Cloud Computing has emerged as a computing paradigm where services are provided through the internet in recent years. Offering on-demand services has transformed the IT companies' working environment, leading to a linearly increasing trend of its usage. The provisioning of the Computing infrastructure is achieved with the help of virtual machines. A great figure of physical devices is required to satisfy the users' resource requirements. To meet the requirements of the submitted workloads that are usually dynamic, the cloud data centers cause the over-provisioning of cloud resources. The result of this over-provisioning is the resource wastage with an increase in the levels of energy consumption, causing a raised operational cost. High CO
2
emissions result from this huge energy consumption by data centers, posing a threat to environmental stability. The environmental concern demands for the controlled energy consumption, which can be attained by optimal usage of resources to achieve in the server load, by minimizing the number of active nodes, and by minimizing the frequency of switching between active and de-active server mode in the data center. Motivated by these actualities, we discuss numerous statistical, deterministic, probabilistic, machine learning and optimization based computational solutions for the cloud computing environment. A comparative analysis of the computational methods, on the basis of architecture, consolidation step involved, objectives achieved, simulators involved and resources utilized, has also been presented. A taxonomy for virtual machine (VM) consolidation has also been derived in this research article followed by emerging challenges and research gaps in the field of VM consolidation in cloud computing environment.
Job scheduling is a fundamental issue in achieving a high performance on the Grids. In grid computing several applications require numerous resources for execution which are not often available for them, thus presence of a scheduling system to allocate resources to input jobs is vital. This paper introduces a model and a job scheduling algorithm in grid computing environments. Computational grids have the potential for solving Large-scale scientific problems using heterogeneous and geographically distributed resources. One problem that is critical to effective utilization of computational grids is the efficient scheduling of jobs. This work addresses this problem by describing and evaluating a grid scheduling architecture and a job-scheduling algorithm. The research work introduces NSA (node-selection algorithm) at the global scheduler and the PSA (processor selection algorithm) at the local scheduler. The NSA is based on the rule that the lightloaded processing node is selected for the job allocation. This technique fetches the jobs from the Global job queue that is ready to execute and assign these jobs to the best nodes of the grid. The PSA (processor selection algorithm) schedule the job to the processor of a selected node having maximum available CPU resource (ACR).The algorithm has been tested in a simulated grid environment.
Fundamentally, a strategy considering the effective utilization of resources results in the better energy efficiency of the system. The aroused interest of users in cloud computing has led to an increased power consumption making the network operation costly. The frequent requests from the users asking for computing resources can lead to instability in the load of the computing system. To perform the load balancing in the host, migration of the virtual machines from the overloaded and underloaded hosts needs to be done, which is considered an important facet concerning energy consumption. The proposed Particle Swarm Optimization based Resource Aware VM Placement (RAPSO_VMP) scheme aims to place the migrated virtual machines. RAPSO_VMP takes into consideration multiple resources like CPU, storage, and memory while trying to optimize the overall resource utilization of the system. According to the simulation analysis, the proposed RAPSO_VMP scheme shows an improvement of 5.51% in energy consumption, reduced the number of migrations by 9.12%, and the number of hosts shutdowns 22.74%.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.