Uber Redesigns M3DB Sharding with Subclusters to Limit Failure Impact

Leela Kumili — Mon, 21 Sep 2026 14:37:00 GMT

Uber has redesigned shard placement in M3DB with fixed size subclusters to limit the impact of node failures, maintenance, and cluster scaling. The approach bounds shard dependencies, preserves replica isolation, and uses a greedy algorithm to select shard migrations while avoiding a separate rebalancing pass and unnecessary data movement.

By Leela Kumili

From Memory-Hungry HNSW to Quantized SPANN: the Technical Evolution of Pinterest's Manas Platform

Olimpiu Pop — Wed, 16 Sep 2026 06:06:00 GMT

Pinterest Engineering has enhanced its Manas search platform to manage vast data, improving efficiency in search and discovery functions. By applying Scalar and Product Quantization, memory usage decreased significantly while maintaining high recall rates. The platform utilizes SSDs for optimized performance, and is transitioning to multi-vector models for refined relevance matching.

By Olimpiu Pop

InfoQ - Algorithms - News

Uber Redesigns M3DB Sharding with Subclusters to Limit Failure Impact

From Memory-Hungry HNSW to Quantized SPANN: the Technical Evolution of Pinterest's Manas Platform