![]() | Only 14 pages are availabe for public view |
Abstract Cloud computing is emerging as a replacement for traditional physical hardware computing in the area of parallel and distributed computing. Clouds consist of a collectionof virtualized resources that can be provisioned on demand, depending on the users’ needs. Cloud computing faces the grand quantity of the user groups, as well as the quantity of tasks and massive data, so the processing is also very significant. Scheduling tasks efficiently has become an important problem to be solved in the field of cloud computing. All over the years, task scheduling was a major research area in different architectures and environments starting from single processor, passing by multiprocessor and ending with cloud computing. Cloud computing is a model for enabling ubiquitous network access to a shared pool of configurable computing resources where available resources must be checked and scheduled using an efficient task scheduler to be assigned to clients. Most of the existing task schedulers did not achieve the required standards and requirements. In this thesis, we propose a novel hybrid task scheduling algorithm named (SRDQ) based on both of shortest-job-first and round robin schedulers using a dynamic variable task quantum considering splitting the ready queue into two sub-queues, Q1, and Q2. Assigning tasks to resources from Q1 or Q2 are done mutually two tasks from Q1 and one task from Q2. The proposed algorithm was implemented in two different environments C# and Cloud Sim where the experimentations results and tests proved that the proposed algorithm had improved the average waiting and response times and also partially reduced the starvation over the state of art algorithms. |