Given the growing demand for real-time gathering and processing of vast amounts of data from data-intensive services such as autonomous driving, digital industry applications, environmental monitoring, and augmented reality (AR), we are witnessing a visible trend to push computing and data analytics closer to the edge of networks for benefits in low latency, reliability, throughput, security, and privacy. Supported by lightweight virtualization technologies such as Docker Containers, Unikernels, and Serverless Functions, the decentralized edge computing paradigm aims to offer efficient access to various geographically distributed computing resources. In this context, the EdgeSys workshop focuses on the confluence of edge computing, decentralized communication, distributed computing, and edge AI. For example, there are several industry-led edge computing and orchestration platforms such as KubeEdge, Firecracker, and Azure Sphere. In the decentralized communication space, systems such as Mastodon and Matrix have been developed. There are also numerous academic research activities on mobile data and computation offloading (e.g., Cloudlet, Talket, ThinkAir, MAUI, and MADNet), as well as systems that combine decentralization and computation/storage offloading, such as IPFS. In addition, as data analytics and machine learning are increasingly offered as a service, the next phase of evolution is to extend the offerings beyond basic algorithms and push analytics and ML models closer to the edge, where the data from users and devices is first captured. In this regard, edge computing has the potential to enable a new class of privacy-preserving and communication-efficient data analytics applications and services. We expect such edge-driven data analytics and edge AI will emerge and reshape the existing cloud-based data processing pipelines. The knowledge from this new edge pipeline can further power future cyber-physical and network services, such as cognitive assistance and proactive accident prevention for autonomous driving.

The 9th Workshop on Edge Systems, Analytics and Networking (EdgeSys'26) aims to bring together system researchers, data scientists, engineers, and practitioners to identify open directions and discuss the latest research ideas and results on edge systems, analytics, and networking, especially those related to novel and emerging technologies and use cases. EdgeSys focuses on systems, analytics, and networking aspects, covering system architecture, distributed ML algorithms, decentralized networking, distributed consensus and ledger techniques, edge services, and data analysis.

The topics include, but are not limited to, the following:

System Architecture for Edge Computing
Serverless, In-network Computing
Edge Storage Systems
Distributed ML for Edge Analytics and Services
Edge-driven Data Analytics
Federated Learning at the Edge
Lightweight Virtualization for Edge Computing
Distributed Consensus Algorithms and Ledgers
Infrastructure and Toolkits for Edge AI
Trustworthy Edge AI
Sustainable Edge Computing
Energy-driven Edge Computing
IoT Networking and Communications
Sensor Networks and Processing
(Autonomous) Management for Edge Systems
System Performance and Measurement
Security and Trust Management
Reliability and Fault Tolerance
Privacy-Preserving Analytics
Data Provenance for Distributed Processing
Heterogeneous & Scalable Orchestration
Edge Architectures for 5G and Beyond

Important Dates

Submission deadline: 15 March 2026
Acceptance notification: 23 April 2026
Camera ready: 30 April 2026
Workshop date: 25 June 2026

Submission Guidelines

Authors are encouraged to submit papers describing original and unpublished research, not currently under review in other venues. In particular, innovative, early-stage ideas and preliminary results are welcome. Papers must be submitted electronically as PDF files, formatted for 8.5x11-inch paper. The length of the paper must be no more than 6 pages in the ACM double-column format (10-pt font), excluding references and appendices. Authors are suggested to use the ACM Master article template with the \documentclass[sigconf,10pt]{acmart} option. The reviews will be single blind. The first page must contain an abstract, the name(s) and affiliation(s) of the author(s), as well as the corresponding contact information. Each submission will receive independent, blind peer reviews from the program committee. Accepted papers will be published in the ACM Digital Library. At least one author of each accepted paper must register and present the paper at the workshop. The program committee will elect one paper for the Best Paper Award.

Submission: https://edgesys26.hotcrp.com/