Databases 8 min read

How JD Cloud’s JCHDB Powered the 11.11 Shopping Festival’s Massive Data Surge

This article explains how JD Cloud’s JCHDB database handled PB‑level data growth during the 11.11 shopping festival, detailing the high‑availability architecture, performance optimizations, scaling techniques, and the eight‑step preparation process that enabled millions of queries per second and terabit‑level traffic.

JD Cloud Developers
JD Cloud Developers
JD Cloud Developers
How JD Cloud’s JCHDB Powered the 11.11 Shopping Festival’s Massive Data Surge

During JD's 11.11 shopping festival, JD Cloud's database handled PB‑level data growth, achieving a 258% increase compared to the 618 peak, with total order amount exceeding 271.5 billion yuan, up 33% YoY.

JCHDB, an OLAP service built on ClickHouse, provides a distributed architecture with multi‑core, multi‑node parallel queries, delivering 1‑2 orders of magnitude faster performance than traditional open‑source databases, meeting real‑time analysis needs during large‑scale promotions.

Real‑time order data flows through Kafka and Flink into JCHDB; the system supports batch writes of 50‑200 MB/s, but frequent small‑batch writes during the event required tuning of ZooKeeper JVM parameters and cloud‑disk concurrency to ensure stability.

The team optimized write patterns to favor larger, less frequent batches while maintaining query performance, and adjusted cluster parameters to sustain service under high pressure.

Key operational steps for the 11.11 preparation include: (1) scope identification, (2) traffic estimation, (3) plan drafting, (4) monitoring and alert design, (5) stress testing, (6) rehearsal, (7) on‑call duty, (8) post‑event review.

High‑availability architecture offers single‑zone and multi‑zone deployments, anti‑affinity for primary‑replica, sub‑2 ms replication, and a custom sentinel system for automatic failover.

Fault‑tolerant design enables second‑level automatic switch‑over, ensuring data consistency by replaying logs before VIP switch.

Elastic storage expansion leverages cloud‑disk incremental snapshots, allowing expansion to any size within 3‑5 minutes for cloud‑disk instances and near‑instant vertical scaling for local‑disk instances.

During the “opening red” period, the database peaked at 5.024 million QPS and 1.183 Tbps traffic, demonstrating the effectiveness of the architecture.

Overall, JD Cloud’s database provides end‑to‑end services from provisioning to monitoring, supporting JD Retail, Logistics, AI, Health, and other core businesses, making it a reliable choice for enterprise cloud migration.

e-commerceperformancedatabaseHigh AvailabilitycloudScaling
JD Cloud Developers
Written by

JD Cloud Developers

JD Cloud Developers (Developer of JD Technology) is a JD Technology Group platform offering technical sharing and communication for AI, cloud computing, IoT and related developers. It publishes JD product technical information, industry content, and tech event news. Embrace technology and partner with developers to envision the future.

0 followers
Reader feedback

How this landed with the community

login Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.