Energy efficient data centres

Data centres are the core of future industrial and customer services, especially taking into account a fast development of clouds, mobile applications, Internet of Things (IoT) and artificial intelligence (AI) techniques. Consequently data centres energy efficiency and sustainability has becone a major research, business and societal problem. To address this problem the main goal of the E2DC 2020 workshop is to study innovative methods to improve efficiency and sustainability of data centres - important and ever growing source of energy consumption.


8th International Workshop on Energy Efficient Data Centres (E2DC 2020)

The main goal of the E2DC workshop is to study innovative methods to improve efficiency and sustainability of data centres - important and ever growing group of energy consumers. Importantly, E2DC concentrates on a broader context of data centre ecosystem by investigation of interactions with smart cities and smart grids. Hence, the workshop discusses topics such as the use of renewable energy, optimisation of cooling and heat re-use, demand response techniques, federation of distributed data centres and many more. At E2DC 2020, we would like to discuss how emerging AI technologies can help in improvements of DCs efficiency, reliability, and their interaction with external dynamically growing systems such as smart cities, smart grids, heat re-use networks, renewable energy sources, and edge computing systems.


Important Dates

April 1 st , 2020
April 10th, 2020
April 30th , 2020


Paper submission

April 24th , 2020
May 21st , 2020


Notification of acceptance

May 18th , 2020
May 28th , 2020


Submission of camera-ready papers

Call for Papers

The workshop invites original papers of up to 8 pages that were not previously published and are not currently under review for publication elsewhere. Topics include, but are not limited to:

  1. New architectures of the whole data centre ecosystems to cope with dynamic demands and requirements of large data processing:
    • New architectures and management methods for distributed data centres, edge data centres, fog computing
    • Novel approaches for data centric processing (computations closer to data)
    • Integration of data centres into existing infrastructure and facilities (e.g. buildings, wind plants and other novel locations)
    • Data centre integration with smart cities, smart grids and renewable energy sources
  2. Achieving energy efficiency by the innovative use of new hardware and software technologies, e.g.:
    • Microservers, low power CPUs, embedded SoCs, modular DCs
    • Methods to facilitate the use of hardware accelerators at large scale, the use and integration of heterogeneous systems
    • Virtualization and service architectures, containers, lightweight virtualization, microservices
    • Novel cooling approaches, direct liquid cooling, heat re-use, thermal storage
  3. Methods to improve uptake of energy efficient solutions in data centres, incentives that increase focus on energy efficiency and reduction of emissions:
    • New business models and programs to make the use of energy efficient technologies profitable
    • Metrics beyond PUE
    • Green and energy-related SLAs
    • Engaging end users in limiting energy consumption and carbon footprint
    • Other incentives leading to increased adoption of energy efficient technologies in data centres
  4. Applications of AI:
    • Prediction of load in data centers
    • The use of AI techniques for prediction of temperatures and energy consumption of DCs
    • Automated anomaly detection and predictive maintenance
    • Optimization of flexibility offers for demand response programs
    • Multi-criteria selection of configuration for trade-off between performance, efficiency, and gains from interaction with other stakeholders (e.g. smart grids, heat operators, consumers of heat, etc.)


Submission and publication of the work

Important notice!
In light of the Covid-19 outbreak, the organising committee of the ACM e-Energy 2020 conference, in conjunction with the steering committee and ACM has decided to run the entire conference virtually. All the connection details are provided to the registered users (please remember to select E2DC2020 workshop during registration)..

