DSD Special Session

Scope

Modern applications require more and more interconnected systems, merging IoT, Edge, and Cloud technologies into a unified computing paradigm called the IoT-to-Edge-to-Cloud continuum. Many application domains can benefit from the adoption of this paradigm, ranging from automotive (including autonomous driving) to intelligent industrial systems, smart manufacturing, smart cities, robotics, smart farming, and more.

These application domains often require the deployment of complex cyber-physical systems that need to accurately sense the environment, perform convoluted processing, and actuate control decisions. These activities are very likely to occur on different processing nodes. For example, some advanced computations may rely on heavyweight algorithms based on artificial intelligence, which cannot be entirely handled on embedded platforms, and Edge computing in general, due to resource constraints. Furthermore, modern applications are expected to be characterized by the presence of highly heterogeneous nodes that exchange massive amounts of data across the continuum, and inevitably end up needing to comply with non-functional requirements such as privacy preservation, security, safety, energy consumption, and real-time computation.

The IoT-to-Edge-to-Cloud continuum poses new challenges that need to be properly addressed with ad-hoc hardware and software technologies, as well as tools allowing for designing and managing such massively distributed cyber-physical systems.

Topics

Papers on any of the following and related topics can be submitted to the special
session:

  • Hardware technologies for ultra-low latency communication (5G and more);
  • Hardware-software codesign techniques for the compute continuum;
  • Hardware acceleration technologies for distributed offloading;
  • Orchestrators and tools for managing the IoT-to-Edge-to-Cloud continuum;
  • Middleware framework extensions (DDS, MQTT, ROS, CyberRT,…) to cope
    with context-specific non-functional requirements;
  • Mechanisms for the interoperability between different communication
    standards in heterogeneous distributed systems;
  • Tool for optimizing the deployment of largely-distributed applications;
  • Tool for analyzing performance criteria related to non-functional requirements
    in the compute continuum;
  • Energy-aware hardware and software to foster a greener compute continuum;
  • Safety, security, and real-time performance in distributed applications;
  • Virtualization technologies;
  • Hardware and software technologies for implementing distributed AI
    applications on the compute continuum;
  • AI techniques for managing distributed applications.

Special Session Chairs

Technical Program Committee

TBD