Meitu migrated its high‑traffic push‑service storage from fragmented Redis nodes to the TiKV‑based Titan system, using a dual‑write rollout, batch‑operation tuning, and TiKV configuration tweaks, achieving required QPS, cutting costs by about 60%, and delivering stable, maintainable service after six months.
This article outlines the evolution of 360’s Titan big‑data processing platform, describing the challenges of traditional script‑based development, the three architectural stages (pre‑Titan, Titan 1.0, Titan 2.0), the functional modules, the DITTO component framework, and key takeaways for building flexible, self‑service data pipelines.
The article explains how relational databases struggle with many‑to‑many and deep relationship queries, compares popular graph databases, details Titan's modular architecture, data model, Gremlin query examples, storage layout, and demonstrates its successful deployment at Paipaidai for large‑scale fraud detection, achieving over 25% efficiency gains.
