Distributed Scheduling Framework - A Load Distribution Facility on Mach

With a network of workstations, load distribution strategies can be employed to better utilize resources. In practice, there is no single load distribution algorithm that outperforms all others in all cases. Therefore , it is desirable that diierent systems can use diierent load distribution algorithms. Even in the same system, the necessity of changing the load distribution algorithm may arise when the system connguration or other parameters, such as task types, are changed. Generally speaking, changing of load distribution algorithm is not easy since the scheduler running the algorithm is an operating system protected object. This paper presents the Distributed Scheduling Framework (DSF) | a load distribution facility on Mach, which allows various kinds of load distribution algorithms to be installed at user-level, and minimizes the eeorts needed with their implementation. The DSF thus provides a basis for experimentation with new load distribution algorithms, and allows such algorithms to be used to improve system performance as well.

[1]  Joseph S. Barrera A Fast Mach Network IPC Implementation , 1991, USENIX MACH Symposium.

[2]  Richard Draves,et al.  A Revised IPC Interface , 1990, USENIX MACH Symposium.

[3]  Alessandro Forin,et al.  UNIX as an Application Program , 1990, USENIX Summer.

[4]  R. Chawla,et al.  The Stealth distributed scheduler , 1991, [1991] Proceedings. 11th International Conference on Distributed Computing Systems.

[5]  David L. Black Scheduling support for concurrency and parallelism in the Mach operating system , 1990, Computer.

[6]  Philip H. Enslow What is a "Distributed" Data Processing System? , 1978, Computer.

[7]  Fred Douglis,et al.  Transparent process migration: Design alternatives and the sprite implementation , 1991, Softw. Pract. Exp..

[8]  Dejan S. Milojicic,et al.  Task Migration on the top of the Mach Microkernel , 1993, USENIX MACH Symposium.

[9]  David L. Black,et al.  Microkernel operating system architecture and Mach , 1991 .

[10]  Mukesh Singhal,et al.  Load distributing for locally distributed systems , 1992, Computer.

[11]  Thomas L. Casavant,et al.  Effects of Response and Stability on Scheduling in Distributed Computing Systems , 1988, IEEE Trans. Software Eng..

[12]  Thomas L. Casavant,et al.  A Taxonomy of Scheduling in General-Purpose Distributed Computing Systems , 1988, IEEE Trans. Software Eng..

[13]  Edward D. Lazowska,et al.  A Comparison of Receiver-Initiated and Sender-Initiated Adaptive Load Sharing , 1986, Perform. Evaluation.

[14]  Edward D. Lazowska,et al.  Adaptive load sharing in homogeneous distributed systems , 1986, IEEE Transactions on Software Engineering.

[15]  Miron Livny,et al.  Condor-a hunter of idle workstations , 1988, [1988] Proceedings. The 8th International Conference on Distributed.

[16]  Phillip Krueger,et al.  Two adaptive location policies for global scheduling algorithms , 1990, Proceedings.,10th International Conference on Distributed Computing Systems.

[17]  Dejan S. Milojicic Load distribution - implementation for the Mach microkernel , 1994 .

[18]  Michael Stumm,et al.  The design and implementation of a decentralized scheduling facility for a workstation cluster , 1988, [1988] Proceedings. 2nd IEEE Conference on Computer Workstations.

[19]  Miron Livny,et al.  Load Balancing in Homogeneous Broadcast Distributed Systems , 1982, SIGMETRICS.

[20]  John K. Ousterhout,et al.  Medusa: An experiment in distributed operating system structure (Summary) , 1979, SOSP '79.

[21]  Raphael A. Finkel,et al.  A Stable Distributed Scheduling Algorithm , 1981, IEEE International Conference on Distributed Computing Systems.