Metrics

Metrics is a cloud-agnostic monitoring service that utilizes VictoriaMetrics as its central component. It employs MetricsQL to extract data from its Time Series Database (TSDB), enabling teams to monitor and generate alerts based on identified application and infrastructure metrics.

Features

• Monitor AWS and private cloud applications, infrastructure and services Focus on delivering your application and leave the “How” on monitoring and alerting to the Metrics platform. Monitor your instances or integrate your Cloudwatch metrics with Metrics for alerting.
• Reliable and real-time alerting Alerting from scraped logs is inefficient and adds delays to identifying mission critical incidents. With your monitoring solution hosted with your application, alerting is near real-time bypassing network latencies and log aggregation times.
• Service discovery Metrics will discover any new application instances when following the guidelines around tagging and discovery. No registering required with any additional services like iTRC.
• Long term storage capabilities Save business and analytics metrics for a year. With each application team responsible owning the costs of collection, and a show back model to enable teams to keep an eye on spending.
• Push and Pull Metric Collection Metrics supports the ability to pull metriucs via VMAgent, or recieve metrics from a compatible protocol via push.
• Out of Order Ingest Metrics allows for the storage of out of order metrics, meaning backfilling metrics is supported.
• User experience managed via the ControlPlane Onboarding, learning, and management capabilities will soon be available!
• Visualization via Grafana Centralized highly available Grafana deployed for customers to create dashboards from their metrics datasources. Support for graphing other operational datasets is also available.

Components

• VictoriaMetrics VictoriaMetrics is an advanced time-series database (TSDB) that is designed to provide a comprehensive solution for storing, querying, and analyzing time-series data. The VictoriaMetrics cluster is a specific configuration of VictoriaMetrics that is designed for high-availability, horizontal scalability, and long-term storage.
• AlertManager for raising alerts based on metrics. (While Grafana also has this capability, we encourage AlertManager as a more natural and manageable integration choice.) This comes included with the installation.
• Telegraf can be used to expose metrics from the hosts. Telegraf supports a variety of plugins for integration with various application technologies: the standard setup will expose the usual system metrics, such as CPU, memory and disk utilization.
• Grafana for visualizations. Our on-boarding will provide a location in Grafana where you can build dashboards from your collected metrics.