This article explains the evolution from Pika, a RocksDB‑backed Redis‑compatible KV store, to ZestKV’s serverless, storage‑compute separated architecture, detailing design shortcomings, core components, multi‑tenant support, elastic scaling, performance benchmarks, and future roadmap.
This article details how Bilibili's KV storage system evolved from early solutions like Redis and MySQL to a highly scalable, high‑availability distributed architecture, describing its overall design, node components, data splitting, multi‑active disaster recovery, typical use cases, and operational challenges.
Pika, a disk‑based KV database fully compatible with the Redis protocol, was created by 360 to cut storage costs, has been adopted by major Chinese internet firms, and is now an OpenAtom open‑source project aiming to become a global benchmark for cloud‑native KV databases.
Vivo’s feature‑storage platform, built on Nebula Graph’s Raft‑based, storage‑compute‑separated architecture and exposed via Redis‑compatible proxies, meets massive, low‑latency AI data demands while offering strong consistency, horizontal scalability, backup, active‑active replication, and a roadmap toward general‑purpose KV, cloud‑native integration, and advanced storage engines.
This article introduces Zhuanzhuan's multi‑service storage layer—including MySQL, TiDB, ZZRedis, ZZKV/ZZList, and ZZOS—detailing their applicable scenarios, architectural designs, and the improvements made to address scalability, security, and resource‑usage challenges.
