Networking for AI Factories: Systematic Review Corpus, Evidence Coding, and Selected-Venue Coverage Audit

Description

This dataset accompanies the survey “Networking for AI Factories: A Systematic Survey of Scale-Up, Scale-Out, and Scale-Across Fabrics for Large-Scale AI Training and Inference.” It provides the systematic-review corpus, final eligibility decisions, evidence coding, evaluation metadata, claim-source mappings, and an independent selected-venue coverage audit. The dataset supports reproducibility, secondary analysis, evidence synthesis, and future research on AI networking, AI fabrics, large-scale AI training and inference, and scale-up, scale-out, and scale-across interconnects. A Python validation script is included to independently verify the principal review counts and integrity constraints.

Institutions

• Lancaster University

  England, Lancaster
• King Fahd University of Petroleum and Minerals

  Eastern Province, Dhahran
• Nanjing University of Information Science and Technology

  Jiangsu, Nanjing

Categories

Licence