What is Application Monitoring eBPF Edition
Alibaba Cloud Application Monitoring eBPF Edition is a one-stop observability product developed for Kubernetes clusters, based on metrics, application traces, logs, and events. Alibaba Cloud Application Monitoring eBPF Edition is designed to provide IT developers and O&M engineers with a holistic observability solution.
Application Monitoring eBPF Edition is in public preview free of charge. If you have any questions, join the DingTalk group (ID: 35568145) to obtain technical support.
Alibaba Cloud Application Monitoring eBPF Edition Features
Feature | Description |
No code intrusion | Uses bypass technology to obtain rich network performance data without requiring code instrumentation. |
Language-agnostic | Parses network protocols at the kernel layer and supports any language or framework. |
High performance | Obtains rich network performance data with minimal overhead using eBPF technology. |
Resource association | Shows relationships between resources using network and resource topologies. |
Data diversity | Supports various types of observable data, including monitoring metrics, traces, logs, and events. |
Holistic | Integrates architecture-aware topology, Prometheus monitoring, and alert configurations through scenario-based designs in the console. |
Features
Resource performance monitoring
Alibaba Cloud Application Monitoring eBPF Edition uses Extended Berkeley Packet Filter (eBPF) to non-intrusively collect performance data for Rate, Errors, and Duration (RED) from containers. This lets you quickly identify performance issues in target containers and their associated pods. Additionally, Alibaba Cloud Application Monitoring eBPF Edition automatically associates the corresponding service and Controller Workload resources, such as deployments, StatefulSets, and DaemonSets, which further reduces the time required to identify issues.
Cluster network topology and network tracing
Application Monitoring eBPF Edition analyzes network requests, automatically parses network protocols, and builds a network topology that supports protocols such as HTTP and MySQL.
The network topology shows the network performance between specific containers or between containers and specific cloud product instances. This helps you quickly identify performance issues in associated services.
By default, Application Monitoring eBPF Edition stores the details of slow or failed requests that have an HTTP return code of 300 or higher, or a response time greater than 500 ms. This lets you view the specific causes of errors and slowdowns.
Alert configuration
Application Monitoring eBPF Edition provides out-of-the-box alert templates. You can create alert rules from these templates or customize rules for specific Kubernetes clusters. When an alert rule is triggered, a notification policy notifies the specified contacts. This prompts the contacts to take the necessary action to resolve the issue.
Host and workload resource monitoring
You can monitor basic metrics of Elastic Computing Service (ECS) instances, such as CPU usage, memory usage, and disk usage, to ensure that host resources are sufficient.
You can monitor basic metrics for pods and containers to ensure that workload resources are sufficient. These metrics include CPU usage, request rate, limit rate, memory usage, and disk usage.
Advantages of Alibaba CloudApplication Monitoring eBPF Edition
Compared to open source Kubernetes monitoring solutions, Application Monitoring eBPF Edition offers the following advantages:
Innovative resource association and interaction
Application Monitoring eBPF Edition monitors network requests to build a network topology. This helps you view service dependencies.
Rich data diversity
Application Monitoring eBPF Edition provides visualizations for various types of data.
Contact us
If you have any questions while using Application Monitoring eBPF Edition, contact the DingTalk group (ID: 35568145) for assistance.