2016 International Conference on Parallel Architecture and Compilation Techniques (PACT) (2016)
Sept. 11, 2016 to Sept. 15, 2016
Younghyun Cho , Department of Computer Science and Engineering, Seoul National University, Korea
Surim Oh , Department of Computer Science and Engineering, Seoul National University, Korea
Bernhard Egger , Department of Computer Science and Engineering, Seoul National University, Korea
We present an accurate online scalability prediction model for data-parallel programs on NUMA many-core systems. Memory contention is considered to be the major limiting factor of program scalability as data parallelism limits the amount of synchronization or data dependencies between parallel work units. Reflecting the architecture of NUMA systems, contention is modeled at the last-level caches of the compute nodes and the memory nodes using a two-level queuing model to estimate the mean service time of the individual memory nodes. Scalability predictions for individual or co-located parallel applications are based solely on data obtained during a short sampling period at runtime; this allows the presented model to be employed in a variety of scenarios. The proposed model has been implemented into an open-source OpenCL and the GNU OpenMP runtime and evaluated on a 64-core AMD system. For a wide variety of parallel workloads and configurations, the evaluations show that the model is able to predict the scalability of data-parallel kernels with high accuracy.
Computational modeling, Scalability, Data models, Runtime, Adaptation models, Predictive models, Kernel
Y. Cho, S. Oh and B. Egger, "Online scalability characterization of data-parallel programs on many cores," 2016 International Conference on Parallel Architecture and Compilation Techniques (PACT), Haifa, Israel, 2016, pp. 191-205.