The workshop will be co-located with ACM eEnergy 2020 Please submit your manuscript via submission portal. Manuscripts will be published in the ACM proceedings and must adhere to ACM format. Submission implies the willingness of at least one author to register and present the paper at E2DC. The organizers reserve the right to exclude a paper from publication if it is not presented. For more information please contact Ariel Oleksiak (ariel at man.poznan.pl), Gunnar Schomaker (schomaker at sicp.de)


KEYNOTE SPEAKERS


Title: The energy demand of data centres - current calculations, future trends and selected savings potentials

Speaker: Simon Hinterholzer, Borderstep Institute

Abstract: Within the presentation, some current studies about the energy consumption of data centres are discussed briefly. Our own calculations are presented with a brief introduction into the methodology. In addition, some trends are discussed, which are affecting the need for data centres and the way they are operated. Finally, selected technological approaches are presented to show how the CO2 emissions caused by the energy demand can be reduced.


E2DC WORKSHOP PROGRAM 2020

9:00 - 9:05 CEST, 08:00-08:05 London (BST), 17:00-17:05 Melbourne (AEST)

Introduction

Speakers: Ariel Oleksiak, Gunnar Schomaker

9:05 - 9:45 CEST, 08:05-08:45 London (BST), 17:05-17:45 Melbourne (AEST)

Keynote

Title: The energy demand of data centres - current calculations, future trends and selected savings potentials
Speaker: Simon Hinterholzer, Borderstep Institute

9:45 - 10:00 CEST, 08:45-09:00 London (BST), 17:45-18:00 Melbourne (AEST)

Coffee break

10:00 - 10:25 CEST, 09:00-09:25 London (BST), 18:00-18:25 Melbourne (AEST)

SpinSmart: Exploring Optimal Server Fan Speeds to Improve Overall System Energy Consumption
M. Tian, A. Vishwanath, G. Venkataramani, S. Subramaniam

10:25 - 10:50 CEST, 09:25-09:50 London (BST), 18:25-18:50 Melbourne (AEST)

EDGE: Microgrid Data Center with Mixed Energy Storage
R. Brännvall, M. Siltala, J. Gustafsson, J. Sarkinen, M. Vesterlund, J. Summers

10:50 - 11:15 CEST, 09:50-10:15 London (BST), 18:50-19:15 Melbourne (AEST)

Energy Efficient Scheduling Based on Marginal Cost and Task Grouping in Data Centers
Kaixuan Ji, Ce Chi, Avinab Marahatta, Fa Zhang, Zhiyong Liu

11:15 - 11:40 CEST, 10:15-10:40 London (BST), 19:15-19:40 Melbourne (AEST)

Jointly Optimizing the IT and Cooling Systems for Data Center Energy Efficiency based on Multi-Agent Deep Reinforcement Learning
Ce Chi, Kaixuan Ji, Avinab Marahatta, Penglei Song, Fa Zhang, Zhiyong Liu

11:40 - 12:00 CEST, 10:40-11:00 London (BST), 19:40-20:00 Melbourne (AEST)

Summary and closing
Ariel Oleksiak, Gunnar Schomaker

Workshop Chairs

Ariel Oleksiak


Poznan Supercomputing and Networking Center, Poland

Gunnar Schomaker


Software Innovation Campus Paderborn, Germany

Technical Programme Committee

Hermann de Meer


University of Passau, Germany

Sonja Klingert


University of Mannheim, Germany

Boudewijn Haverkort


Tilburg School of Humanities & Digital Sciences, Netherlands

Evgenia Smirni


College of William and Mary, United States of America

Hongyang Sun


Vanderbilt University, United States of America

Jean-Marc Pierson


University of Toulouse,
France

Xavier Hesselbach-Serra


Universitat Politecnica de Catalunya, Spain

Diego Arnone


Engineering, Italy

John Booth


Green IT Amsterdam, Netherlands

Jon Summers


RISE, Sweden

Tudor Cioara


Technical University Cluj-Napoca, Romania

Georges Da Costa


University of Toulouse, France

Robert Basmadjian


TU Clausthal, Germany

Suresh Subramaniam


The George Washington University, United States of America

Guru Venkataramani


The George Washington University, United States of America

Contact

Ariel Oleksiak (ariel at man.poznan.pl)
Gunnar Schomaker (schomaker at sicp.de)