ClickHouse Cloud Rebuilds Autoscaling Orchestration for Near Real-Time Reactivity
mgill25 · x · 2026-08-04
ClickHouse Cloud's engineering team detailed the rebuild of their underlying autoscaling orchestration system.
- Old Pain Point: The previous recommendation service ran on a fixed schedule, unable to scale up immediately during sudden workload spikes.
- New Architecture: The team rebuilt the orchestration layer using controller-runtime, gaining native reactivity, deduplication, backoff, and concurrency control.
- Data-Driven: To feed this machinery with near real-time signals, they used ClickHouse itself as a signals table, reducing fixed cadence to seconds and significantly improving reactive scaling.
More from Infra
- Runware launches modular AI inference pods with 10x lower cost — aziz4ai · 2026-08-04
- Running DeepSeek-V4 1M Context on a Single RTX 5090 with vLLM — BlackBeardAI · 2026-08-04
- MiniMax H3 Acceleration Benchmark: TE-Speed Delivers up to 1.785x Speedup — Commercial_Board9219 · 2026-08-04
- Cloudflare Launches CI SDK with AI Self-Healing Code Fixes — dinasaur_404 · 2026-08-04
- Gavin Baker Reveals SSI to Launch Model in August, Discusses AI Infra & GPU Prices — zephyr_z9 · 2026-08-04
- AI Trade Enters Stock-Picker Phase as Compute Supply Defies Narrative — tengyanAI · 2026-08-04