International Journal of Computer Applications (0975 – 8887) Volume 182 – No.4, July 2018 17 Proposing Priority based Dynamic Resource Allocation [PDRA] Model in Cloud Computing Amit Chaturvedi, PhD MCA Deptt Govt. Engineering College, Ajmer Rajasthan, India Praveen Sengar Ph.D. Scholar, Bhagwant Univ., Ajmer Rajasthan, India Kalpana Sharma, PhD Computer Sc. Deptt Bhagwant Univ., Ajmer Rajasthan, India ABSTRACT Scaling of resources in cloud computing is essential for the better utilization of resources. Dynamic allocation of the resources / VMs in the multi-tenant environment is the need of the cloud computing. Virtualization technologies evolved to help IT organizations and to improve the efficiency of their hardware resources by partitioning hardware to provide simultaneous support to multiple applications and their corresponding software stacks. If the resource utilization is not properly allocated to applications, it will lead to the faulty services to the customers. The Cloud is the hub of resources, and can be used by any client on rental bases and on no demand resources can left with no usage. Clients/ Brokers may request for the multiple VMs/ other resources like, applications, database, operating system etc, but the resources are limited. So, there is the need of such a system to handle this allocation and deallocation of resources or VMs. By this PDRA model, authors have presented an idea to handle the resources/ VMs allocation and deallocation system. General Terms Your general terms must be any term which can be used for general classification of the submitted material such as Pattern Recognition, Security, Algorithms et. al. Keywords Virtual Machine, dynamic allocation, elasticity, cloud, scaling.. 1. INTRODUCTION Introduction: Scaling of resources in cloud computing is essential for the better utilization of resources. As we know that the main purpose of the cloud computing is to share computing resources like applications, database, operating system, etc. to the world of computers for maximizing the utilization and return. Cloud scaling enables scale-up and down automatically. We can create thousands of server instances, brokers, virtual machines etc and allocate them simultaneously. These instances can be controlled separately by the medium of middleware known as Virtual Machine Monitors or Hypervisors. Flexibility in creating and hosting the resources and its services can provide multiple choices for instance creating and could be configured for the memory requirements, operating system, processors, virtual machines, etc. As the number of virtual machines increases, they want more operating memory, stable storage, processing capability, and obviously more bandwidth to meet the required performance. To tackle the workload on servers, there is the requirement of dynamic provisioning of different data centres with guarantee of availability of the resources and services required. In virtualized multi-tenant cloud environment, applications encapsulate and segregate application performance by using virtual machines (VMs). Virtualization technologies evolved to help IT organizations and to improve the efficiency of their hardware resources by partitioning hardware to provide simultaneous support to multiple applications and their corresponding software stacks. If the resource utilization is not properly allocated to applications, it will lead to the faulty services to the customers. Resource scaling can resolve issues of application migration conflicts as well, which are due to compatibility issues in multi-tenant cloud environment. Resource scaling should do check of compatibility of applications during application transfer to cloud. Resource scaling is also very important in case of processor availability so that there should be no issue of service due to technical snag during execution of applications. Multi-tenant clouds are central resources for computing power, and ultimately the cloud organization can distribute same resource among multiple clients without mingling each other‟s data. The Cloud is the hub of resources, and can be used by any client on rental bases and on no demand resources can left with no usage. Running huge data centres is unsustainable for organizations with server‟s resources being idle most of the time. Small companies can upload their data to cloud and take‟s cloud resources on rental bases. Resource scaling issues are caused by clients most of the time as they are not aware of the workload demand by their users. Customers can be increased at the times or stayed low sometimes. Resources allocation and de-allocation is important to schedule resources perfectly while entertaining requests from clients. The whole point of cloud computing is to achieve economies of scale by managing a very large pool of computing resources in a highly economic and efficient fashion. By using the elasticity of computing resources, systems may occupy and release resources to dynamic workloads and paying for those only which are needed and used. This characteristic of cloud computing is basically known elasticity and the allocation and de-allocation dynamically by controlling through an automated system is the rapid elasticity. The multi-tier applications services that allocates and relieves resources in segments, such as virtual server instances of predefined sizes. It highlights on elastic control of the storage tier, in which storage and removing from machines or brick needs re-balancing stored data on all the machines .the new challenges for storage tier presents for elastic controls. 2. RELATED WORK In cloud computing various types of resources like CPU, Memory, OS, Application Software etc. are used. A cloud server, which has sufficient resources all the time for its clients as resource pools, efficiently and dynamically allocates or deallocates these resources, is considered good for its