InKS: a programming model to decouple algorithm from optimization in HPC codes

Abstract

Existing programming models tend to tightly interleave algorithm and optimization in HPC simulation codes. This requires scientists to become experts in both the simulated domain and the optimization process and makes the code difficult to maintain or port to new architectures. In this paper, we propose the InKS programming model that decouples these concerns with two distinct languages: InKS pia to express the simulation algorithm and InKS pso for optimizations. We define InKS pia and evaluate the feasibility of defining InKS pso with three test-languages: InKS o/C++ , InKS o/loop , InKS o/XMP . We evaluate the approach on synthetic benchmarks (NAS and heat equation) as well as on a more complex example (6D Vlasov-Poisson solver). Our evaluation demonstrates the soundness of the approach as it improves the separation of algorithmic and optimization concerns at no performance cost. We also identify a set of guidelines for the later full definition of the InKS pso language.

The NAS Parallel Benchmarks for evaluating C++ parallel programming frameworks on shared-memory architectures

“…The source code of NPB-CPP is provided within a repository made accessible to the community. 1 NPB-CPP is licensed under the terms of the MIT license and has recently been used by other independent research works as a baseline for experimental comparisons [14][15][16][17][18][19][20]. It highlights the importance of having a reliable C++ implementation of the distinguished NPB suite.…”

PAMPAR: A new parallel benchmark for performance and energy consumption evaluation

“…An example of this is the Threading Building Blocks (TBB), which is a C++ template library recently developed by Intel® for parallel programming on multicore processors . Another example is the InKS, which is a new programming model to decouple performance from algorithms in HPC codes.…”