End-to-End Schedulability Analysis for Bi-directional Real-Time Multistage Pipeline

In this paper we discuss the end-to-end schedulability analysis for the bi-directional multistage pipeline systems, in which the arriving tasks could traverse the pipeline nodes in either of the two opposite directions. By accounting for the task execution overlap among nodes, we could derive an end-to-end delay bound formula for each task, which transforms the schedulability analysis of the task in the pipeline system into the schedulability analysis of that task in a corresponding single node system essentially, then the uniprocessor schedulability analysis theory could be applied to test the schedulability of this task in the original pipeline system. This transformation makes no assumptions on task arrival patterns and thus could be applied to both periodic and a periodic arrival patterns. We compare this technique with previous pipeline schedulability test techniques and show by simulations that our technique could derive better end-to-end delay bound except when node count is small and the ratio of deadline to period is large.

[1]  Chenyang Lu,et al.  A utilization bound for aperiodic tasks and priority driven scheduling , 2004, IEEE Transactions on Computers.

[2]  Tarek F. Abdelzaher,et al.  Transforming Distributed Acyclic Systems into Equivalent Uniprocessors under Preemptive and Non-Preemptive Scheduling , 2008, 2008 Euromicro Conference on Real-Time Systems.

[3]  Tarek F. Abdelzaher,et al.  Towards feasible region calculus: an end-to-end schedulability analysis of real-time multistage execution , 2005, 26th IEEE International Real-Time Systems Symposium (RTSS'05).

[4]  Tarek F. Abdelzaher,et al.  A Delay Composition Theorem for Real-Time Pipelines , 2007, 19th Euromicro Conference on Real-Time Systems (ECRTS'07).

[5]  John A. Clark,et al.  Holistic schedulability analysis for distributed hard real-time systems , 1994, Microprocess. Microprogramming.

[6]  Patrick Lardieri,et al.  End-to-End Scheduling Strategies for Aperiodic Tasks in Middleware , 2005 .

[7]  Tarek F. Abdelzaher,et al.  End-to-End Delay Analysis of Distributed Systems with Cycles in the Task Graph , 2009, 2009 21st Euromicro Conference on Real-Time Systems.

[8]  Jia Xu,et al.  Multiprocessor Scheduling of Processes with Release Times, Deadlines, Precedence, and Exclusion Relations , 1993, IEEE Trans. Software Eng..

[9]  John P. Lehoczky,et al.  The rate monotonic scheduling algorithm: exact characterization and average case behavior , 1989, [1989] Proceedings. Real-Time Systems Symposium.

[10]  Krithi Ramamritham,et al.  Allocation and scheduling of complex periodic tasks , 1990, Proceedings.,10th International Conference on Distributed Computing Systems.

[11]  Jay K. Strosnider,et al.  Distributed pipeline scheduling: end-to-end analysis of heterogeneous, multi-resource real-time systems , 1995, Proceedings of 15th International Conference on Distributed Computing Systems.

[12]  Gerhard Fohler,et al.  Static scheduling of pipelined periodic tasks in distributed real-time systems , 1997, Proceedings Ninth Euromicro Workshop on Real Time Systems.

[13]  Alan Burns,et al.  Applying new scheduling theory to static priority pre-emptive scheduling , 1993, Softw. Eng. J..

[14]  Michael González Harbour,et al.  Offset-based response time analysis of distributed systems scheduled under EDF , 2003, 15th Euromicro Conference on Real-Time Systems, 2003. Proceedings..

[15]  Giorgio C. Buttazzo,et al.  A hyperbolic bound for the rate monotonic algorithm , 2001, Proceedings 13th Euromicro Conference on Real-Time Systems.