Serpens for Kepler
Data Centres Optimization for Energy-Efficient and EnvironmentalLy Friendly Internet
DOLFIN aims to significantly contribute towards improving the energy efficiency of Data Centres and stabilizing of Smart Grids, a holistic approach, across networks of Data Centres and Smart Grids. Today, most DCs are part of computing and storage clouds, offering their customers Virtual Machines (VMs) as a virtual operating environment. DOLFIN will model, monitor, and measure energy consumption and enable seamless, autonomic migration of VMs between servers of the same DC or across a group of Energy-conscious, Synergetic DCs, aiming to:
- optimize the overall energy consumption by dynamically changing the percentage of active versus stand-by servers and the load per active server in a DC
- stabilize the Smart Grid energy distribution, under peak load and increased demand, by dynamically changing the energy consumption/ production requirements of the local DCs
Scheduled finish date: 2016-10-01
The DOLFIN high-level objectives for improving Data Centre energy efficiency and for stabilizing smart grids:
1) Data Centre energy Consumption Optimisation Platform (eCOP)
The eCOP will provide for continuous monitoring, energy benchmarking, dynamic control and adaptive optimisation of the Data Centre infrastructure, including ICT and HVAC equipment. DOLFIN will create a components/modules utilisation database of the ICT infrastructure, either by directly collecting utilisation information from the ICT devices or by interfacing with existing management operation systems.
2) Energy-conscious Synergetic Data Centres
DOLFIN foresees a group of energy-aware Data Centres, operating in geographically dispersed locations, in a collaborative manner. Over this group, DOLFIN will design, implement, and deploy a solution that will allow the dynamic reallocation of the workload and Virtual Machines to Data Centres, where ICT resources and/or cooling facilities are more energy/cost effective (e.g., they employ solar energy during daytime across different time zones, have more efficient cooling system due to climate conditions, etc), without compromising the established SLA.
3) Smart-Grid Energy stabilisation
DOLFIN data centers will be inter-connected with the smart grid network, providing responses on the changing demands for energy. In this respect, DOLFIN will contribute in energy stabilisation of the power distribution system. More particularly, DOLFIN will respond to the smart grid demands by lowering its power consumption, within the limits of the Data Centre infrastructure, provided that SLAs are still honoured, or by transferring workload to synergetic Data Centres, which are located in areas where the power distribution system is not overloaded and have available ICT resources.
4) Heat absorption
A large portion of the energy consumed by a Data Centre, especially when no or limited energy management strategies are applied, is simply wasted as heat. Additional energy is required and consumed through the air-conditioning system, so as to equalise the heat. Considering that water is a much better coolant that the air, DOLFIN is targeting to deliver a prototype cooling system for servers that is based on the utilisation of cold water.
5) Energy-aware SLA modelling and negotiation
DOLFIN will research and propose business models and a technical framework for online SLA negotiation, taking into account not only the dimension of the quality of offered services, but also the energy expected to be consumed, while providing a certain level of service quality, together with the respective cost.
6) Influential contribution to standard and industry bodies on energy consumption measurements, revealing the energy and environmental footprint of a Data Centre
DOLFIN participants will collaborate with the following bodies:
• IETF Energy Management (EMAN) Working Group
• ETSI Technical Committee Environmental Engineering (TC-EE)
• EU Code of Conduct for DCs
• ITU-T Future Networks
• ITU-T SG5 Group
To this end, the project will provide a clear list of actions that can contribute in reducing the energy footprint of a Data Centre:
• technical and administrative requirements for incorporating low Energy consumption model proposed by DOLFIN
• implementation cost considering technical and human limitations
• energy and carbon dioxide emission savings