Can I read Traffic management: a holistic approach to memory placement on NUMA systems on EtoBox?

Traffic management: a holistic approach to memory placement on NUMA systems by Mohammad Dashti; Alexandra Fedorova; Justin Funston; Fabien Gaud; Renaud Lachaize; Baptiste Lepers; Vivien Quema; Mark Roth is a Computer Science article available to read on EtoBox.

What is Traffic management: a holistic approach to memory placement on NUMA systems about?

NUMA systems are characterized by Non-Uniform Memory Access times, where accessing data in a remote node takes longer than a local access. NUMA hardware has been built since the late 80's, and the operating systems designed for it were optimized for access locality. They co-located memory pages with the threads that accessed them, so as to avoid the cost of remote accesses. Contrary to older systems, modern NUMA hardware has much smaller remote wire delays, and so remote access costs per se are not the main concern for performance, as we discovered in this work. Instead, __congestion on memory controllers and interconnects__ , caused by memory traffic from data-intensive applications, hurts performance a lot more. Because of that, memory placement algorithms must be redesigned to target traffic congestion. This requires an arsenal of techniques that go beyond optimizing locality. In this paper we describe __Carrefour__ , an algorithm that addresses this goal. We implemented __Carrefour__ in Linux and obtained performance improvements of up to 3.6 relative to the default kernel, as well as significant improvements compared to NUMA-aware patchsets available for Linux. __Carrefour__ n

Who reads Traffic management: a holistic approach to memory placement on NUMA systems?

Author

Mohammad Dashti; Alexandra Fedorova; Justin Funston; Fabien Gaud; Renaud Lachaize; Baptiste Lepers; Vivien Quema; Mark Roth

Publisher

ACM

Published

2013

Language

EN

Field

Computer Science (Physical Sciences)