How Ctrip Revolutionized IDC Management with Visual Automation

Introduction

In recent years the rapid growth of internet services has dramatically increased demand for data‑center IT equipment. Operating a data centre is labor‑intensive, and Ctrip faced many challenges while scaling its infrastructure.

1. Ctrip IDC Development

Ctrip’s equipment grew from a few dozen units to tens of thousands, with rack count reaching thousands. The IDC expansion progressed through three stages:

Stage 1 – Self‑built IDC

High investment and long construction cycles : building an IDC requires power approvals and can take 2–3 years, sometimes longer.

Limited scalability : early capacity estimates become insufficient as business expands.

Stage 2 – Self‑built + Vendor IDC

To avoid heavy‑asset constraints, Ctrip began leasing vendor data centres, but encountered issues such as delayed access to room conditions, coarse monitoring granularity, and slow fault response.

How can limited manpower keep up with fast‑growing data‑center operations? How to achieve fine‑grained operational management? How to spend each yuan efficiently while responding quickly to operational demands?

2. IDC Management Visualization

Ctrip built a visual platform that shows space usage, power, temperature, rack layout, device status, alerts, and historical metrics, enabling engineers to detect failures instantly and improve efficiency.

Room space utilization

Power consumption

Temperature monitoring

Rack layout and device information

Hardware health and alert handling

3. Automation of Device Intake

Receiving thousands of servers each month required a streamlined process. Ctrip introduced barcode scanning, hardware auto‑verification, and a Baremetal workflow that discovers devices, records them in CMDB, validates specifications, and deploys operating systems automatically.

Own standardization: room temperature, cabling, information collection.

Parallel cabling and network device installation improves efficiency.

4. Future Directions

Planned improvements include Redfish‑based server management, deeper analytics of monitoring data to optimize room utilization, dynamic resource scaling, automated alarm handling, and mobile‑first operations for engineers.