VikingDB is a cloud‑native vector database that originated at ByteDance, offering high‑performance ANN search, hybrid dense‑sparse retrieval for Retrieval‑Augmented Generation, extensive scaling and filtering capabilities, and ready‑to‑use SDKs for real‑world AI applications.
This article presents a large‑scale distributed algorithm for guaranteed‑delivery advertising that incorporates user‑level personalization, solves a bipartite matching problem with complex constraints, and introduces a real‑time pacing module to smooth delivery while maximizing platform revenue and advertiser satisfaction.
This article recounts Huang Jie’s presentation on the evolution of Ele.me’s monitoring system, detailing its three development stages, the challenges faced, the layered monitoring architecture, the design of a unified platform supporting both PC and mobile, and the underlying LinDB time‑series database.
Ant Financial’s Kubernetes team built one of the world’s largest clusters—hundreds of thousands of nodes—by applying cloud‑native operators, GitOps, and extensive performance optimizations to achieve rapid, automated, and reliable large‑scale workloads during the 2019 Tmall 618 promotion.
