Developing and Evaluating a Flexible Framework for Dynamic Distributed Real-Time Scheduling

1998 USENIX Student Research Grant Project



Project Participants:



Project Description:

In theory, dynamic scheduling can alleviate many of the limitations of static scheduling. In practice, however, implementations of conventional dynamic scheduling techniques, such as earliest deadline first (EDF) and minimum laxity first (MLF), do not preserve the schedulability guarantees offered by static techniques. To address the limitations of existing dynamic scheduling techniques, we propose to enhance the existing static scheduling framework found in The ACE ORB (TAO) to support dynamic and hybrid static/dynamic scheduling of operations for ORB middleware and real-time applications.

The scope of this project consists of two main objectives. First, we propose to complete current work on dynamic scheduling features in TAO's existing event service and scheduling service, and to add features for distributed dynamic scheduling to these services using a flexible strategized scheduling framework. Second, we propose to conduct a set of experiments to examine the effects of varying platform, application, and middleware factors on scheduling behavior.


Related Publications: