Cluster Based Real Time Scheduling for Distributed System

Real time tasks scheduling on a distributed system is a complex problem. The existing real time tasks scheduling techniques are primarily based on partitioned and global scheduling. In partitioned based scheduling the tasks are assigned on a dedicated processor. The advantages of partitioned based approach is existing uni-processor scheduling techniques can be used; no migration overheads but task assignment is NP hard problem and optimal utilization of processing nodes is not possible. In global scheduling all tasks are maintained in a single tasks queue and allocated to multiple processing nodes. The advantage of global scheduling is optimal utilization of processing nodes but suffer from high migration and preemption overheads. This paper proposed cluster based real time tasks scheduling on a distributed system which is a hybrid scheduling approach where processing nodes group into cluster and scheduling using global scheduling. The simulation result shows that the proposed scheduling increases the tasks acceptance ratio, resource utilization as compared to partitioned and global scheduling and reduces migration as well as preemption overheads.

Processor Scheduling in High-Performance Computing (HPC) Environment

Processor scheduling is one of the thrust areas in the field of computer science. The future technologies use a huge amount of processing for execution of their tasks like huge games, programming software, and in the field of quantum computing. In real-time, many complex problems are solved by GPU programming. The primary concern of scheduling is to reduce the time complexity and manpower. Several traditional techniques exit for processor scheduling. The performance of traditional techniques is reduced when it comes to the huge processing of tasks. Most scheduling problems are NP-hard in nature. Many of the complex problems are recently solved by GPU programming. GPU scheduling is another complex issue as it runs thousands of threads in parallel and needs to be scheduled efficiently. For such large-scale scheduling problems, the performance of state-of-the-art algorithms is very poor. It is observed that evolutionary and genetic-based algorithms exhibit better performance for large-scale combinatorial and internet of things (IoT) problems.