Towards an Adaptive Runtime System for Cloud-Native HPC
Aditya Bhosale, Advait Tahilyani, Laxmikant Kale, Sara Kokkila-Schumacher · 2026
Read Towards an Adaptive Runtime System for Cloud-Native HPC by Aditya Bhosale, Advait Tahilyani, Laxmikant Kale, Sara Kokkila-Schumacher (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-14630.txt.
Description
Bhosale and co-authors adapt the Charm++ runtime for cloud-native high performance computing. HPC codes built for static supercomputers fit cloud settings poorly. Clouds shift underfoot, with mixed hardware and spot pricing. Charm++ uses asynchronous, message-driven tasks that suit that flux. Rate-aware load balancing lifts speed on mixed CPU and GPU fleets. Core Charm++ rules blunt harm from network contention and speed variance. Extra tools push resilience further for coupled science codes, the authors report.
Catalog record
| CREATOR | Aditya Bhosale, Advait Tahilyani, Laxmikant Kale, Sara Kokkila-Schumacher |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33006 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.14630