A Cost-Sensitive Adaptation Engine for Server Consolidation of Multitier Applications

Abstract

Abstract. Virtualization-based server consolidation requires runtime resource reconfiguration to ensure adequate application isolation and performance, especially for multitier services that have dynamic, rapidly changing workloads and responsiveness requirements. While virtualization makes reconfiguration easy, indiscriminate use of adaptations such as VM replication, VM migration, and capacity controls has performance implications. This paper demonstrates that ignoring these costs can have significant impacts on the ability to satisfy response-time-based SLAs, and proposes a solution in the form of a cost-sensitive adaptation engine that weighs the potential benefits of runtime reconfiguration decisions against their costs. Extensive experimental results based on live workload traces show that the technique is able to maximize SLA fulfillment under typical time-of-day workload variations as well as flash crowds, and that it exhibits significantly improved transient behavior compared to approaches that do not account for adaptation costs.

Server Consolidation for Internet Applications in Virtualized Data Centers

“…From the service providers' perspective, a few works [30][31][32] studied on application-to-VM-to-PM mappings to minimize the cost operating in a per-application level. For example, SmartSLA [32] horizontally and vertically tuned VMs according to the average SLA penalty cost predicted using machine learning to minimize SLA penalty cost, rented hardware cost, and action cost.…”

“…For example, SmartSLA [32] horizontally and vertically tuned VMs according to the average SLA penalty cost predicted using machine learning to minimize SLA penalty cost, rented hardware cost, and action cost. Jung et al [30] predicted the behaves of workloads employing an autoregressive moving averages (ARMA) model and then tuned VMs to minimize SLA penalty. These works did not take advantage of consolidating multiple applications for improving resource utilizations and energy efficiency.…”

“…While they do not consider the placement of VMs on PMs, which is critical for improving energy efficiency. A few of elasticity managements [30][31][32] minimized one or more of SLA (service level agreement) penalty cost, rented hardware cost, software cost, and action (e.g. load balancing) cost for service providers.…”

Optimizing Performance and Energy Efficiency in Cloud Data Centers Through SLA-Aware Consolidation of Virtualized Resources

“…The algorithm uses historical data to forecast future demand and relies on periodic executions to minimize the number of physical servers to support the virtual machines. In [4], SLA requirements are represented as the pre-determined response times for each type of transactions specific to the web-application. Based on the utility function, the migration controller decides whether an effective reconfiguration is possible in order to fulfill the SLA.…”

Review of performance metrics for green data centers: a taxonomy study

“…Various research activities have been carried out to manage data centers in a "green" mode (the latter method), such as reducing data center temperature [28,39,42], increasing server utilization [21,26,37], and decreasing power consumption of computing resources [7,8,12,20]. A fundamental research topic for the above study is how we can define performance metrics to identify how "green" a data center is.…”