Evaluation of ARM CPUs for IceCube available through Google Kubernetes Engine

TL;DR

This study evaluates ARM CPUs for IceCube's simulation workloads on Google Kubernetes Engine, finding they are more cost-effective and slightly faster than traditional x86 CPUs, supporting ARM adoption.

Contribution

First comprehensive comparison of ARM and x86 CPUs for IceCube simulations in cloud environments, demonstrating ARM's advantages in cost and performance.

Findings

ARM CPUs are about 20% more cost-effective.

ARM CPUs are less than 10% slower in absolute performance.

Building ARM binaries from source was straightforward.

Abstract

The IceCube experiment has substantial simulation needs and is in continuous search for the most cost-effective ways to satisfy them. The most CPU-intensive part relies on CORSIKA, a cosmic ray air shower simulation. Historically, IceCube relied exclusively on x86-based CPUs, like Intel Xeon and AMD EPYC, but recently server-class ARM-based CPUs are also becoming available, both on-prem and in the cloud. In this paper we present our experience in running a sample CORSIKA simulation on both ARM and x86 CPUs available through Google Kubernetes Engine (GKE). We used the production binaries for the x86 instances, but had to build the binaries for ARM instances from source code, which turned out to be mostly painless. Our benchmarks show that ARM-based CPUs in GKE were not only the most cost-effective but were also the fastest in absolute terms in all the tested configurations. While the advantage is not drastic, about 20% in cost-effectiveness and less than 10% in absolute terms, it is still large enough to warrant an investment in ARM support for IceCube.