Project reference №: 734273

Reducing energy consumption via optimized data centre operation
At present, data centres (DCs) are one of the major energy consumers and source of CO2 emissions globally. The GREENDC project (http://www.greendc.eu) addresses this growing challenge by developing and exploiting a novel approach to forecasting energy demands.
The project bring together five leading academic and industrial partners with the overall aim of reducing energy consumption and CO2 emissions in specific national DCs. It implement a total of 163 person-months staff and knowledge exchanges between industry and academic partners. More specifically, knowledge of data centres operations is transferred from industry to academic partners, whereas simulation based optimization for best practice of energy demand control will be transmitted from academia to industry through the knowledge transfer scheme.
In recent years, the use of the information and communications technologies (ICT), comprising communication devices and/or applications, DCs, internet infrastructure, mobile devices, computer and network hardware and software and so on, has increased rapidly. Internet service providers such as Amazon, Google, Yahoo, etc., representing the largest stakeholders in the IT sector, constructed a large number of geographically distributed internet data centers (IDCs) to satisfy the growing demand and providing reliable low latency services to customers.
These IDCs have a large number of servers, large-scale storage units, networking equipment, infrastructure, etc., to distribute power and provide cooling. The number of IDCs owned by the leading IT companies is drastically increasing every year. As a result, the number of servers needed has reached an astonishing number. According to the European Commission’s JRC (Joint Research Centre) report, IDCs is expected to consume more than 100 TWh by 2020.
Due to large amount of energy implied and the related cost, IDCs can make a significant contribution to the energy efficiency by reducing energy consumption and power management of IT ecosystems. This is why most researchers focus on reducing power consumption of IDCs. Those efforts include designing innovative architecture of DCs to minimise loss of cool airs, heats from IT devices, protecting from outside heats and so on. Also, computer scientists developed energy efficient workload algorithms to minimise overloads of servers to minimise energy consumption and heat generation.
However, efforts to reduce energy consumption via efficient DC operation are still scarce. For example, DC managers are yet to find answers to such questions as:
- What are the optimal temperature of the DCs to minimize energy consumption without affecting the performance of IT devices and meeting the service-level-agreement?
- How many servers or virtual machines need to be on for next 24 hours or a week considering expected workloads?
- What are the optimal schedules of servers and VMs for handling workloads which are various time to time?
- Where are the best operational schedules of cooling devices within the DC to have maximum cooling effect?
The GREENDC project tries to find answers to those questions for DC managers via a multi-disciplinary study. It takes a more holistic view by considering the system as a whole i.e. servers, cooling system, backup power and electrical distribution system. Particularly, a Decision Support System (DSS) that integrates functionalities including workload and energy forecasting, generation of optimal operation scheduling of cooling and IT devices and simulation for impact analysis of DC operation strategies.
The architecture of the GREENDC DSS takes layered architecture to guarantee maximum level of independence of components in different layers. This allows the DSS easily be customised for the different requirements of various types of DCs in different regions. There are four layers: data, mathematical model, business logic and user interface layer.
Data layer contains components that collect energy, workload, and meteorological data from target DCs. Collected data is processed by a data normalization component that converts different format of data into the standard format of the GREENDC to be stored in a data warehouse. The normalized data is consumed by upper layer components including math model and business logic layers. Math model layer contains utility components for forecasting and optimization functionalities of the GREENDC DSS and mainly used for the components in the business logic layer.
Business logic layer components provide the main services of the DSS by using the components in lower layers. Those services include monitoring, estimation, optimization, and simulation of energy and workloads data. The components in the business logic layer also respond to the requests from user interface layer to get requests and respond to those requests.
As shown in the gallery below, GREENDC DSS provides a dashboard style user interface to DC managers for easy use of the system and compliant with majority energy management tools.
DSS Screen Shots gallery
The GREENDC project is implemented by five experts in the field:
- Brunel University London (UK)
- Gebze Technical University (Turkey)
- Turksat (Turkey)
- DAVID Holding (Bulgaria)
- LKKE (UK)
In particular, the GREENDC DSS will be tested through two field trials in Turksat and Davind Holdings respectively. Turksat is operating one of the largest data centre in Turkey to provide eGovernment services. It also hosts large number of IT servers for municipalities in Turkey for public service provision. The GREENDC DSS will be tested using the real data obtained from Turksat’s data centre.