Prediction-based VM provisioning and admission control for multi-tier web applications

Abstract

We present a prediction-based, cost-efficient Virtual Machine (VM) provisioning and admission control approach for multi-tier web applications. The proposed approach provides automatic deployment and scaling of multiple web applications on a given Infrastructure as a Service (IaaS) cloud. It monitors and uses collected resource utilization metrics itself and does not require a performance model of the applications or the infrastructure dynamics. The approach uses the OSGi component model to share VM resources among deployed applications, reducing the total number of required VMs. The proposed approach comprises three sub-approaches: a reactive VM provisioning approach called ARVUE, a hybrid reactive-proactive VM provisioning approach called Cost-efficient Resource Allocation for Multiple web applications with Proactive scaling (CRAMP), and a session-based adaptive admission control approach called adaptive Admission Control for Virtualized Application Servers (ACVAS). Performance under varying load conditions is guaranteed by automatic adjustment and tuning of the CRAMP and ACVAS parameters. The proposed approach is demonstrated in discrete-event simulations and is evaluated in a series of experiments involving synthetic as well as realistic load patterns.

A predictive approach for the efficient distribution of agent-based systems on a hybrid-cloud

“…A set of scheduling algorithms have been studied to predict and optimize the tasks outsource cost. In [31] a prediction approach to ensure effective VM-provisioning and admission control of multi-tiers web-applications is proposed. The approach provides an adaptive automatic deployment solution on a given IaaS provider (Amazon EC2) by using a set of auto-scaling and admission control algorithms.…”

Tackling uncertainty in long-term predictions for host overload and underload detection in cloud computing

“…Ashraf et al [27] propose a load prediction approach for VM resource allocation and admission control of multitier web applications in cloud computing. Their prediction method is based on a two step procedure.…”

Optimized Deployment Plans for Platform as a Service Clouds

“…The platform in PaaS can be simply regarded as middleware, which any IaaS user also can leverage to deploy a service in multiple VMs and then use a scaling and load balancing mechanism to distribute the work among dynamically allocated VMs. Indeed, many scholars have studied the problem of horizontal scaling from the perspective of the IaaS user [2,3,[17][18][19][20]. A common theme in these works is that cloud providers leverage existing IaaS clouds to deploy services in combination with an auto-scaling mechanism for deciding how many VMs a given service will require at any given moment, either now or in the future, as well as a load balancer for distributing the load between these VMs.…”

“…Determining the resource requirements of services is out of scope for this paper, but there are several options, e.g. deriving an empirical performance model at runtime, constructing an analytic model or directly measuring the system [2].…”