Call for Papers

Recent advances in virtualization technologies (e.g., unikernel, containers, VM) and networking technologies (e.g., SDN, NFV, data plane programming) have led to an increased interest in the joint consideration of computation and communication in distributed systems. At the same time, in recent years we have witnessed the proliferation of edge computing, a complementary and at times alternative solution to centralized cloud computing, forming the edge-cloud continuum. Edge computing offers more proximate resources closer to or at the network edge to support emerging applications (e.g., self-driving vehicles, autonomous systems, VR/AR) that require fast response, low delay, high bandwidth, trust-sensitivity, and/or continued operation despite intermittent or complete disconnectivity. These trends have given rise to the concept of Computing in the Network (COIN), the integration of in-network computation and network processing in a common framework. COIN naturally fits within the edge-cloud continuum, where expanded resource distribution and tightly integrated computing-networking capabilities exist from the edge of the network to the back-end cloud infrastructure, as well as at points in between.

The use of both network edge and core resources beyond packet forwarding is often synonymous with taking tasks associated with computation into the network. This enables a more collaborative and better integration of communication and computation resource allocation in the network to respond to diverse application needs. Other than treating the network as a simple end-to-end connector between dependent services, the network can now be considered as an essential constituent of a distributed application: COIN dissolves the boundaries between the networking domain and computing domain. The networking behavior can be dynamically adapted to the required computation task scheduling according to many criteria, including application properties, operator policies, user preferences, network congestion, data availability, location, etc.

We are still at the initial stages of the COIN trend and understanding how the edge-cloud continuum should evolve to accommodate the many challenges to be tackled. Although the joint optimization of communication and computation resource allocation has already been widely investigated, they more or less focus on the high-level abstraction of the resources. When looking deeper into the enabling technologies such as dataplane functionality, unikernel, containers, NFV, SDN, etc., further investigations are needed. For example, an edge-cloud application may consist of a large number of container-based microservices with short lifetimes in milliseconds and also require a set of common shared virtualized network functions (VNFs) with comparatively long lifetimes. This implies the need for both edge and cloud services to collaborate. The microservices and the VNFs coexist in a distributed manner. It is essential to collaboratively manage the communication and computation resources with the consideration of the operation characteristics of NFV orchestration framework (e.g., MANO) and microservice scheduling framework (e.g., Kubernetes). Up to now, the networking domain and computing domain have been developed individually. In the future, joint fine-grain optimization needs to address the mismatch between end-to-end semantics and computing-aware hop-by-hop decision making based on local execution of service-specific functions or telemetry results. It is also required to abstract both the transport and the remote function invocation to provide uniform interfaces for transparent orchestration.

This Special Issue (SI) mainly targets the issues related to emerging directions in COIN, especially in conjunction with edge-cloud. The topics relevant to this special issue include but are not limited to:

• New computing-aware network routing technologies
• Low-latency or delay-sensitive networking technologies for COIN
• Cloud native computing with COIN
• Architecture design toward COIN with edge-cloud continuum
• Crowd computing and mist computing for COIN
• Unikernel, container, VM orchestration for COIN
• Compute-First Networking
• Networking and remote-method-invocation abstraction
• State management and data stewardship for COIN
• Satellite and aerial system for COIN
• Data plane programming and transportation abstraction for COIN
• Security, privacy and trust for COIN in the edge-cloud continuum
• Algorithm design for joint scheduling of communication and computation resources
• Cognitive resource allocation in COIN in the edge-cloud continuum
• Pervasive edge intelligence for COIN
• Interaction between cloud, edge, and 5G/6G networks
• Modeling and performance analysis framework for COIN with the edge-cloud continuum
• System profiling methods for COIN within the edge-cloud continuum
• System telemetry and monitoring in COIN in the edge-cloud continuum
• Load balancing algorithms in COIN for the edge-cloud continuum
• Data-driven COIN and edge/cloud computing management
• ICN/NDN for COIN in the edge-cloud continuum

Submission Guidelines

Manuscripts should conform to the standard format as indicated in the "Information for Authors" section of the Paper Submission Guidelines.

All manuscripts to be considered for publication must be submitted by the deadline through Manuscript Central. Select “September 2021/In-Network Computing: Emerging Trends for the Edge-Cloud Continuum” from the drop-down menu of Topic/Series titles.

Important Dates

Manuscript Submission Deadline: 31 January 2021
Initial Decision: 31 March 2021
Revised Manuscript Due: 30 April 2021
Final Decision: 31 May 2021
Final Manuscript Due: 30 June 2021
Publication Date: September 2021

Guest Editors

Deze Zeng
China University of Geosciences, China

Nirwan Ansari
New Jersey Institute of Technology, USA

Marie-José Montpetit
Concordia University, Canada

Eve M. Schooler
Intel Corporation, USA

Daniele Tarchi
University of Bologna, Italy