COHERENT 2024
The 1st International Workshop on the Orchestration of the Serverless Edge-Cloud Continuum
July 31, 2024, Big Island, Hawaii, USA
Papers Submission Link
We are working on organising a Special Issue for the Journal of Grid Computing, focusing on the topics covered by the workshop. Selected authors will be invited to submit extended versions of their work for consideration.
COHERENT is a workshop focussed on the innovative application of the Serverless Function as a Service model to the Edge-Cloud continuum. It addresses various critical challenges, encouraging research into novel architectural designs for resource-constrained Edge environments, efficient Edge-Cloud distributed scheduling and resource allocation strategies.
The workshop will be an opportunity to bring together researchers working on topics such as Edge/Fog Computing, the Internet of Things, Virtualised Infrastructure Systems and Architectures and Security and Privacy in Communications, Networks, and Testbeds from the ICCCN conference where it will hosted. It will identify new problems, challenges, and potential innovative solutions for extending the Serverless FaaS model to the Edge-Cloud Continuum.
Serverless Function as a Service (FaaS)
The rapid evolution of lightweight virtualisation technologies, including containers, micro VMs, and unikernels, has ushered in a new software development and deployment era. These technologies have facilitated the transition from monolithic software systems to modular solutions built on seamlessly cooperating microservices. Agile software development, continuous integration, and event-driven programming practices have paved the way for Serverless Computing, a paradigm in which third-party providers dynamically allocate resources to enable on-demand execution of computing services.
The Serverless Function as a Service (FaaS) model takes this concept further, allowing developers to write stateless functions that can be executed on-demand within containerised runtime environments. When a function is invoked, the underlying FaaS platform handles resource provisioning, scaling, and operational aspects, billing users only for the actual time their code runs. This approach addresses the limitations of traditional Infrastructure as a Service (IaaS) Clouds, where resources must be provisioned in advance to meet workloads, offering cost savings during idle periods through a pay-per-use model and seamless scalability.
Edge-Cloud Continuum
Simultaneously, the advent of Edge Computing aims to address the demands of mobility support, geo-distribution, location awareness, and low latency, particularly in Internet of Things (IoT) applications. This paradigm shifts computation from centralised data centres to the edge, reducing latency and enabling real-time processing based on collected data. Edge nodes are typically resource-constrained, requiring efficient architectural designs for orchestrating distributed applications. The Edge-Cloud continuum is emerging as a deployment solution, providing near-user computation and access to cloud resources when needed.
However, central resource management and orchestration solutions used in the Cloud may not be suitable for the dispersed nature of the Edge-Cloud continuum. Issues such as central points of failure, unpredictable network connectivity, device volatility, and resource consumption pose challenges.
Scope
Serverless FaaS can address these challenges by saving energy and compute resources, allowing users to share lightweight abstractions that can automatically be deallocated during idleness (i.e., scale-to-zero). Yet, many existing Serverless FaaS frameworks feature central architectures that may not align with geo-distributed environments, making it challenging to harness the potential of distributed computing. We invite researchers, practitioners, and experts in Serverless Computing, Edge Computing, and Cloud Computing to submit papers addressing the integration of Serverless FaaS into the Edge-Cloud continuum.
Important Dates
Paper submission deadline: March 29, 2024 (Hard Deadline)
Notification of acceptance: May 9, 2024
Camera-ready paper due: May 20, 2024 (Hard Deadline)
Workshop date: July 31, 2024
Workshop Co-Chairs
Francesco Tusa, University of Westminster, UK
Stuart Clayman, University College London, UK