Skip to content

HADES: Heterogeneity-Aware Dependent Task Offloading in Collaborative Edge Networks

Oct 2026 · IEEE Transactions on Mobile Computing · Vol 25, pp. 15879-15896 · 0 citations · 58 references

Abstract

Regarding the computational intensity and stringent latency requirements of modern applications comprised of many dependent subtasks, collaborative edge computing (CEC) emerges as a solution to guarantee the service-level objectives (SLOs) by offloading subtasks to multiple distributed edge servers. However, existing offloading schemes remain inadequate in addressing the stochastic task arrival and diverse task-resource affinity that stems from computational efficiency differences when executing various tasks on heterogeneous resources, leading to suboptimal system performance. Moreover, due to continuous updates in both applications and resources (e.g., hardware and runtime environment) in practical environments, comprehensively evaluating task performance profiles across all resource configurations becomes prohibitively expensive. To address these challenges, we propose heterogeneity-aware dependent task offloading scheme (HADES). We first formulate the problem of minimizing the long-term system energy consumption while guaranteeing average application latency. To optimize task offloading policy without relying on prior knowledge of task-resource affinity, HADES adopts a novel generative diffusion model-based graph reinforcement learning approach. Specifically, HADES employs graph neural networks to handle the graph-structured states of both the edge network and dependent applications, and leverages diffusion model-based policy representation for efficient policy exploration in a large action space. In addition, a Sortino ratio-based latency reward is designed for HADES to achieve risk-aware latency optimization. Extensive evaluations on real-world datasets demonstrate that HADES can reduce the latency and energy consumption by at most 98.1% and 86.9% compared to existing baselines.

View source

Similar papers

Open access Aug 2026

CELLServe: An SLO-Aware and Cost Efficient LLMs Serving System for Serverless Computing Environments

CELLServe formalizes SLO-constrained joint resource provisioning as an optimization problem with a dedicated algorithm, and introduces an opportunistic instance merging strategy for decode phase functions to reclaim fragmented resources.

Ze-Jian Wang, Nan Lin, Zi-Nuo Cai et al. · 0 citations
Book Open access Sep 2026

QMScaler: A QoS-Constrained Resource-Efficient Microservice Autoscaling Framework for Edge Environments

QMScaler is proposed, a QoS-driven microservice horizontal scaling framework based on Monte Carlo Tree Search (MCTS) that aims to minimize the number of container instances while meeting the QoS requirements of multiple application functions.

Tian-Yang Zheng, Peng-Fei Yang, Zhe Xu et al. · 0 citations
Open access Aug 2026

Real-Time Task Offloading with Replication Under Node Churn in Volunteer Edge Computing

This article proposes Volunteer Edge, a cost-effective real-time task offloading framework that exploits underutilized computing resources of privately managed nodes to execute offloaded workloads and reduces edge rental costs by 54.0% on average compared with conventional public-edge-based offloading while maintaining...

Jih-Yun Lee, Gahyeon Kwon, Hyokyung Bahn · 0 citations
Preprint Aug 2026

Beyond the Limits: Flexible and Congestion-Aware Cluster Scheduling for the Cloud

The results show that soft SLO limits reduce corrective rescheduling actions by 49% compared to hard-limit approaches while maintaining acceptable performance guarantees, and resource-aware scheduling decreases node-level congestion and further mitigates SLO violations, demonstrating the effectiveness of incorporating...

Oliver Larsson, Thijs Metsch, Cristian Klein et al. · 0 citations
Conference Aug 2026

MADRL-Based Multiobjective Joint Task Offloading and Resource Allocation in Heterogeneous IIoT

In Industrial Internet of Things systems, heterogeneous devices generate tasks with diversified quality-of-service (QoS), including latency-sensitive tasks and energy-sensitive tasks. However, most existing task offloading approaches optimize a single performance metric or convert multi-objective problem into weighted...

Xianhui Wang, Zi-Shuo You, Qiang Li et al. · 0 citations
2026

Workflow-Aware Expert Routing for Distributed LLM Serving Over the Edge-Cloud Continuum

Deploying Large Language Models (LLMs) over the edge-cloud continuum faces severe stability challenges due to the conflict between stochastic network topology and complex workflow dependencies. Existing schedulers, relying either on computationally prohibitive Graph Neural Networks (GNNs) or topology-agnostic heuristic...

Yan Gao, Shaoyuan Huang, Yonghui Ye et al. · 0 citations

Related blog posts

Microsoft Research Blog Jul 13, 2026

Verifying Rust cryptography in SymCrypt, from standards to code

Cryptographic code supports vital protections in modern computing systems. Learn how a new method helps verify code as developers write it while preserving speed and adaptability as it gets implemented and evolves. The post Verifying Rust cryptography in SymCrypt, from standards to code appeared first on Microsoft Research.

Microsoft Research Blog Oct 29, 2026

Microsoft Research Focus – Edition #97

Host Sowbhagya Suresh shares some of the latest news and cutting-edge research from across Microsoft—from understanding the challenges facing coding agents to uncovering how people use AI at work and improving extended-range weather forecasting. The post Microsoft Research Focus – Edition #97 appeared first on Microsoft Research.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.