Official Ctrip Technology account, sharing and discussing growth.
When attribution analysis on over 900 million rows slowed to more than 40 seconds and threatened cluster stability, Ctrip's smart attribution team rebuilt the architecture with Ray and DuckDB, achieving sub‑15‑second query times, 160 % performance gain, and complete resource isolation.
During a gray‑release of JDK 25 on Ctrip's massive Spark and Flink clusters, silent data corruption appeared in Parquet and ORC files, traced to a G1GC Optional Evacuation bug that moved JNI‑pinned objects, a root cause later back‑ported and fixed in JDK 25.0.3.
A recent K8s host‑level deployment triggered massive service timeouts because runc 1.1.5 passed incorrect CPU‑binding masks to systemd, causing containers to share cores, inflating load and starving workloads, a problem uncovered with Perfetto, BPF tracing and a targeted bug‑fix upgrade.
This article explains how Ctrip’s Universal Control Service (UCS) handles global traffic routing, data‑compliance zones, region failover, and gradual migration by using configurable strategy models, partitioning logic, and a lightweight degradation mask to ensure reliable, compliant user experiences worldwide.
This article explains the design and implementation of Ctrip's Data Replication Center (DRC), a MySQL‑based high‑availability system that solves cross‑region data loop, progress tracking, concurrency, DDL handling, and conflict resolution to achieve low‑latency, reliable data replication for global travel services.
Ctrip migrated its massive User Behavior Tracking system from ClickHouse to a compute‑storage separated StarRocks cluster on Kubernetes, achieving millisecond‑level query latency, halving storage usage, reducing node count, and sustaining millions‑of‑rows‑per‑second write throughput while simplifying scaling and operations.
Ctrip transformed its traditional T+1 offline warehouse into a near‑real‑time lakehouse by integrating Flink CDC with Apache Paimon, designing a two‑stage CDC ingestion, optimizing performance, implementing dynamic updates, and deploying the solution across multiple business scenarios, achieving minute‑level latency, reduced costs, and faster data‑driven decisions.
TripCast introduces a masked 2D transformer pre‑training framework that treats travel demand as a two‑dimensional time‑series problem, leveraging time‑patch tokenization, dual masking and RevIN normalization to achieve state‑of‑the‑art forecasting performance on massive real‑world travel data.
This article details a practical transformation from manual to automated testing using Jest and BDD‑generated HTA, explaining the challenges, technical solutions, mock‑data management, pipeline integration, and measurable results such as higher coverage, lower development cost, and faster release cycles.
This article explains how Ctrip’s AI‑driven Transpiler, built on a ReAct‑inspired multi‑agent framework, automates code transpilation from CRN to xTaro and streamlines large‑scale Java 21 upgrades, dramatically reducing manual effort and improving accuracy across the software development lifecycle.
