Qunar Tech Salon is a learning and exchange platform for Qunar engineers and industry peers. We share cutting-edge technology trends and topics, providing a free platform for mid-to-senior technical professionals to exchange and learn.
This document describes the background, problems, requirements, analysis, and detailed design of a high‑performance, highly available, and scalable log‑analysis pipeline for MySQL that leverages Kafka, ClickHouse, MySQL, and custom services to aggregate, enrich, and visualize massive query logs.
This article examines why Redis HSCAN can cause high CPU usage on ziplist‑encoded hash objects, compares its performance with HGETALL through benchmark tests, and provides practical recommendations for avoiding costly full‑scan operations in production environments.
This article explains the purpose and mechanics of PostgreSQL's VACUUM and AUTOVACUUM processes, demonstrates practical examples of table bloat, transaction ID handling, and provides step‑by‑step guidance for precise triggering and performance optimization of vacuum operations.
This case study details how Qunar’s hotel pricing platform was reorganized in 2021 using Domain‑Driven Design and explicit architecture principles to streamline cross‑team collaboration, reduce data‑flow complexity, and improve response time by 25%, while aligning system structure with Conway’s Law.
In this interview, Qunar's technical director Zheng Jimin shares how the company revived Domain‑Driven Design to restructure hotel pricing, discusses strategic and tactical design, implementation challenges, success criteria, and the broader impact of technical influence on organizational growth.
The article analyses why indiscriminately killing slow MySQL queries is risky, enumerates the drawbacks of common kill‑based approaches, and proposes more reliable designs such as registration, signature‑based killing, and source‑code modifications to safely mitigate query‑induced database avalanches.
This article details the background, design rationale, network topology changes, and step‑by‑step procedures—including VPC configuration, BGP setup, and port‑channel adjustments—used to upgrade Qunar's data‑center network for Kubernetes deployments, with practical code examples and operational tips.
This article details Qunar's high‑concurrency flight pricing architecture, covering the origin of pricing data, two‑level caching design, a cache‑management system, CQRS‑based policy storage, data compression techniques, and performance improvements achieved through these optimizations.
This article compiles a selection of Qunar's 2021 technical writings covering architecture design, big data processing, front‑end engineering, and cloud‑native practices, providing titles, authors, brief abstracts, and direct links for readers seeking in‑depth engineering insights.
This article compiles the 2021 Qunar Technical Salon infrastructure series, presenting original technical writings on databases, operational practices, and core components, each linked to detailed posts that share real‑world experiences, design guidelines, and performance insights for engineers and practitioners.
