Metoro

Observability for Microservices in Kubernetes

Freemium

Description

Metoro is an observability platform specifically designed for monitoring microservices within Kubernetes environments. It leverages eBPF (extended Berkeley Packet Filter) technology to automatically collect detailed telemetry data directly from the Linux kernel. This approach eliminates the need for manual code instrumentation or restarting application containers, allowing for instant visibility upon installation.

The platform gathers a wide range of data, including Application Performance Monitoring (APM) metrics, logs, traces, container profiles, events, and network topology information. Metoro presents this data in a unified dashboard, offering features like custom visualizations, automated performance regression detection, and AI-driven root cause analysis to help teams quickly identify and resolve issues within their clusters. It supports various Kubernetes distributions, including managed services like EKS, GKE, AKS, and bare metal setups.

Key Features

Automatic APM: Provides full APM data derived from eBPF traces with zero setup or code changes.
Kubernetes Log Management: Collects and analyzes logs from all containers within the cluster.
Kubernetes Container Profiling: Delivers deep insights into container performance, down to the function level.
eBPF-based Data Collection: Utilizes eBPF to gather observability data at the kernel level without instrumentation.
Integrated Observability Data: Combines Logs, Metrics, Traces, and Profiling in a single interface.
Custom Dashboards and Metrics: Allows creation of custom visualizations using built-in or external metrics (Otel/Prometheus).
Automated Performance Regression Monitoring: Periodically profiles services to detect and alert on performance regressions.
AI-Driven Root Cause Analysis: Proactively monitors changes and uses AI (OpenAI models) to investigate root causes.
Network Topology Visualization: Maps out network communication within the cluster.
1 Minute Install: Simple setup process using a single command.

Use Cases

Monitoring microservice performance in Kubernetes.
Debugging application issues within Kubernetes clusters.
Analyzing container resource usage and performance.
Centralizing observability data (logs, metrics, traces, profiles).
Tracking application performance changes over time.
Automating the identification of root causes for incidents.
Visualizing network interactions between services.

Frequently Asked Questions

What is eBPF?

eBPF is a linux kernel technology that allows running code in response to kernel events. Metoro uses it to collect telemetry data at the kernel level for each container without needing instrumentation or restarts.

What happens if there's a bug in Metoro? Will it crash my host?

No, eBPF is designed such that an eBPF program crash will not cause the kernel to crash, unlike older kernel modules.

How does Metoro work without manual instrumentation?

Metoro uses its own eBPF programs loaded into the kernel of all nodes in the Kubernetes cluster, collecting data automatically without requiring container restarts.

Which LLM providers does Metoro support for AI features?

Metoro currently uses OpenAI models for its AI features. The cloud offering utilizes Microsoft's hosted OpenAI API.

Do you offer on-premises installations?

Yes, on-premises installations are available as part of the enterprise offering.