|
|
|
|
|
A Thread Pool Scheduling Optimization Method of Real-time System |
|
PP: 607-610 |
|
Author(s) |
|
Yi Liu,
|
|
Abstract |
|
To deal with concurrent requests, thread pool technology is wildly used in multi-task real-time systems whose chief demand is timeout avoidance. In order to reduce requesting timeout ratio, least square support vector machine(LS-SVM) algorithm whose parameters optimization was based on hybrid quantum-behaved particle swarm optimization(HQPSO) was applied to estimate the execution time of requests. Also, a scheduling priority algorithm of thread pool was designed based on the estimation results. In performance test, a node of wireless sensor network is implemented to test the requesting timeout ratio of HQPSO LS-SVM based on thread pool and other ones in different states. Test result indicates that HQPSO LS-SVM based on thread pool has remarkable superiority of timeout avoidance while the thread amount of thread pool is set properly.
|
|
|
|
|
|