, set SELECT_TRACE = -DTRACER_BIGSIM_TRACES=1, and set CHARMPATH to the Charm++ installation in tracer/Makefile.common.
- Set the ARCH variable in tracer/Makefile.common or alternatively set the CXX and ARCH_FLAGS variables. Then type:
cd tracer
make
More detailed build instructions are available on TraceR's documentation pages.
Refer to TraceR's User Guide for details on how to do network simulations using TraceR.
Any published work that utilizes TraceR should cite the following paper:
Nikhil Jain, Abhinav Bhatele, Samuel T. White, Todd Gamblin, and Laxmikant V. Kale. Evaluating HPC networks via simulation of parallel workloads. In Proceedings of the ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis, SC '16. IEEE Computer Society, November 2016. LLNL-CONF-690662.
TraceR is distributed under the terms of the MIT license.
Copyright (c) 2015, Lawrence Livermore National Security, LLC. Produced at the Lawrence Livermore National Laboratory.
Written by:
Nikhil Jain <nikhil.jain@acm.org>
Bilge Acun <acun2@illinois.edu>
Abhinav Bhatele <bhatele@llnl.gov>
LLNL-CODE-740483. All rights reserved.
SPDX-License-Identifier: MIT