Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network

  Advanced Search  

Fast and Scalable Mutual Exclusion ? James H.Anderson and YongJik Kim

Summary: Fast and Scalable Mutual Exclusion ?
James H.Anderson and Yong­Jik Kim
Department of Computer Science
University of North Carolina at Chapel Hill
Abstract. We present an N­process algorithm for mutual exclusion un­
der read/write atomicity that has O(1) time complexity in the absence
of contention and \Theta(log N) time complexity under contention, where
``time'' is measured by counting remote memory references. This is the
first such algorithm to achieve these time complexity bounds. Our algo­
rithm is obtained by combining a new ``fast­path'' mechanism with an
arbitration­tree algorithm presented previously by Yang and Anderson.
1 Introduction
Recent work on mutual exclusion [3] has focused on the design of ``scalable'' algo­
rithms that minimize the impact of the processor­to­memory bottleneck through
the use of local spinning . A mutual exclusion algorithm is scalable if its perfor­
mance degrades only slightly as the number of contending processes increases.
In local­spin mutual exclusion algorithms, good scalability is achieved by requir­
ing all busy­waiting loops to be read­only loops in which only locally­accessible
shared variables are accessed that do not require a traversal of the processor­
to­memory interconnect. A shared variable is locally accessible on a distributed


Source: Anderson, James - Department of Computer Science, University of North Carolina at Chapel Hill


Collections: Computer Technologies and Information Sciences