Loading…
A Methodology for Partitioning Real-Time ADA (Tradename) Software for Distributed Targets
Task allocation, the assignment of tasks to processors, is an important problem in the design of distributed real-time systems. A task allocation scheme is required in order to produce a feasible partition of tasks across processors in the system, and to ensure high performance, especially for syste...
Saved in:
Main Authors: | , |
---|---|
Format: | Report |
Language: | English |
Subjects: | |
Online Access: | Request full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | Task allocation, the assignment of tasks to processors, is an important problem in the design of distributed real-time systems. A task allocation scheme is required in order to produce a feasible partition of tasks across processors in the system, and to ensure high performance, especially for systems with real time operational requirements. Several researchers have studied the task allocation problem for distributed system. One of the problems of current interest in real-time Ada software for distributed systems. This document discuss source code allocation, the development of multitasking Ada software, which is then partitioned; this approach allows development and testing of the software as a whole before allocation. The source code allocation approach has a number of important advantages. In an allocation scheme. it is preferable to place minimum design restrictions on software development, especially since the target architecture may, in most cases, not be known. It is also preferable to minimize the burden on the complier. Additionally, since the underlying system constraints may vary. a good partition can be achieved only through iteration, and therefore, creating new partitions should be inexpensive. |
---|