Performance Evaluation is a leading journal in the area of modeling, measurement, and evaluation of performance aspects of computing and communication systems. As such, it aims to present a balanced and complete view of the entire field of performance evaluation. The journal audience encompasses academics, industrial researchers, and professional figures such as performance engineers, network managers, and computer system designers.

The journal solicits from its audience papers that focus on one or more of the following dimensions:

    Define new performance evaluation tools, including measurement and monitoring tools as well as modeling and analytic techniques
    Provide new insights into the performance of computing and communication systems
    Present novel control and learning methods for performance management
    Introduce new application areas where performance evaluation tools and methods can play an important role

Contributions to the journal often focus on theoretical, methodological, or applied aspects of:

    Resource allocation (e.g., routing and flow control, bandwidth allocation, load-balancing, deployment,...
    System tuning, sizing and optimal configuration
    System reliability and dependability
    Energy efficiency and energy harvesting
    System architecture design and implementation

Application areas of interest include, but are not limited to, the following:

    Computer and communication networks (e.g., software-defined, wireless, cognitive radio, ad-hoc, quantum,...
    Data centers
    Cloud, IoT, edge, and cyber-physical systems
    Social networks, social media, metaverse
    AI-based services
    Big data and storage systems
    Blockchain and distributed ledgers
    Autonomous and self-organizing systems
    Embedded systems

Commonly featured performance evaluation methods and techniques include:

    Stochastic models (e.g., queueing theory, mean-field models, stochastic geometry,...
    Data-driven methods (e.g., AI/machine learning, inference, statistical analysis, ...)
    Scheduling and load balancing theory
    Simulation methods
    Game theory and pricing
    Measurement techniques (e.g. software and hardware monitors) and workload characterization

Note that the above lists are not all inclusive or restrictive and submissions with creative applications of performance evaluation tools outside of those above and/or applications outside of those above are also welcome.

A variety of types of submissions are possible, primarily original work, tutorials & surveys.

Special Issue on Performance in the Edge-to-Cloud Continuum
Submission Date: 2025-03-18

Guest editors:

Daniel Sadoc Menasché, UFRJ, Brazil
Francesco de Pellegrini, University of Avignon, France
Marco Ajmone Marsan, IMDEA Networks Institute, Spain

Special issue information:

Networks are evolving into interconnected computing systems that manage network operations and provide end-user services. These facilities vary in size and distance from users: Clouds are large and distant from end-users, Edges are smaller and closer, and Fogs are small facilities near or within user devices.

The ensemble of the computing facilities within a network has been recently termed the Computing Continuum, or the Edge-to-Cloud Continuum. The emergence of 5G and Beyond 5G (B5G) technologies has significantly advanced this continuum, enabling ultra-low latency, high bandwidth, and reliable connectivity. These advancements enhance the potential for distributed services, allowing seamless deployment of chains of microservices and network functions across diverse computing elements within the continuum.

The availability of a number of alternative computing elements and the possibility to implement services in a distributed fashion, as chains of microservices and network functions possibly residing on different computing elements, raises a number of interesting research questions in which performance plays a key role. Some of the research problems have been investigated for a number of years, most notably the server selection problem (where a system user must decide which computing element is the best choice for her computation, according to some performance metric of interest) and the server placement problem (where a network operator must decide the best locations where to distribute its computing facilities over the network). Other problems are new and not yet well investigated.

This special issue solicits unpublished works on performance measurements, analysis and evaluation of the Edge-to-Cloud Continuum, as well as of the applications running over the available computing facilities.

This special issue is intended for researchers, engineers, and practitioners who study and work on Cloud and Edge Systems, as well as those interested in performance measurement, analysis and modeling in general. We welcome submissions that study performance and resource management in the Edge-to-Cloud Continuum or those that present novel algorithms, techniques, or solutions to improve efficiency and sustainability, noting that sustainability is central to the operation of these systems. While both theoretical and experimental approaches are welcome, and measurements of existing systems are of special interest, attention will be paid in the review process on rigor and quantitative analysis.

Topics of interest for this special issue include, but are not limited to, the following:

Techniques employed:

    Performance modeling, including probabilistic techniques, queueing models, and simulation
    AI/ML for the optimization of the Edge-to-Cloud Continuum
    Measurements of existing Cloud and Edge system performance

Application domains:

    Robustness reliability and availability in the Edge-to-Cloud Continuum
    AI/ML execution over Cloud, Edge and Fog infrastructures
    Integrated inference over Cloud, Edge and Fog
    Edge-based networked games and augmented/extended reality
    Real-time services over Cloud, Edge and Fog infrastructures
    Edge-based video distribution

Specific examples of submissions include, but are not limited to the following research themes in Edge to Cloud Continuum:

Resource Management and Scheduling:

    Server selection and placement algorithms​
    Performance analysis of microservice orchestration frameworks
    Energy-efficient resource management techniques
    Adaptive scheduling for AI/ML tasks across Edge-to-Cloud infrastructures
    Optimal offloading strategies
    Memory-computation-network tradeoffs for edge-based resource allocation

Data and Traffic Management:

    Network-aware routing, data compression and deduplication techniques​
    Network-aware and similarity caching techniques
    Network-aware AI/ML tasks, e.g. early-exit deep neural networks, federated learning and distributed AI
    Age of Information (AoI) in dynamic Edge-to-Cloud systems

Security and Resilience:

    Performance impact of security protocols in Edge-to-Cloud systems
    Fault-tolerance mechanisms for the Edge-to-Cloud Continuum

Emerging Domains:

    OpenRAN and federated Edge-to-Cloud systems​
    Performance of 5G and beyond-5G (B5G) technologies in the Edge-to-Cloud Continuum
    Performance of edge-based augmented and virtual reality applications
    Edge-based blockchain technologies
    Performance of edge-based applications in vertical domains, e.g., healthcare, industry 4.0, smart cities

Submissions focusing on real-world measurements, comprehensive simulations, and/or novel theoretical approaches are particularly encouraged. Researchers are also invited to explore interdisciplinary solutions that span network performance, application design, sustainability, security, and energy efficiency. Note that surveys are not in scope of the special issue and only original full-length papers will be considered.

Manuscript submission information:

Expected timeline:

    Submission deadline: 18 March, 2025
    First reviews: 8 August, 2025
    First revision due: 15 November, 2025
    Second reviews: 15 January, 2026
    Publication: 18 March, 2026