Phoenix - A Novel Technique for Performance-Aware Orchestration of Thread and Page Table Placement in NUMA Systems

Mohammad Siavashi, Alireza Sanaee, Mohsen Sharifi, Gianni Antichi · 2025

Read Phoenix - A Novel Technique for Performance-Aware Orchestration of Thread and Page Table Placement in NUMA Systems by Mohammad Siavashi, Alireza Sanaee, Mohsen Sharifi, Gianni Antichi (2025) 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-2502-10923.txt.

Description

Mohammad Siavashi, Gianni Antichi, and co-authors present Phoenix, a system tool that joins thread and memory placement on large shared machines, in a 2025 paper. On such machines, memory far from a core is slower to reach. Past tools set threads and page tables apart, which left much speed on the table.

Phoenix links the scheduler and the memory manager in one Linux module. It copies page tables on demand and guards shared bandwidth to keep service fair. On real hardware, it cut core cycles by more than half versus top past tools. Page walk cycles also fell by a wide margin.

Catalog record

CREATORMohammad Siavashi, Alireza Sanaee, Mohsen Sharifi, Gianni Antichi
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE56 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-10923

Browse the full library · About the Tower of Babel