Observability Platform functions as a centralized observability solution provided by Volcano Engine for cloud\-native environments. It aims to aggregate cloud\-native observability services and build a complete observability ecosystem that systematically displays all observability services. This simplifies user operations, enhances user experience, and improves integration among services. Observability Platform provides you with an easy\-to\-use portal for observability access and query. It achieves this by aggregating a series of services, such as AI application monitoring, Server\-side Monitoring, VMP, Cloud Monitor, RUM, and CloudDetect, and integrating general modules, such as custom dashboards, Alert Center, and Integration Center. ![图片](https://portal.volccdn.com/obj/volcfe/cloud-universal-doc/upload_342f26ef697e48fad2f68c1ca3f04b4f.png) |Service |Feature overview |Common scenario | |---|---|---| |AI application monitoring |Supports transparent instrumentation for mainstream inference frameworks, parses agent traces, and quickly identifies end\-to\-end issues from agents to inference engines. Provides aggregated views centered on LLMs for observability, supports LLM\-based session monitoring and upstream/downstream monitoring of MCP services, and provides monitoring capabilities for AI business scenarios. |* Large AI model root cause analysis

* LLM trace issue identification

* Comprehensive agent monitoring and tracing | |Server\-side Monitoring |Provides an end\-to\-end monitoring service for the server\-side application backend, which involves performance monitoring for basic resources, service and code layer monitoring, and online application health monitoring. |* Inference fault detection

* Model metric monitoring

* Intelligent root cause identification | |Volcengine Managed Service for Prometheus |Deeply compatible with cloud native and community ecosystems to provide an operation and maintenance\-free, highly available VMP service. Supports configuration with a few clicks to quickly write metrics from AI frameworks, open\-source add\-ons, cloud services, and multi\-cloud/hybrid cloud services. Provides pre\-configured dashboards for metrics of AI frameworks, open\-source add\-ons, and cloud services. These capabilities help you quickly create monitoring dashboards. |* Machine learning platform observability

* Inference suite and training suite observability

* Multi\-cloud observability and custom observability | |Cloud Monitor |Collects and displays the resource status of various cloud services, and provides flexible alerting and resource dashboards. Allows you to view the quantity and health status of resources corresponding to various monitoring metrics across regions and cloud services. |* Infrastructure monitoring

* Service event handling

* Alert and fault management | |RUM |Provides comprehensive monitoring and analysis capabilities across all major client platforms, including App, WebPro, Mini Program Pro, Mini Program, PC, and OS. |* App crash event collection and stability analysis

* Browser white screen monitoring

* Performance analysis of mini program page loading | |CloudDetect |Simulates user access to cloud service by using monitoring nodes distributed around the world to achieve end\-to\-end observability of network availability, stability, and user experience. The observability capabilities include on\-demand synthetic monitoring, network quality monitoring, DNS monitoring, API monitoring, and web page monitoring.

|* Network quality alerting

* Workplace network quality management

* CDN assessment and competitor analysis |