TY - GEN
T1 - The AmpereOne A192-32X in Perspective
T2 - 2025 International Conference on High Performance Computing in the Asia-Pacific Region, HPC Asia 2025
AU - Carlson, David
AU - Simakov, Nikolay
AU - Hadlich, Rodrigo Ristow
AU - Curtis, Anthony
AU - Martin, Joshua
AU - Verma, Gaurav
AU - Chheda, Smeet
AU - Coskun, Firat
AU - Gonzalez, Raul
AU - Wood, Daniel
AU - Zhang, Feng
AU - Harrison, Robert
AU - Siegmann, Eva
N1 - Publisher Copyright:
© 2025 Copyright held by the owner/author(s).
PY - 2025/4/19
Y1 - 2025/4/19
N2 - This study presents a comprehensive benchmarking analysis of the Arm-based AmpereOne A192-32X CPU, a high-performance but low power processor designed for cloud-native workloads characterized by high core occupancy, imperfectly-vectorized or even pure scalar software, limited need for high floating-point performance, and, increasingly, AI inference. These traits also characterize much of academic research computing. Hence a thorough investigation of this novel CPU seeking to characterize its strengths and weaknesses on academic workloads, including traditional HPC codes for which it was not designed, will shed light on its relevance in a research setting. We report comparative analyses with contemporary CPUs (Intel Sapphire Rapids, AMD EPYC, NVIDIA Grace-Grace) and illustrate AmpereOne’s architectural advantages in handling parallel workloads and optimizing power consumption. The CPUs are compared in terms of performance and power consumption using a wide range of applications covering different workloads and disciplines.
AB - This study presents a comprehensive benchmarking analysis of the Arm-based AmpereOne A192-32X CPU, a high-performance but low power processor designed for cloud-native workloads characterized by high core occupancy, imperfectly-vectorized or even pure scalar software, limited need for high floating-point performance, and, increasingly, AI inference. These traits also characterize much of academic research computing. Hence a thorough investigation of this novel CPU seeking to characterize its strengths and weaknesses on academic workloads, including traditional HPC codes for which it was not designed, will shed light on its relevance in a research setting. We report comparative analyses with contemporary CPUs (Intel Sapphire Rapids, AMD EPYC, NVIDIA Grace-Grace) and illustrate AmpereOne’s architectural advantages in handling parallel workloads and optimizing power consumption. The CPUs are compared in terms of performance and power consumption using a wide range of applications covering different workloads and disciplines.
KW - Arm
KW - Benchmarking
KW - HPC
UR - https://www.scopus.com/pages/publications/105007283116
U2 - 10.1145/3703001.3724384
DO - 10.1145/3703001.3724384
M3 - Conference contribution
AN - SCOPUS:105007283116
T3 - Proceedings of International Conference on High Performance Computing in Asia-Pacific Region Workshops, HPC Asia 2025 Workshops
SP - 23
EP - 35
BT - Proceedings of International Conference on High Performance Computing in Asia-Pacific Region Workshops, HPC Asia 2025 Workshops
PB - Association for Computing Machinery, Inc
Y2 - 19 February 2025 through 21 February 2025
ER -