The use of cloud computing data centers is growing rapidly to meet the tremendous increase in demand for high-performance computing (HPC), storage and networking resources for business and scientific applications. Virtual machine (VM) consolidation involves the live migration of VMs to run on fewer physical servers, and thus allowing more servers to be switched off or run on low-power mode, as to improve the energy consumption efficiency, operating cost and CO 2 emission. A crucial step in VM consolidation is host overload detection, which attempts to predict whether or not a physical server will be oversubscribed with VMs. In contrast to the majority of previous work which use CPU utilization as the sole indicator for host overload, a recent study has proposed a multiple regression host overload detection algorithm, which takes multiple factors into consideration: CPU, memory and network BW utilization. This paper provides further improvement along two directions. First, we provide Multi-Dimensional Regression Host Utilization (MDRHU) algorithms that combine CPU, memory and network BW utilization via Euclidean Distance (MDRHU-ED) and absolute summation (MDRHU-AS), respectively. This leads to improved results in terms of energy consumption and service level agreement violation. Second, the study explicitly takes real-world HPC workloads into consideration. Our extensive simulation study further illustrates the superiority of our proposed algorithms over existing methods. In particular, as compared to the most recently proposed multiple regression algorithm that is based on Geometric Relation (GR), our proposed algorithms provide an improvement of at least 12% in energy consumption, and an improvement of at least 80% in a metric that combines energy consumption, service-level-violation, and number of VM migrations.
The adoption of High-Performance Computing (HPC) applications has gained an extensive interest in the Cloud computing. Current cloud vendors utilize separate management tools for HPC and non-HPC applications, missing out on the consolidation benefits of virtualization. Non-HPC applications executed in the cloud may interfere with resource-hungry HPC applications, which is a key performance challenge. Furthermore, correlations between application major performance indicators, such as response time and throughput, with resource capacities reveal that conventional placement strategies are impacting virtual machine efficiency, resulting in poor resource optimization, increased operating expenses, and longer wait times. Since applications often underutilized the hardware, smart execution of HPC and Non-HPC applications on the same node can boost system and energy efficiency. This research incorporates proactive dynamic VM consolidation to enhance the resource usage and performance while maintaining energy efficiency. The proposed algorithm generates a workload-aware fine-grained classification by employing machine learning techniques to generate complimentary profiles that alleviate cross-application interference by intelligently co-locating non-HPC and HPC applications. The research used CloudSim to simulate real HPC workloads. The results verified that the proposed algorithm outperforms all heuristic methods with respect to the metrics in key areas.
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.