Computational Performance and Energy Efficiency of ARM based HPC servers
Oskar Schirmer

TL;DR
This paper evaluates the computational performance and energy efficiency of ARM-based HPC servers to explore their potential as alternatives to traditional x86 systems, aiming to diversify hardware options and promote co-design benefits.
Contribution
It provides an analysis of ARM-based HPC servers' performance and energy efficiency, highlighting their viability as alternative platforms and encouraging broader hardware diversity in HPC.
Findings
ARM servers show competitive performance levels.
Energy efficiency of ARM servers is promising.
Potential for hardware diversity in HPC environments.
Abstract
HPC world is dominated by x86 ISA CPUs. This monoculture is not necessarily justified by best performance evaluation, but may inherit from e.g. SW related restrictions on the choice of HW platforms. To avoid running (further) into path dependency, alternate HW platforms need to be evaluated for performance compared to existing HPC setup. As a result, it may turn out alternate HW platforms are more efficient for HPC. In any case, even if performance differences are low, avoiding path dependencies that stem from HW choice restrictions simplifies switching to different HW platforms in future, should suitable systems evolve. Moreover, broadening the perspective to generic HW platforms may trigger cooperation and wield influence on HW platform development, resulting in HW/SW co-design advantages.
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.
Taxonomy
TopicsCloud Computing and Resource Management · Distributed and Parallel Computing Systems · Parallel Computing and Optimization Techniques
