LIWSA-GNN: Graph Neural Network-Based Workflow Scheduling in WorkflowSim
Year: 2026
Type: software
Abstract: This compute capsule implements and evaluates LIWSA-GNN, a Graph Neural Network-driven planning algorithm for scientific workflow scheduling within the WorkflowSim framework. It features a multi-seed training pipeline in Python using PyTorch Geometric and integrates exported model weights into Java-based WorkflowSim/CloudSim simulations across standard workflow benchmarks (Montage, CyberShake, Sipht, Epigenomics, Inspiral).
Source: Code Ocean
Author: Mohammed Alaa Ala'anzy
Institutions: Shandong University, University of Southern Denmark
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7203785195