Skip to content

ohoroyoi/SnuMAP

 
 

Repository files navigation

================================================================================
SnuMAP

SnuMAP is an open-source OS-level multi/many-core application profiler.
This framework contains the software package for SnuMAP, and provide
installation and SnuMAP testing guide.

SnuMAP is developed by Younghyun Cho, Camilo A. Celis Guzman, Heesik Shin
at Seoul National University as part of Computer Systems and Platforms
Laboratory (http://csap.snu.ac.kr) under the direction of Prof. Bernhard Egger.

Current website:
http://csap.snu.ac.kr/software/snumap

================================================================================
Access to a test machine

If you want to test SnuMAP on our platform, contact us.
For the testing, we can provide you the access to our one multi-core platform,
32-core (2-socket) AMD Opteron server.

Contact point: Younghyun Cho <yhcho614@gmail.com; younghyun@csap.snu.ac.kr>

================================================================================
Package Organization

[bin]
SnuMAP binaries and tools.

[examples]
Example open-source test applications, application implementations are not ours.
This directory also contains our experimental results on our target platforms.

[lib]
SnuMAP library files.

[src]
SnuMAP software source code.
We have several modules inside the directory.
1. kernel-module
2. kernel-patches
3. snumap-application interface
4. snumap-main (trace-profiler)
5. snumap-numa (numa-performance profiler)

================================================================================
Installation Guide

Please refer to the INSTALL file.

================================================================================
How to use SnuMAP?

Please refer to the USAGE file.

================================================================================

Last modification: 2017-01

About

Lightweight Trace Profiler for Multi/Manycore Systems

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 41.7%
  • C++ 28.7%
  • C 27.1%
  • Makefile 1.4%
  • Shell 1.1%