Green Coding II

Green Coding - Session II

Measuring Resource Efficiency of LaTeX Paper Templates
Scientific work is mostly communicated via scientific papers, which are often published in journals or conference proceedings, either in print or digital form. These journals and conferences usually demand that submitted papers follow a specific formatting style, for which they provide style templates. The choice of a template influences different properties of the generated document, like its file size or the number of pages that it would use in printed form, directly affecting its impact on the environment. We built a system to automatically compare different LaTeX templates with regard to different factors relevant to the environmental impact. We test our approach with seven templates used by different conferences and journals, and find that the most efficient templates have roughly one third of the file size, and require about one half of the resources for paper production of the least efficient.
Energy and resource comparison of current applications with a focus on statistical analyses and evaluations using the example of Matlab and R/RStudio
This paper presents a comprehensive comparison of the energy and resource efficiency between MATLAB and R, two widely used programming languages in scientific computing and data analysis. The study utilizes a system under test, comprising a load driver and an automation software, PowerBI, to measure and evaluate the performance of both languages. Prior to the experiment, specific mathematical operations and execution methods were implemented in MATLAB and R scripts. The measurement steps and evaluation were conducted using the Oscar framework. The findings indicate that R outperforms MATLAB in baseline and statistical operations, while MATLAB excels in matrix calculations. These results provide valuable insights into selecting the most suitable programming language based on specific computational requirements, thereby optimizing energy consumption and resource utilization.
Power Consumption of Common Symmetric Encryption Algorithms on Low-Cost Microchips
In the Internet of Things (IoT), many devices are battery-operated, making them particularly susceptible to power-hungry applications. Symmetric encryption is a regularly performed task on such devices, as it ensures the confidentiality of the data they send. While previous work has compared the power consumption of common symmetric encryption algorithms on commodity hardware, no such evaluation exists for low-cost microchips, which are often used in IoT devices. In this paper, we compare the power consumption of an ESP8266 executing common symmetric encryption algorithms with varying parameters such as key size, data authentication, or payload size. We find that the power consumption depends on several factors, but that overall AES-GCM has the lowest power consumption when the encrypted data is also authenticated, while Blowfish-CTR has the lowest power consumption when no authentication is applied.
Measuring the energy consumption of software with simple tools
In my talk, I would like to present a simple and low-threshold approach to measuring the energy consumption of software on local development environments. With the help of a measurement script executed in parallel to one’s own application, which is documented at, it is easy to examine the energy efficiency of one’s own code in any programming language or complete applications. The measurement results can then be used to make the code more energy-efficient. I present the results of a test series in which the “softwarefootprint” measurement script was executed on different hardware platforms with different load drivers and show its possible areas of application and limitations.