An Approximate Computing Stack Based on Computation Reuse

Sato, Yuuki; Tsumura, Takanori; Tsumura, Tomoaki; Nakashima, Yasuhiko

doi:10.1109/candar.2015.35

Cited by 7 publications

(1 citation statement)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hara and Hanawa utilized field programming gate array (FPGA) flexibility to offload some tasks in an HPC application [7]. In addition, certain research works memorize the input/output of repeated computations and reuse them while allowing some accuracy degradation [11,12]. As a result, AC can be applied to various applications; however, we still need to consider the characteristics of each thread/rank in a parallel application.…”

Section: Related Workmentioning

confidence: 99%

Data Transfer API and its Performance Model for Rank-Level Approximate Computing on HPC Systems

Morie

Wada

Kobayashi

et al. 2023

IJNC

View full text Add to dashboard Cite

The application of approximate computing (AC) in optimizing tradeoffs among performance, power consumption, and accuracy of computation results can be improved by adjusting data precision in applications. The importance of AC has increased over the years as it is used to maximize performance even with limited power budget and hardware resources in high performance computing (HPC) systems that require more precise computations. To apply AC for HPC applications effectively, we must consider the character of each message passing interface (MPI) rank in an application and optimize it by adjusting its data precision. This rank-level AC ensures that ranks and threads in an application run with data precision and perform data transfer while converting the precision of target data. In this paper, we have proposed and evaluated data pack/unpack application programming interfaces (APIs), which are applicable for standard MPI programs run on HPC systems, for converting the precision of target data. The proposed APIs enable us to express data transfer among ranks with different precisions. In addition, we have also developed a reasonable performance model to select an appropriate data transfer API for maximizing performance with rank-level AC based on performance evaluation with various HPC systems.

show abstract

Section: Related Workmentioning

confidence: 99%