Please use this identifier to cite or link to this item:
|Optimized Hybrid Task Scheduling Algorithm in Cloud
|Cloud computing;Task Scheduling;Scheduling;Service Provider;Internet
|Cloud Computing is one of the buzzwords of the 21st century. Cloud has transformed a colossal part of the IT industry, making software more alluring than before. Cloud Computing is termed as a network of remote servers that are used to store, manage and process data over the Internet. The most captivating feature of Cloud is the distribution of on-demand computing resources. The Cloud end-users need to pay a certain amount of money to the Service Provider in order to attain the various services of the Cloud. As such, the end-users expect the Provider to provide services having the most Quality of Service (QoS) along with less cost. On the other hand, the Cloud Providers also need to optimize costs to maintain the data center. In order to achieve efficient QoS, a proper scheduling technique is required. Maintaining the resources is very important as scaling up or down of the resources occurs quite often due to the fluctuation of the resources used by the consumers. Therefore, a proper utilization of the resources becomes a key factor to obtain maximum QoS. Therefore, an efficient scheduling technique is required for utilizing the resources through parallelism such that the makespan time and the cost can be improved. In this thesis work, an enhanced scheduling algorithm- Optimized Hybrid Task Scheduling (OHTS) algorithm is proposed which is based on the hybrid approach of MaxMin and Min-Min strategy that helps in allocating tasks to appropriate resources. Min-Min algorithm always starts with task with minimum time and ignores the task with longer execution time. Again, Max-Min algorithm starts with tasks that have highest execution time. As such, the processes with shorter execution time are not processed by the system. Hence an algorithm is needed which would overcome the shortcomings of these algorithms and provide better solutions in scheduling. The proposed algorithm helps in achieving better optimized results. The simulation results obtained shows that the performance of this proposed algorithm is more efficient as compared to Min-Min and Enhanced Load Balancing Min-Min strategy in terms of the makespan time and the costs.
|Master of Engineering -CSE
|Appears in Collections:
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.