Dynamic subrings: Consistent hashing without the tradeoff

Axiom's MetricsDB employs a novel approach to consistent hashing by utilizing dynamic subrings, which helps balance write distribution and read locality in a distributed database. This method involves nesting smaller rings, or subrings, within a main ring for each metric, allowing for scalability and fault tolerance without the need for static trade-offs between write efficiency and read performance. The subrings automatically adjust in size based on the number of series they handle, ensuring that even as metrics scale from thousands to millions of series, the distribution remains optimal. This technique enhances data locality, leading to better compression and reduced storage costs, while also maintaining resilience by shifting data to neighboring nodes in case of failures. By facilitating efficient data distribution across a multi-tenant fleet, MetricsDB can support customers with varying workloads without bottlenecking individual metrics, thereby streamlining queries and lowering latency.

No tracked trend matches for this post yet.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.