Cloud Energy Broker: Towards SLA-Driven Green Energy Planning for IaaS Providers

Abstract

Demand for Green services is increasing considerably as people are getting more environmental conscious to build a sustainable society. Therefore, enterprise and clients want to shift their workloads towards green Cloud environment offered by the Infrastructure-asa-Service (IaaS) provider. The main challenge for an IaaS provider is to determine the best trade-off between its profit while using renewable energy and customers satisfaction. In order to address this issue, we provide a Cloud energy broker, which can adjust the availability and price combination to buy Green energy dynamically from the market to make datacenter green. We investigate a simplified power model from where we can formulate and predict power demand. Our energy broker tries to maximize of using renewable energy under strict budget constraint whereas it also tries to minimize the use of brown energy by capping the limit of overall energy consumption of datacenter. The energy broker was evaluated with a real workload traced by PlanetLab. Experimental results show that our energy broker successfully keeps the best trade-off.

Exploiting Renewable Sources: When Green SLA Becomes a Possible Reality in Cloud Computing

“…Usually, GEaaS provider add their green energy production to the Grid and IaaS provider consume the purchased energy from the same Grid, since there is only a single Global Grid in European electricity market. How IaaS and GEaaS provider can have a formal contract regarding energy procurement, is elaborately discussed in our previous work [4] as well. Therefore, we purchased green energy before hand by forecasting the green energy demand of SaaS providers.…”

“…Though some of the state-ofthe-art techniques claim to have reduced this value closer to 1.20, still most of today's data center have higher PUE values than 1.7 [24]. Therefore, in the planning phase [4], for transforming CPU utilization to power consumption, we traced CPU utilization for 7 days of 30 servers from PlanetLab [21] where CPU utilization has been traced for 500 different servers from across the world. As, building precise analytical models for modeling power consumption by modern multi-core CPUs makes a complex research problem, instead of using an analytical model of power consumption by a server, we utilize real data on power consumption provided by the results of the SPECpower benchmark.…”

“…Day-ahead pricing (DAP) is one of the many pricing methodology, which is widely used in deregulated electricity pricing market. In our previous work [4], we consider different GEaaS providers, which update DAP with information including probable generation of Green energy, price and availability. Usually, GEaaS provider add their green energy production to the Grid and IaaS provider consume the purchased energy from the same Grid, since there is only a single Global Grid in European electricity market.…”

“…In planning phase [4], we proposed a Cloud energy broker, which can adjust the availability and price combination to buy green energy dynamically 24 hours before from the REC market to make data center green for a specific (as a example 30%) portion. We investigated a simplified power model from where we could formulated and predicted power demand of a data center for next 24 hours by evaluating 7 days of real data, traced from PlanetLab [21].…”

“…To address this problem, we investigated Cloud Energy Broker in our previous work [4] and now are extending our research to propose Green SLA based on the availability of green energy by introducing the concept of Virtualization of green energy. The energy can be virtually green for a specific period of time if abundance of green energy is available aperiodically in shorter time interval along with the deficit of green energy in rest of the time frame.…”