OPTIMA will present the paper “Feature-based SpMV Performance Analysis on Contemporary Devices” at the 37th IEEE International Parallel & Distributed Processing Symposium (IPDPS 23) in Parallel Technical Sessions 6B on Wednesday, 17 May, 2023.
Below is the abstract of the paper:
The SpMV kernel is characterized by high performance variation per input matrix and computing platform. While GPUs were considered State-of-the-Art for SpMV, with the emergence of advanced multicore CPUs and low-power FPGA accelerators, we need to revisit its performance and energy efIciency. This paper provides a high-level SpMV performance analysis based on structural features of matrices related to common bottlenecks of memory-bandwidth intensity, low ILP, load imbalance and memory latency overheads. Towards this, we create a wide articial matrix dataset that spans these features and study the performance of different storage formats in nine modern HPC platforms; five CPUs, three GPUs and an FPGA. After validating our proposed methodology using real-world matrices, we analyze our extensive experimental results and draw key insights on the competitiveness of different target architectures for SpMV and the impact of each feature/bottleneck on its performance.
More info: https://www.ipdps.org/ipdps2023/2023-advance-program.html#wednesday
See you in Florida!