GAMER with out-of-core computation
Hsi-Yu Schive, Yu-Chih Tsai, and Tzihong Chiueh

TL;DR
This paper enhances the GAMER astrophysical simulation code by integrating advanced hydrodynamic methods and out-of-core computation, enabling larger simulations with improved accuracy and performance on GPU clusters.
Contribution
The paper introduces out-of-core computation to GAMER, allowing larger simulations, and implements the MUSCL-Hancock method with Roe's Riemann solver for better accuracy and speed.
Findings
Improved simulation accuracy with MUSCL-Hancock and Roe's solver.
Enhanced GPU-CPU speed-up performance.
Enabled large-scale simulations using out-of-core computation.
Abstract
GAMER is a GPU-accelerated Adaptive-MEsh-Refinement code for astrophysical simulations. In this work, two further extensions of the code are reported. First, we have implemented the MUSCL-Hancock method with the Roe's Riemann solver for the hydrodynamic evolution, by which the accuracy, overall performance and the GPU versus CPU speed-up factor are improved. Second, we have implemented the out-of-core computation, which utilizes the large storage space of multiple hard disks as the additional run-time virtual memory and permits an extremely large problem to be solved in a relatively small-size GPU cluster. The communication overhead associated with the data transfer between the parallel hard disks and the main memory is carefully reduced by overlapping it with the CPU/GPU computations.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
