On Using Linux Kernel Huge Pages with FLASH, an Astrophysical Simulation Code

TL;DR

This paper investigates the impact of using Linux kernel huge pages on FLASH, an astrophysical simulation code, finding reduced TLB misses but only marginal overall performance improvements on a specific hardware platform.

Contribution

It demonstrates the feasibility of enabling huge pages with FLASH on a particular compiler and hardware, highlighting the limited performance gains achieved.

Findings

Huge pages reduced translation lookaside buffer misses.

Overall performance gains were marginal.

Successful huge page usage was compiler-dependent.

Abstract

We present efforts at improving the performance of FLASH, a multi-scale, multi-physics simulation code principally for astrophysical applications, by using huge pages on Ookami, an HPE Apollo 80 A64FX platform. FLASH is written principally in modern Fortran and makes use of the PARAMESH library to manage a block-structured adaptive mesh. We explored options for enabling the use of huge pages with several compilers, but we were only able to successfully use huge pages when compiling with the Fujitsu compiler. The use of huge pages substantially reduced the number of translation lookaside buffer misses, but overall performance gains were marginal.