Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Recurso digital |
| Sprache: | |
| Veröffentlicht: |
Zenodo
2025
|
| Online-Zugang: | https://doi.org/10.5281/zenodo.17971086 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Inhaltsangabe:
- <p>This paper proposes a next-generation computational resource sharing system that enables a large number of users to simultaneously utilize a small number of high-performance computing devices.</p> <p>Its key feature is the integrated control of second-level dynamic reallocation of computational resources such as CPU, GPU, and memory, combined with AI-based demand prediction, automatic switching of processing models, and live migration.</p> <p>By overcoming the performance degradation and high costs associated with conventional VDI and DaaS, the paper demonstrates the feasibility of a new PC infrastructure in which many users can perform high-load processing on a single machine.</p>