Green software engineering: the curse of methodology

Abstract

Computer Science often seems distant from its natural science cousins, especially software engineering which feels closer to sociology and psychology than to physics. Physical measurements are often rare in software engineering, except in a few niches. One such important niche is that of software energy consumption, green mining, green IT, and sustainable computing, which all fall under the umbrella of green software engineering.With the physical measurement of energy consumption comes all of the limitations of measurement and experimentation that exist in the natural sciences and engineering.Issues abound, from attribution of energy use, isolation of components, to replicable experiments. These get further complicated by cloud computing whereby systems are virtualized and attribution of resource usage is a serious issue. Thus in this work we discuss the current state of software energy consumption, and where will it go. Abstract-Computer Science often seems distant from its natural science cousins, especially software engineering which feels closer to sociology and psychology than to physics. Physical measurements are often rare in software engineering, except in a few niches. One such important niche is that of software energy consumption, green mining, green IT, and sustainable computing, which all fall under the umbrella of green software engineering.With the physical measurement of energy consumption comes all of the limitations of measurement and experimentation that exist in the natural sciences and engineering. Issues abound, from attribution of energy use, isolation of components, to replicable experiments. These get further complicated by cloud computing whereby systems are virtualized and attribution of resource usage is a serious issue.Thus in this work we discuss the current state of software energy consumption, and where will it go.

A Survey of App Store Analysis for Software Engineering

“…Several papers included in this section relate to energy usage [139], [236], although much of this field of research relates only indirectly to app stores. For those who wish to learn more on the subject, we point the reader to the recent paper by Hindle [96].…”

Helping Programmers Improve the Energy Efficiency of Source Code

“…In fact, software developments are keen on developing energy-efficient software [2], and a long list of (mostly recent) efforts that include [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15] has tried to provide developers with the libraries, tools, techniques and data to support energy-aware development. Even considering these efforts, the green computing research area is still at an early stage where research issues, challenges and opportunities abound [16], [17], [18]. This paper introduces an in development technique and tool, named SPELL -SPectrum-based Energy Leak Localization, to determine red (energy inefficient) areas in software.…”

Principles of Green Data Mining

“…Green IT discusses institutional perspectives [39], the role of users, including their behavior and beliefs when using IT-based systems [38] as well as technical concerns [1,19,24]. Topics include computational methods [1], their implementation in software [8,21], hardware components of computers [24,44], datacenters [39], cloud computing [18,33], parallel data processing (for big data) [19,22,40], as well as organizational and business aspects such as sustainable value chains, green oriented procurement [7], and adoption of Green IT [28]. Loeser et al [30] discussed constructs and practices from Green IT (and IS) with respect to sourcing, operations, disposal, governance and end products.…”