This article walks through implementing a compact, 300‑line Go version of Olric—a distributed key‑value store—covering core data structures, shard routing, simplified RPC, TTL handling, node replication, rebalancing, concurrency safety, and performance experiments with benchmarks, profiling, and memory optimizations.
This article explains Ozone's object storage architecture, its evolution of metadata management using distributed KV stores like Apache Cassandra, and the performance optimizations—read/write separation, unlimited scaling, and partitioning—that enable high‑throughput, low‑latency handling of massive datasets.
This article compares the consistency read mechanisms of the distributed key‑value stores Consul and etcd, detailing Consul’s three read modes and leader‑forwarding logic, and explaining etcd’s serialize and linearizable reads, including the internal notification and index‑checking processes.
UCloud’s US3 object storage metadata service, originally built on a chained MongoDB architecture, faced scalability, performance, and cost challenges, prompting a redesign that introduces a high‑compatibility DB‑Gateway, a distributed KV store (UKV) with custom RocksDB, delivering faster reads, zero list‑service latency, 80% cost reduction, and simpler operations.
RedKV, Xiaohongshu’s self‑developed distributed NoSQL key‑value store, combines a Redis‑compatible interface, RocksDB‑backed storage, and a shard‑managed multi‑cloud architecture that delivers three‑times higher write throughput, 1.5‑times read throughput, 40 % cost savings versus HBase, and sub‑5 ms P999 latency for petabyte‑scale workloads.
This article explains the core concepts and detailed design of a custom message queue built on a distributed key‑value store, covering terminology, architecture, broker metadata, topic metadata, message format, sending, storage, retrieval, delay handling, retry mechanisms, dead‑letter queues, and TTL policies.
This article details the challenges of WeChat Pay’s original transaction‑record system, the architectural redesign using distributed key‑value storage and data partitioning, and the resulting improvements in scalability, reliability, security, and user experience.
This article presents a comprehensive case study of how WeChat Pay rebuilt its transaction record storage to handle massive data growth, improve performance, ensure data completeness, and strengthen security through a distributed key‑value architecture, hierarchical archiving, and robust operational safeguards.
