PaaS-IaaS Inter-Layer Adaptation in an Energy-Aware Cloud Environment

Abstract

Cloud computing providers resort to a variety of techniques to improve energy consumption at each level of the cloud computing stack. Most of these techniques consider resource-level energy optimisation at IaaS layer. This paper argues energy gains can be obtained by creating a cooperation between the PaaS layer (in charge of hosting the application/service) and the IaaS layer (in charge of handling the computing resources). It presents a novel method based on steering information and decision taking to trigger the PaaS and IaaS layers to adapt their energy mode in service operation, therefore enabling the Cloud stack to actively adapt to changing situations. Experimental results demonstrate such adaptation achieves dynamic energy management in each of the PaaS and IaaS cloud layers.

Rapid and accurate energy models through calibration with IPMI and RAPL

“…The device supervisor in this case when performing scheduling queries the energy modeler for the projected power usage of the deployment. At runtime, a self‐adaptation manager which can be used to manage QoS goals of the applications can utilize the energy modeler to gain current power consumption information for applications that are running . This is achieved through the interaction of the energy modeler and the monitoring infrastructure.…”

“…The first is at deployment time when workload is being placed upon the physical infrastructure. In Clouds the VM manager can utilize power consumption predictions for the placement of VMs . Similarly in HPC environments RJMS such as SLURM may also use power consumption predictions to place jobs.…”

“…At runtime, a self-adaptation manager which can be used to manage QoS goals of the applications can utilize the energy modeler to gain current power consumption information for applications that are running. 51 This is achieved through the interaction of the energy modeler and the monitoring infrastructure. The monitoring infrastructure provides information about the physical hosts such as utilization and power consumption, while the energy modeler determines the power consumption of each application.…”

“…In Clouds the VM manager can utilize power consumption predictions for the placement of VMs. 51 Similarly in HPC environments RJMS such as SLURM 3 may also use power consumption predictions to place jobs. The second is at operation time when the environment is monitored and this information is utilized to aid adaptation.…”

Energy - Efficient IaaS-PaaS Co-Design for Flexible Cloud Deployment of Scientific Applications

“…Unfortunately, their proposition does not present validation results. In [15], Djemame et al propose a self-adaptive IaaS-PaaS architecture that allows to run energy-efficient cloud operations. Applications are defined with complex user-defined SLA rules that express constraints on performance, energy and price aspects.…”