Scheduling Algorithms for Multiprogramming in a Hard-Real-Time Environment
Journal of the ACM (JACM)
The Aperiodic Multiprocessor Utilization Bound for Liquid Tasks
RTAS '02 Proceedings of the Eighth IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS'02)
Analyzing Fixed-Priority Global Multiprocessor Scheduling
RTAS '02 Proceedings of the Eighth IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS'02)
Schedulability Analysis and Utilization Bounds for Highly Scalable Real-Time Services
RTAS '01 Proceedings of the Seventh Real-Time Technology and Applications Symposium (RTAS '01)
Static-Priority Scheduling on Multiprocessors
RTSS '01 Proceedings of the 22nd IEEE Real-Time Systems Symposium
Stochastic Analysis of Expected Schedulability for Real-Time Tasks on a Single Computing System
DS-RT '08 Proceedings of the 2008 12th IEEE/ACM International Symposium on Distributed Simulation and Real-Time Applications
| Hi-index | 0.00 |
We provide a constant time schedulability test for amultiprocessor server handling aperiodic tasks. Dhall麓seffect is avoided by dividing the tasks in two priorityclasses based on task utilization: heavy and light. Weprove that if the load on the multiprocessor server staysbelow Uthreshold = 3 - \[\sqrt 7 \] \approx 35.425%, the server canaccept incoming aperiodic tasks and guarantee that thedeadlines of all accepted tasks will be met. 35.425% utilizationis also a threshold for a task to be characterized asheavy.The bound Uthreshold = 3 - \[\sqrt 7 \] \approx 35.425% is easy-to-use,but not sharp if we know the number of processors inthe multiprocessor. For a server with m processors, wecalculate a formula for the sharp bound Uthreshold(m), whichconverges to Uthreshold from above as m \rightarrow \infty.The results are based on a utilization function um(x) =2(1 - x)/(2 + \[\sqrt {2 + 2x} \]) + x/m. By using this function, theperformance of the multiprocessor can in some cases beimproved beyond Uthreshold(m) by paying the extra overheadof monitoring the individual utilization of the current tasks.