Kernel-mode scheduling server for CPU partitioning: a case study using the Windows research kernel

Authors:
Michael Schöbel;Andreas Polze
Affiliations:
Hasso Plattner Institute at University of Potsdam, Potsdam, Germany;Hasso Plattner Institute at University of Potsdam, Potsdam, Germany
Venue:
Proceedings of the 2008 ACM symposium on Applied computing
Year:
2008

Citing 5
Cited 1

CPU reservations and time constraints: efficient, predictable scheduling of independent activities

Proceedings of the sixteenth ACM symposium on Operating systems principles
A Soft Real Time Scheduling Server in UNIX Operating System

IDMS '97 Proceedings of the 4th International Workshop on Interactive Distributed Multimedia Systems and Telecommunication Services
Teaching operating systems: the windows case

Proceedings of the 37th SIGCSE technical symposium on Computer science education
A soft real-time scheduling server on the Windows NT

WINSYM'98 Proceedings of the 2nd conference on USENIX Windows NT Symposium - Volume 2
CPU reservations and time constraints: implementation experience on windows NT

WINSYM'99 Proceedings of the 3rd conference on USENIX Windows NT Symposium - Volume 3

Building a time- and space-partitioned architecture for the next generation of space vehicle avionics

SEUS'10 Proceedings of the 8th IFIP WG 10.2 international conference on Software technologies for embedded and ubiquitous systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

The ability to partition available computing power lays the foundation for predictable computing in different domains (e.g. multimedia applications or virtual machine execution). Fine-grained computing power control can be used to implement applications which adapt themselves to changing performance requirements. The scheduling server concept can be used for implementing fine-grained CPU partitioning. In the past, several user-mode implementations of this idea were realized without any kernel modification. This paper describes a kernel-mode implementation and compares two different design alternatives. Experiences of using the Windows Research Kernel (WRK) for experiments with the scheduler are given. Furthermore, benchmark results and possible applications are described.