5th International Workshop on Green and Sustainable Software
– May 16th, 2016 –
Austin, Texas, USA
Workshop in conjunction with ICSE 2016
– Keynote –
Engineering Future Transactive Energy Systems: Challenges and Directions
Hong-Mei Chen, University of Hawaii, USA
Abstract: The sea change in the Electric Power industry is posing many new challenges to green software engineering. The disruptive developments in smart grid distributed energy generation and distribution, the rapid growth and viability of renewable energies, the rise of “energy communities”, the proliferation of demand response-enabled smart appliances/ devices and the threat from new battery technologies (e.g. graphene) all contribute to force energy utilities out of their “natural monopoly” status. The future of utilities is now seen to be in Transactive Energy: market-based transaction-oriented exchanges between energy producers, prosumer and consumers, with economic and control mechanisms that allow the dynamic balance of supply and demand across the entire electrical infrastructure using value as a key operational parameter. Existing utility control systems are not able to manage the physical infrastructure being added to the grid (e.g., solar panels, wind turbines, customer-owned microgrid systems, smart devices, etc.), let alone dictate realtime market exchanges. To survive, the utilities must change their business models and rethink their role in the value proposition, moving from an electricity supplier (i.e., a goods-dominant perspective) to a smart service provider in the new Transactive Energy ecosystem (i.e., a service-dominant perspective). This talk will provide a state of the art overview of the challenges, paradigm shifts required, and future directions for engineering for Transitive Energy systems. A promising Eco-architecture approach will be discussed based on an empirical case study with a large IT service consulting firm.
(To download the PDF version of the program click here)
|9:00 – 10:30||Session 1: Welcome
Welcome to GREENS 2016
Hausi Müller, University of Victoria, CanadaSession 2: Keynote
Engineering Future Transactive Energy Systems:
Challenges and Directions
Hong-Mei Chen, University of Hawaii, USA
|10:30 – 11:00||Nutrition Break|
|11:00 – 11:45||Session 3:
What do programmers know about software energy consumption?
Candy Pang, Abram Hindle, University of Alberta; Bram Adams, École Polytechnique de Montréal; Ahmed Hassan, Queen’s University, CanadaAbstract: Traditionally, programmers have received a wide range of training on programming languages and methodologies, but rarely about software energy consumption. Yet, the popularity of mobile devices and cloud computing require increased awareness about software energy consumption. On a mobile device, computation is often limited by the battery life. Under the demands of cloud computing, data centers struggle to reduce energy consumption through virtualization and data center infrastructure management (DCIM) systems. Efficient energy consumption of software is increasingly becoming an important non-functional requirement for programmers. However, are programmers knowledgeable enough about software energy consumption? Do programmers base their implementation decision on popular beliefs? In this article, we survey over 100 programmers for their knowledge of software energy consumption. We find that programmers have limited knowledge about energy efficiency, lack the knowledge about the best practice to reduce energy consumption of software, and are often unsure about how software consumes energy. Education about the importance of energy effective software will benefit the programmers. Our results highlight the need for training about energy consumption and efficiency.
|11:45 – 12:30||Session 4:
Green Software Design: Challenges and Opportunities
Ziliang Zong, Texas State University, USAAbstract: In the past decades, substantial efforts have been made in improving hardware energy efficiency. Software developers benefit automatically without improving code energy efficiency. This trend will change as hardware is now approaching its physical limitation on further improving energy efficiency due to the transistor density wall, the heat wall, and the voltage scaling wall. The excessive power consumption of data centers and the ubiquitous usage of battery-driven devices require innovative research and creative practices in improving software energy efficiency, which brings great opportunities for promoting green software design. On the other hand, green software design faces enormous challenges due to the lack of 1) standards in software industry on designing green software and evaluating software energy efficiency; 2) government-recognized programs to incentivize green software products; 3) in-depth research and mature theories on improving software energy efficiency; 4) easy-to-use infrastructures/tools that allow software developers write green code on their daily programming practices; 5) lack of education and training at the college level and post-degree level. We recently interviewed over 100 people (from government, industry, and academia) on the challenges and opportunities green software is facing. This talk will present our findings, discuss the current ecosystem of green software engineering, and demonstrates our recent practices on building the cloud-based Greensoft system that supports green software research and education.
|12:30 – 14:00||Lunch|
|14:00 – 14:30||Session 5:
Lightweight Measurement and Estimation of Mobile Ad Energy Consumption
Jiaping Gui, Ding Li, Mian Wan and William Halfond
University of Southern California, USAAbstract: Mobile ads are an important component of the app ecosystem. Typically, developers use ads to generate revenue and, in return, end users get a “free” app. However, recent work has shown that apps with ads actually have significant hidden costs to end users in terms of energy, network usage, and performance. These can affect the ratings and reviews an app receives. Therefore, it is desirable for developers to balance the usage of ads with these potential negative costs. However, for energy developers lack techniques to help them measure the cost to their apps. To address this problem, we propose and evaluate several lightweight statistical approaches for measuring and predicting ad related energy consumption. We evaluate our approaches on real-world market apps and find that they are able to accurately and quickly estimate the energy cost without requiring expensive infrastructure or extensive developer effort.
|14:30 – 15:00||Session 6:
Adaptive Virtual Machine Migration Mechanism for Energy Efficiency
Sahar Sohrabi, Antony Tang, Irene Moser, Swinburne University of Technology, Australia and Aldeida Aleti, Monash University, AustraliaAbstract: Cloud systems have become a popular platform for business applications due to the flexibility in usage and payment they offer. One of the caveats of Cloud systems is their high energy consumption. Minimizing energy consumption while maintaining a high service level has become a relevant optimization task for Cloud providers. Opportunities for energy savings arise when server hosts are overloaded, which also entails unnecessary delays. To address the problem, researchers have devised strategies how to choose the server host to deploy an application to and how to choose a running application for migration when a host has been identified as overloaded. In this work, we introduce a Bayesian Belief Network which learns over time which of the virtual machines are best removed from a host that has been identified as overloaded. The probabilistic choice is made among virtual machines that are grouped by their degree of CPU usage. Given the feedback in the form of the computing resources saved, the system learns which virtual machine profiles should be shifted for best performance. This strategy compares favourably to two existing methods for load balancing.
