What problem does it solve?
This Skill eliminates the guesswork of designing ClickHouse systems by providing workload-specific architecture guidance instead of generic one-size-fits-all rules, ensuring your ingestion, partitioning, and query patterns are optimized for your exact use case.
Core Features & Use Cases
- Workload Classification: Automatically categorizes your use case (observability, SIEM, financial services, IoT, etc.) to tailor recommendations.
- Decision Frameworks: Provides structured guidance for ingestion strategy, time-series partitioning, enrichment path selection, late-arriving data handling, and real-time pre-aggregation.
- Provenance-Labeled Recommendations: Every recommendation is explicitly marked as official (backed by ClickHouse documentation), derived (logically reasoned from official docs), or field (experience-based heuristics) to ensure transparency.
- Use Case Example: Use this Skill to design a high-throughput observability ClickHouse architecture that handles 300K events per second with sub-second freshness, including Kafka-based decoupled ingestion and incremental materialized views for dashboard rollups.
Quick Start
Use the clickhouse-architecture-advisor skill to design a ClickHouse architecture for your real-time market surveillance workload that handles 80 million daily order events with support for late-arriving corrections and cancellations.
