Summary: An Optimal k-Exclusion Real-Time Locking Protocol
Motivated by Multi-GPU Systems
Glenn A. Elliott and James H. Anderson
Department of Computer Science, University of North Carolina at Chapel Hill
Graphics processing units (GPUs) are becoming increas-
ingly important in today's platforms as their increased gen-
erality allows for them to be used as powerful co-processors.
In previous work, we have found that GPUs may be in-
tegrated into real-time systems through the treatment of
GPUs as shared resources, allocated to real-time tasks
through mutual exclusion locking protocols. In this pa-
per, we present an optimal k-exclusion locking protocol for
globally-scheduled job-level static-priority (JLSP) systems.
This protocol may be used to manage a pool of GPU re-
sources in such systems.
The widespread adoption of multicore technologies in
the computing industry has prompted research in a wide
variety of computing fields with the goal of better under-