LOI/KRD
LOI / KRD (Low-overhead Instrumentation and Kernel Reuse Distance) is a toolchain for low overhead kernel instrumentation and reuse distance analysis. The goal is to analyze parallel execution on large multicore NUMA machines. More information on LOI/KRD can be found in the following two publications
LOI / KRD (Low-overhead Instrumentation and Kernel Reuse Distance) is a toolchain for low overhead kernel instrumentation and reuse distance analysis. The goal is to analyze parallel execution on large multicore NUMA machines. More information on LOI/KRD can be found in the following two publications
- Scalable Analysis of Multicore Data Reuse and Sharing. Miquel Pericàs, Kenjiro Taura and Satoshi Matsuoka, The 2014 International Conference on Supercomputing (ICS’14), Munich, Germany, June 2014
- Analysis of Data Reuse in Task-Parallel Runtimes. Miquel Pericàs, Abdelhalim Amer, Kenjiro Taura and Satoshi Matsuoka. 4th International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS13), Denver, November 2013
The code can be accessed at: https://bitbucket.org/miquelp/loikrd/src/master
To clone (using git):
git clone git@bitbucket.org:miquelp/loikrd.git
I appreciate any feedback and comments on your experience using this code!
LOI/KRD is published under the terms of the New BSD License
Enjoy LOI/KRD!