ProductAI Software / Framework

ScaleAcross Explorer

Optimizer developed by Meta and Harvard researchers that searches over parallelism placement, scheduling, and network design for cross-building AI training.

Made by Meta

Coverage

Aug 20, 2026
AIAnalysis
Gigawatt Supercomputers Span Buildings. Scale-Across Networking Has a Benchmark. Heterogeneity Doesn't.

NVIDIA has priced training across 1,000 km; Google trains synchronously across data centers. Missing: what one job pays to span mixed silicon and fabrics.

Aug 20, 2026 · By SCN Staff · 9 min read