Richard E. Kavanagh

Towards an Energy-Aware Cloud Architecture for Smart Grids

Energy consumption in Cloud computing is a significant issue in regards to aspects such as the cost of energy, cooling in the data center and the environmental impact of cloud data centers. Smart grids offers the prospect of dynamic costs for a data center’s energy usage. These dynamic costs can be passed on to Cloud users providing incentives for users to moderate their load while also ensuring the Cloud providers are insulated from fluctuations in the cost of energy. The first step towards this is an architecture that focuses on energy monitoring and usage prediction. We provide such an architecture at both the PaaS and IaaS layers, resulting in energy metrics for applications, VMs and physical hosts, which is key to enabling active demand in cloud data centers. This architecture is demonstrated through our initial results utilising a generic use case, providing energy consumption information at the PaaS and IaaS layers. Such monitoring and prediction provides the groundwork for providers passing on energy consumption costs to end users. It is envisaged that the resulting varying price associated with energy consumption can help motivate the formation of methods and tools to support software developers aiming to optimise energy efficiency and minimise the carbon footprint of Cloud applications.

A grid broker pricing mechanism for temporal and budget guarantees

We introduce a pricing mechanism for Grid computing, with the aim of showing how a broker can accept the most appropriate jobs to be computed on time and on budget. We analyse the mechanisms performance via discrete event simulation, and illustrate its viability, the benefits of a new admission policy and to how slack relates to machine heterogeneity.

Accuracy of Energy Model Calibration with IPMI

Energy consumption in Cloud computing is a significantissue and affects aspects such as the cost of energy, coolingin the data center and the environmental impact of cloud data centers. Monitoring and prediction provides the groundwork for improving the energy efficiency of data centers. This monitoring however is required to be fast and efficient without unnecessary overhead. It is also required to scale to the size of a data center where measurement through directly attached Watt meters is unrealistic. This therefore requires models that translate resource utilisation into the power consumed by a physical host. These models require calibrating and are hence subject to error. We discuss the causes of error within these models, focusingupon the use of IPMI in order to gather this data. We makerecommendations on ways to mitigate this error without overly complicating the underlying model. The final result of these models is a Watt meter emulator that can provide values for power consumption from hosts in the data center, with an average error of 0.20W.

Towards energy aware cloud computing application construction

The energy consumption of cloud computing continues to be an area of significant concern as data center growth continues to increase. This paper reports on an energy efficient interoperable cloud architecture realised as a cloud toolbox that focuses on reducing the energy consumption of cloud applications holistically across all deployment models. The architecture supports energy efficiency at service construction, deployment and operation. We discuss our practical experience during implementation of an architectural component, the Virtual Machine Image Constructor (VMIC), required to facilitate construction of energy aware cloud applications. We carry out a performance evaluation of the component on a cloud testbed. The results show the performance of Virtual Machine construction, primarily limited by available I/O, to be adequate for agile, energy aware software development. We conclude that the implementation of the VMIC is feasible, incurs minimal performance overhead comparatively to the time taken by other aspects of the cloud application construction life-cycle, and make recommendations on enhancing its performance.

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

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 optimization 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.

Energy Efficiency Support Through Intra-layer Cloud Stack Adaptation

Energy consumption is a key concern in cloud computing. The paper reports on a cloud architecture to support energy efficiency at service construction, deployment, and operation. This is achieved through SaaS, PaaS and IaaS intra-layer self-adaptation in isolation. The self-adaptation mechanisms are discussed, as well as their implementation and evaluation. The experimental results show that the overall architecture is capable of adapting to meet the energy goals of applications on a per layer basis.

Towards an interoperable energy efficient Cloud computing architecture - practice & experience

The energy consumption of Cloud computing continues to be an area of significant concern as data center growth continues to increase. This paper reports on an energy efficient interoperable Cloud architecture realized as a Cloud toolbox that focuses on reducing the energy consumption of Cloud applications holistically across all deployments models. The architecture supports energy efficiency at service construction, deployment, and operation and interoperability through the use of the Open Virtualization Format (OVF) standard. We discuss our practical experience during implementation and present an initial performance evaluation of the architecture. The results show that the implementing Cloud provider interoperability is feasible and incurs minimal performance overhead during application deployment in comparison to the time taken to instantiate Virtual Machines.

Towards an Energy-Aware Framework for Application Development and Execution in Heterogeneous Parallel Architectures

The Transparent heterogeneous hardware Architecture deployment for eNergy Gain in Operation (TANGO) project’s goal is to characterise factors which affect power consumption in software development and operation for Heterogeneous Parallel Hardware (HPA) environments. Its main contribution is the combination of requirements engineering and design modelling for self-adaptive software systems, with power consumption awareness in relation to these environments. The energy efficiency and application quality factors are integrated into the application lifecycle (design, implementation and operation). To support this, the key novelty of the project is a reference architecture and its implementation. Moreover, a programming model with built-in support for various hardware architectures including heterogeneous clusters, heterogeneous chips and programmable logic devices is provided. This leads to a new cross-layer programming approach for heterogeneous parallel hardware architectures featuring software and hardware modelling. Application power consumption and performance, data location and time-criticality optimization, as well as security and dependability requirements on the target hardware architecture are supported by the architecture.

The ISQoS grid broker for temporal and budget guarantees

We introduce our Grid broker that uses SLAs in job submission with the aim of ensuring jobs are computed on time and on budget. We demonstrate our brokers ability to perform negotiation and to select preferentially higher priority jobs, in a tender market and discuss the architecture that makes this possible. We additionally show the effects of rescheduling and how careful consideration is required in order to avoid price instability. We therefore make recommendations upon how to maintain this stability, given rescheduling.