Adaptive QoS and resource management using a posteriori workload characterizations

Certain real-time applications must operate in highly dynamic environments (e.g., battle environments), thereby precluding accurate characterization of the applications' workloads by static models. Thus, static guarantees of real-time performance based on a priori characterizations are not possible. However, potential benefits of a posteriori approaches are significant, including the ability to function correctly in dynamic environments (through adaptability to unforeseen conditions), and higher actual utilization of computing resources. In this paper, we present an approach that is appropriate for systems which experience large variations in workload. A distributed collection of computing resources is managed by continuously computing and assessing QoS and resource utilization metrics that are determined a posteriori. The utility of our approach is shown by applying it to a large, experimental distributed Navy computing system.

[1]  Robert D. Harrison.  Combat System Prerequisites on Supercomputer Performance Analysis , 1992, NATO ASI RTC.

[2]  Kang G. Shin,et al.  Application of real-time monitoring to scheduling tasks with random execution times , 1989, [1989] Proceedings. Real-Time Systems Symposium.

[3]  Azer Bestavros,et al.  Statistical rate monotonic scheduling , 1998, Proceedings 19th IEEE Real-Time Systems Symposium (Cat. No.98CB36279).

[4]  Jun Sun,et al.  Probabilistic performance guarantee for real-time tasks with varying computation times , 1995, Proceedings Real-Time Technology and Applications Symposium.

[5]  John P. Lehoczky,et al.  Real-time queueing theory , 1996, 17th IEEE Real-Time Systems Symposium.

[6]  Lonnie R. Welch,et al.  Modeling resource contention among distributed periodic processes , 1992, [1992] Proceedings of the Fourth IEEE Symposium on Parallel and Distributed Processing.

[7]  Dieter K. Hammer,et al.  Incorporating Temporal Considerations during Assignment and Pre-run-time Scheduling of Objects and Processes , 1996, J. Parallel Distributed Comput..

[8]  Lonnie R. Welch,et al.  Response time prediction for distributed processes specified in CaRT-Spec , 1995 .

[9]  Lui Sha,et al.  Rate Monotonic Analysis for Real-Time Systems , 1991 .

[10]  Jun Sun,et al.  Bounding completion times of jobs with arbitrary release times and variable execution times , 1996, 17th IEEE Real-Time Systems Symposium.

[11]  Giorgio C. Buttazzo,et al.  Integrating multimedia applications in hard real-time systems , 1998, Proceedings 19th IEEE Real-Time Systems Symposium (Cat. No.98CB36279).

[12]  Krithi Ramamritham,et al.  Distributed Scheduling of Tasks with Deadlines and Resource Requirements , 1989, IEEE Trans. Computers.

[13]  David Lorge Parnas,et al.  Scheduling Processes with Release Times, Deadlines, Precedence, and Exclusion Relations , 1990, IEEE Trans. Software Eng..

[14]  Pradeep K. Khosla,et al.  Mechanisms for detecting and handling timing errors , 1997, Commun. ACM.

[15]  Scott A. Brandt,et al.  A dynamic quality of service middleware agent for mediating application resource usage , 1998, Proceedings 19th IEEE Real-Time Systems Symposium (Cat. No.98CB36279).

[16]  Wei-Kuan Shih,et al.  Algorithms for scheduling imprecise computations , 1991, Computer.

[17]  H. Streich,et al.  On the design of a dynamic distributed real-time environment , 1997, Proceedings of 5th International Workshop on Parallel and Distributed Real-Time Systems and 3rd Workshop on Object-Oriented Real-Time Systems.

[18]  Binoy Ravindran,et al.  Specification and modeling of dynamic, distributed real-time systems , 1998, Proceedings 19th IEEE Real-Time Systems Symposium (Cat. No.98CB36279).

[19]  Tei-Wei Kuo,et al.  Incremental Reconfiguration and Load Adjustment in Adaptive Real-Time Systems , 1997, IEEE Trans. Computers.

[20]  Chung Laung Liu,et al.  Scheduling Algorithms for Multiprogramming in a Hard-Real-Time Environment , 1989, JACM.

[21]  Marc Donner,et al.  Run-time monitoring of real-time systems , 1991, [1991] Proceedings Twelfth Real-Time Systems Symposium.