Scheduler-Activated Dynamic Page Migration for Multiprogrammed DSM Multiprocessors

The performance of multiprogrammed shared-memory multiprocessors suffers often from scheduler interventions that neglect data locality. On cache-coherent distributed shared-memory (DSM) multiprocessors, such scheduler interventions tend to increase the rate of remote memory accesses. This paper pres...

Full description

Saved in:
Bibliographic Details
Published in:Journal of parallel and distributed computing Vol. 62; no. 6; pp. 1069 - 1103
Main Authors: Nikolopoulos, Dimitrios S., Polychronopoulos, Constantine D., Papatheodorou, Theodore S., Labarta, Jesús, Ayguadé, Eduard
Format: Journal Article Publication
Language:English
Published: San Diego, CA Elsevier Inc 01.06.2002
Elsevier
Subjects:
Applied sciences
Arquitectura de computadors
Arquitectures paral·leles
Computer science; control theory; systems
Data locality
Distributed shared-memory multiprocessors
Exact sciences and technology
Informàtica
Memory and file management (including protection and security)
Memory management
Memory organisation. Data processing
Multiprocessadors
Multiprocessors
Multiprogramming
NUMA
Operating systems
Page migration
Parallel programming (Computer science)
Programació en paral·lel (Informàtica)
Runtime systems
Software
Telecommunications
Telecommunications and information theory
Teleprocessing networks. Isdn
Àrees temàtiques de la UPC
memory management
data locality
distributed shared-memory multiprocessors
multiprogramming
operating systems
page migration
runtime systems
NUMA
Performance evaluation
Cache memory
Scheduling
Iterative program
Storage access
Competitive algorithms
Remote access
Multiprocessor
Distributed memory
Distributed memory multiprocessor system
Storage management
Virtual memory
Parallel program
Multiprogramming
Shared memory
Operating system
Biased estimation
ISSN:0743-7315, 1096-0848
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The performance of multiprogrammed shared-memory multiprocessors suffers often from scheduler interventions that neglect data locality. On cache-coherent distributed shared-memory (DSM) multiprocessors, such scheduler interventions tend to increase the rate of remote memory accesses. This paper presents a novel dynamic page migration algorithm that remedies this problem in iterative parallel programs. The purpose of the algorithm is the early detection of pages that will most likely be accessed remotely by threads associated with them via a thread-to-memory affinity relation. The key mechanism that enables timely identification of these pages is a communication interface between the page migration engine and the operating system scheduler. The algorithm improves previously proposed competitive page migration algorithms in many aspects, including accuracy, timeliness and cost amortization. Most notably, the algorithm is not biased by obsolete memory access history that may be accumulated in the page access counters at runtime. Experiments on the SGI Origin2000 show that the algorithm outperforms by far the best static page placement algorithm and a customized page migration engine implemented in IRIX, the Origin2000 operating system. The algorithm is implemented at user-level and its functionality is orthogonal to the scheduling policy of the operating system.
ISSN:0743-7315
1096-0848
DOI:10.1006/jpdc.2001.1817