Customer background

In the global digital marketing field, Yidian Tianxia Network Technology Co., Ltd. plays a key role. As a global leader in international intelligent marketing services for enterprises, Yidiantianxia is committed to focusing on "overseas services" and "advertising technology," providing enterprises with global integrated marketing, digital marketing, programmatic advertising monetization, and other overseas marketing services including brand case management, performance advertising, and leading media account management, as well as digital services like AI digital creativity, BI decision-making, and CI intelligent multi-cloud management. It has successfully helped 10,000+ high-quality enterprises achieve global business growth, driving the overseas expansion of new technologies, products, and models for many of China's leading industries, including cross-border e-commerce, gaming and entertainment, tool applications, cultural tourism, new energy, and high-end manufacturing.

As the business expanded, the technical team encountered three specific issues at the monitoring level:

Infrastructure complexity

The introduction of large-scale Kubernetes container clusters has made the collection and systematic construction of monitoring data from nodes, containers, to services more complex.

Low efficiency in problem localization

When applications experience performance anomalies, because links, logs, and metric data are isolated, troubleshooting root causes takes too long.

Pressure to ensure service quality

As a service provider, any platform instability can directly impact the client's business, so maintaining high-level SLAs is crucial.

Solutions

Build a unified observable system across the entire domain

Build a unified observation platform to efficiently manage rapidly growing complex business flows, enhance global business visibility and controllability, and continuously ensure that global user experience and service stability remain at the industry-leading level.

Building an efficient K8S cluster monitoring system

By Guance DataKit's collection capabilities, it achieves comprehensive collection of all dimensional metrics and log data for K8s cluster nodes, containers, and cluster services, building a foundational monitoring system covering the entire container environment at a low cost, ensuring data real-time and accuracy.

Full-link tracing and correlation analysis

By leveraging DataKit's Java application automatic injection mechanism, business applications can be easily integrated into APM for link analysis, including scenarios such as abnormal latency, root cause response failures, and backend service performance statistics. Through correlation analysis, APM links, business logs, and performance metrics are linked to enable deep monitoring of core services and improve fault response efficiency.

Customer outcomes

Upgraded troubleshooting efficiency, leaving no place for faults to hide

With Guance's full-link observability, teams can quickly pinpoint root causes, accurately diagnose faults, effectively improve system stability and reliability, reduce downtime, and provide solid support for business continuity.

Data-driven: Decision-making with "data" in mind

Through multidimensional data correlation analysis, teams can flexibly build data analysis dashboards, which not only detect potential anomalies but also provide data support for architecture optimization and business decision-making.

Unified monitoring architecture simplifies operations and maintenance

Guance helped Yidiantianxia build a unified container monitoring system with extremely low construction complexity, unleashing the productivity of operations teams and allowing them to focus more on strategic architecture optimization and innovation.

Yi Dian Tian Guance