Task Scheduling in Geo-Distributed Computing: A Survey

TL;DR

The paper surveys task scheduling in geo-distributed computing, spanning four environments—geo-distributed cloud, edge, cloud-edge, and HPC—focusing on performance, fairness, and fault-tolerance. It classifies scheduling methods into heuristic, AI-based, mathematical, and hybrid approaches, and analyzes per-environment techniques that address data locality, energy and cost, and network dynamics. Key contributions include a comprehensive taxonomy, environment-specific analyses, and identified challenges and future directions to guide researchers and practitioners. The work highlights practical implications for designing scalable, energy-aware, and privacy-conscious schedulers across globally distributed infrastructures.

Abstract

Geo-distributed computing, a paradigm that assigns computational tasks to globally distributed nodes, has emerged as a promising approach in cloud computing, edge computing, cloud-edge computing and supercomputer computing (HPC). It enables low-latency services, ensures data locality, and handles large-scale applications. As global computing capacity and task demands increase rapidly, scheduling tasks for efficient execution in geo-distributed computing systems has become an increasingly critical research challenge. It arises from the inherent characteristics of geographic distribution, including heterogeneous network conditions, region-specific resource pricing, and varying computational capabilities across locations. Researchers have developed diverse task scheduling methods tailored to geo-distributed scenarios, aiming to achieve objectives such as performance enhancement, fairness assurance, and fault-tolerance improvement. This survey provides a comprehensive and systematic review of task scheduling techniques across four major distributed computing environments, with an in-depth analysis of these approaches based on their core scheduling objectives. Through our analysis, we identify key research challenges and outline promising directions for advancing task scheduling in geo-distributed computing.

Figures (7)

• Figure 1: An overview of geo-distributed task scheduling. We categorize scheduling strategies across Geo-Distributed Cloud Computing, Cloud-Edge Computing, Edge Computing, and Geo-Distributed Supercomputer Computing (HPC). In each scheduling infrastructure, we focus on objectives including performance, fault tolerance, and fairness, with scheduling methods including heuristic, AI-based, mathematical, and hybrid techniques.
• Figure 2: Taxonomy of studies on optimizations under geo-distributed cloud computing infrastructure.
• Figure 3: An example of a geo-distributed computing architecture exploiting spatial-temporal diversity. Every geo-distributed region has its own power supply and power price. After tasks are submitted, the scheduler will assign tasks to or offload tasks from certain computing regions considering each region's power supply diversity.
• Figure 4: An example of a geo-distributed computing environment focusing on distributed network architecture. After users submit tasks, the scheduler will assign tasks to different computing nodes according to computing and networking status. The thickness of the lines between routers and switches represents the relative size of the bandwidth. The length of the lines indicates the relative transmission distances.
• Figure 5: An overview of specific edge computing scenarios.