Paper on Multi-Application Energy Analysis in Edge Computing Accepted at FMEC 2024

Good news everyone! Our paper “Analysing Per-Application Energy Consumption in a Multi-Application Computing Continuum” was accepted at the 9th International Conference on Fog and Mobile Edge Computing (FMEC 2024). This paper was first-authored by Saeedeh Baneshi, a PhD student at the University of Amsterdam, and complements her earlier work “Estimating the Energy Consumption of Applications in the Computing Continuum with iFogSim“. Congratulations on another accepted paper Saeedeh!

The paper addresses the challenge of analyzing the energy consumption of applications distributed over edge devices and data centers in the compute continuum. The goal is to enable stakeholders, such as cloud providers, developers, users, and researchers, to improve energy efficiency, optimize resource usage, and reduce the environmental impact of such applications. To this end, the work proposes a fine-grained simulation approach for analyzing application energy behavior in edge/cloud environments, based on the iFogSim framework. The three main contributions of the work are: 1) An extension is proposed to iFogSim’s energy model to also consider the energy consumption of communication, 2) iFogSim’s reporting is improved to collect finer-grained data, an essential improvement for analysis of multi-application scenarios, and 3) The effectiveness of the approach is demonstrated by evaluating different multi-application scenarios and configurations for a distributed video surveillance application.

Call for Papers: Special Issue on Model-Driven System-Performance Engineering for CPS

I’m honored to serve as Guest Editor for a special issue of IET Cyber-Physical Systems: Theory and Applications focused on Model-Driven System-Performance Engineering for CPS. This issue is a collaboration with Twan Basten (Eindhoven University of Technology), Arvind Easwaran (Nanyang Technological University), and Marilyn Wolf (University of Nebraska-Lincoln).

We invite submissions from both academia and industry across various application domains. If you’re working in this area, consider contributing your research! The submission deadline is November 1, 2024. Feel free to reach out if you have any questions!


Model-Driven System-Performance Engineering for CPS

Submission deadline: Friday, 1 November 2024
Expected Publication Month: June 2025

System performance refers to the amount of useful work done by a system within predefined quality constraints. System performance often brings the competitive advantage for cyber-physical systems in domains like autonomous driving, chip manufacturing and production systems in general, healthcare, the smart grid, precision agriculture, and so on. To meet market demands for product and system quality, system customization, and a low total cost of ownership, systems need to meet ever more ambitious targets relating to system performance. Performance is a cross-cutting system-level concern, with intricate relations to other system-level concerns like quality, cost, energy efficiency, security, reliability, and customizability. Model-driven system-performance engineering (MD-SysPE) for CPS is essential to improve time-to-quality and the cost-performance ratio of complex systems.

This special issue invites any contributions in model-driven system-performance engineering for CPS that are of interest to the academic and industrial CPS community at large. Original research papers, industrial applications and case studies, and surveys on relevant topics are welcome.

Topics for this call for papers include but are not restricted to:

  • Multi-domain modelling, analysis, and optimization of performance aspects
  • Performance views in system architecture
  • Modelling and analysis of trade-offs with other system qualities
  • Modelling and analysis across abstraction levels
  • Design-space exploration methods
  • Synthesis methods targeting performance
  • Scheduling, control in relation to performance
  • Time-predictable (software) execution
  • Data-driven performance analysis and optimization
  • AI methods for performance analysis, optimization, diagnostics
  • Performance monitoring
  • Run-time adaptation and optimization
  • Performance debugging and diagnostics
  • Model learning for performance
  • Performance validation, verification, and testing