Deadlock Preventıon Usıng Schedulabılıty In Vırtualızed Cloud Envıronment

Abstract

Cloud computing is to provide a on demand IT resources/services like server, storage, database, networking, analytics, software etc. over internet. It provide a diversity  of  distributed computing  system, and it has a group of virtualized  and   inter-connected  systems  based on provider–stage agreements  between  the  consumers  and  service providers. Here, a computer architecture and algorithms were developed for cloud platform   via the dynamic provisioned of virtual systems to perform and support soft and real -time task scheduling. The architecture integrated soft real-time task scheduling algorithms, namely master node and virtual machine node schedules. In addition Adaptive Adaptive Scoring Job scheduling algorithm is also applied. The motive of cloud computing is to tie together the power of both distributed computing and parallel computing as well as   aggregate idle resources on the Internet such as Central Processing Unit (CPU) cycles and storage spaces for better consumption. Cloud computing, which connects a commodity hardware with high speed networks, that can meet the same computing power as a supercomputer does, with a lower cost. However, cloud is a heterogeneous system. Independent tasks scheduling in cloud is more complicated. To make use of the power of cloud entirely, we require an well-organized job scheduling algorithm to allocate jobs to resources in a cloud. This project proposes an Adaptive Scoring Job scheduling algorithm (ASJS) for the cloud environment. Compared to other algorithms, it can reduce the time of job completion, which may compose of computing-intensive jobs and data-intensive jobs. Python 3.6 is used as the front end language to develop the application.