|15:00 – 15:30||Session 7:
The Influence of the Java Collection Framework on Overall Energy Consumption
Rui Pereira, Marco Couto, Jácome Cunha, Joao Fernandes and João Saraiva, HASLab/INESC TEC, Universidade do Minho, NOVA LINCS, DI, FCT, Universidade NOVA de Lisboa, RELEASE, Universidade da Beira Interior, PortugalAbstract: This paper presents a detailed study of the energy consumption of the different Java Collection Framework (JFC) implementations. For each method of an implementation in this framework, we present its energy consumption when handling different amounts of data. Knowing the greenest methods for each implementation, we present an energy optimization approach for Java programs: based on calls to JFC methods in the source code of a program, we select the greenest implementation. Finally, we present preliminary results of optimizing a set of Java programs where we obtained 6.2% energy savings.
|15:30 – 16:00||Nutrition Break|
|16:00 – 17:00||Session 8: Breakout Groups|
|17:00 – 17:30||Session 9: Reports from Breakout Groups|
Motivation and Goals
Engineering green software-intensive systems is critical in our drive towards a sustainable, smarter planet. The goal of green software engineering is to apply green principles to the design and operation of software-intensive systems. Green and self-greening software systems have tremendous potential to decrease energy consumption. Moreover, enterprise software can and should be re-thought to address sustainability issues using innovative business models, processes, and incentives. Monitoring and measuring the greenness of software is critical towards the notion of sustainable and green software. Demonstrating improvement is paramount for users to achieve and affect change. The GREENS 2016 theme is The Software Sustainability Challenge. Analysis of the sustainability of a specific software system requires software developers weigh four major dimensions of sustainability—economic, social, environmental, and technical—affecting their related trade-offs. The software engineering community must assume leadership in this important challenge.
The GREENS workshop series brings together researchers and practitioners to discuss both the state-of-the-art and state-of-the-practice in green software, including novel ideas, research challenges, methods, experiences, and tools to support the engineering of sustainable and energy efficient software systems. GREENS 2016 encourages contributions from industry, government, and academia on all topics related to greener software engineering. Topics include, but are not limited to:
- Requirements and design methods for green software
- Best practices to increase energy efficiency and sustainability (including software and process improvement)
- Instrument and monitor software systems to key green indicators (KGIs) and green improvement
- Energy-aware adaptation of software-intensive systems
- Energy challenges and solutions in cyber physical systems
- Energy efficient IoT and sensor networks
- Self-adaptive and self-managing systems for green computing
- Green architectural knowledge, green design patterns
- Sustainable data management
- Monitoring, verification and validation of green software
- Creating user awareness about energy consumption
- Analytics tools for green decision making
- Green key performance indicators
- Quality & risk assessments, tradeoff analyses between energy efficiency, sustainability and traditional quality requirements
- Business models for green software (e.g., SaaS, IaaS, PaaS, and cloud computing)
- Formulating challenges for a green software industry
- Return on investments and economic aspects of green software development
- Case studies and industry experience reports
- Incentives to invest in greener software
- Patricia Lago (VU University Amsterdam, The Netherlands)
- Maurizio Morisio (Politecnico di Torino, Italy)
- Hausi A. Müller (University of Victoria, Canada)
- Giuseppe Scanniello (Università della Basilicata, Italy)
- Giuseppe Procaccianti (VU University Amsterdam, The Netherlands)
- Sedef Akinli Kocak (Ryerson University, Canada)
- Luca Ardito (Politecnico di Torino, Italy)
- Ayse Bener (Ryerson University, Canada)
- Coral Calero (Universidad de Castilla-La Mancha, Spain)
- Ivica Crnkovic (Chalmers University of Technology, Sweden)
- Marios Fokaefs (York University, Canada)
- Sudhukar Ganti (University of Victoria, Canada)
- William G.J. Halfond (University of Southern California, USA)
- Abram Hindle (University of Alberta, Canada)
- Zhenjiang Hu (National Institute of Informatics, Japan)
- Thomas Ledoux (Ecole des Mines de Nantes, France)
- Grace Lewis (Carnegie Mellon, Software Engineering Institute, USA)
- Marin Litoiu (York University, Canada)
- Tommi Mikkonen (Tampere University of Technology, Finland)
- Stefan Naumann (Trier University of Applied Sciences, Germany)
- Birgit Penzenstadler (California State University Long Beach, USA)
- Filippo Seracini (Microsoft, USA)
- Gabby Silberman (Dell Research, USA)
- Kenji Tei (National Institute of Informatics, Japan)
- Marco Torchiano (Politecnico di Torino, Italy)
- Colin Venters (University of Huddersfield, United Kingdom)
- Norha M. Villegas (Icesi University, Colombia)
Workshop papers must follow the ICSE 2016 Format and Submission Guidelines and must have a maximum length of seven (7) pages. Accepted papers will be published as an ICSE 2016 Workshop Proceedings in both the ACM and IEEE Digital Libraries.
The official publication date of the workshop proceedings is the date the proceedings are made available in the ACM Digital Library. This date may be up to two weeks prior to the first day of ICSE 2016. The official publication date affects the deadline for any patent filings related to published work.
Submissions may be directed through here.