Paper argues nested von Neumann architecture can make a million processors act as one computer
bronzeagepapi · x · 2026-10-10
A new arXiv paper by Heng Liao, Many Processors, Still One Computer, argues the real contest in large-scale AI computing is whether a million processors can behave as a single computer, not whether one chip gets stronger. It proposes two extensions: recursively nested BSP as an execution plan holding at every scale (parallel work, then exchange/aggregation, then a barrier, with peer-equal units at each layer), and the Nested Parallel von Neumann Architecture — a nest of peer-equal layers from chip package to autonomous zone joined by a memory-semantic interconnect, the Unified Bus. A τ Scaling law folds time at every layer. The paper analyzes workloads the structure serves (AI training above all, plus much of classic HPC) and the hardware decisions that make the nesting physical.
More from Infra
- Supabase outpaced Vercel in site visits for 12 straight months as AI builder layer doubles — FinanceYF5 · 2026-10-10
- Google AI Pro plans bundle Colab A100 80GB access, enabling full fine-tunes of Gemma up to 31B — algo_diver · 2026-10-10
- 'CUDA cores' are a marketing scam since Kepler: SM count is the only metric that matters — blelbach · 2026-10-10
- Wasted Highway Cloverleaf Land Could Host Data Centers, Speculates Viral Thread — gregmushen · 2026-10-10
- Industrially, What You Do More Of Gets Cheaper: The Learning Curve Behind Data Center Costs — Afinetheorem · 2026-10-10
- SGLang on NVIDIA Vera Rubin: up to 20% faster Kimi K3 inference, 4.8x gains for Cognition — ying11231 · 2026-10-10