Huawei's Sept 19 event: long-running agents bottleneck on data movement, not FLOPs
krishnan · x · 2026-09-22
A reading of Huawei's September 19 announcement: the more useful part is its diagnosis — long-running agents stress movement between compute, memory, storage, and tools, not just the accelerator.
The mechanism: (1) agents repeatedly cycle between CPU work, NPU inference, memory retrieval, storage, and tool execution; (2) longer trajectories expand KV cache and worsen time to first token; (3) conventional clusters pay communication overhead every time state crosses those boundaries; (4) Huawei's stack uses UnifiedBus, composable SuperPoD APIs, and a ThinkProcess abstraction to cut that friction, plus shared 10,000-NPU-scale resources and a baseline 100 NPU-hour program for developers.
These are Huawei's architecture claims, not independent production results.
More from Companies & People
- AI game jam with Anthropic, Google, OpenAI teams: cash prizes plus $50K continuation grants — chrisfirst · 2026-11-12
- Free AI game jam in SF Nov 14–15 with teams from Anthropic, Google, Meshy, OpenAI — chrisfirst · 2026-11-12
- Epic AI MegaJam finals in SF Nov 14-15: teams of 4 plus unlimited AI agents — chrisfirst · 2026-11-12
- VulnCheck audit: Anthropic's Project Glasswing logged 37,966 findings, only 217 fixed — GaryMarcus · 2026-11-12
- YC-backed Monte launches post-training platform to build proprietary models from your agents' real work — ycombinator · 2026-11-12
- Investor: AI diffusion will be slowest in consumer software — Oura, Strava, Labcorp still have no real AI strategy — MartinGTobias · 2026-11-12