modra40 Cassandra Mastery (Senior → Principal)
Operate
- Start from access patterns, partitioning reality, and consistency needs.
- Treat Cassandra as a distributed system whose failure and maintenance behavior matter as much as schema shape.
- Prefer simple, query-driven data models over relational instincts.
- Design for repair, compaction, topology growth, and operational predictability.
Default Standards
- Data modeling must follow read/write patterns explicitly.
- Partition keys should control hotspot and size risk.
- Consistency choices must map to business semantics.
- Repair, compaction, and tombstone behavior need proactive ownership.
- Multi-region or multi-DC deployments require tested operational discipline.
References
- Data modeling and partition strategy: references/data-modeling-and-partition-strategy.md
- Consistency, replication, and topology: references/consistency-replication-and-topology.md
- Read/write performance and query patterns: references/read-write-performance-and-query-patterns.md
- Compaction, tombstones, and storage behavior: references/compaction-tombstones-and-storage-behavior.md
- Repair, anti-entropy, and operational safety: references/repair-anti-entropy-and-operational-safety.md
- Multi-tenant governance and capacity planning: references/multi-tenant-governance-and-capacity-planning.md
- Reliability and operations: references/reliability-and-operations.md
- Incident runbooks: references/incident-runbooks.md