VictoriaMetrics

VictoriaMetrics is a fast, cost-effective, and scalable open-source time series database and monitoring solution. It works as a Prometheus-compatible remote storage with its own query language, enabling long-term storage and high-speed queries of metrics. This allows visualization of the overall system performance in large-scale monitoring environments and facilitates quick identification and resolution of issues.

Key Facts

Architecture

Key Features

Key features of VictoriaMetrics include:

• High-speed writes and queries: Handles massive amounts of time series data with minimal resources.
• Prometheus compatibility: Supports PromQL and the extended MetricsQL, allowing a smooth migration from existing Prometheus environments.
• Low resource consumption: Uses little memory and CPU, with high storage compression that reduces operating costs.
• Horizontal scalability: Supports both single-node and cluster configurations for flexible scaling.
• Long-term storage: Suitable for long-term retention of metrics and usable as remote storage for Prometheus.

VictoriaMetrics is used in combination with popular open-source tools like Prometheus, Grafana, and OpenTelemetry, balancing ease of use and flexibility. It also supports integration with container orchestration platforms like Kubernetes and Docker.

VictoriaMetrics's GitHub repository can be found at the following URL: https://github.com/VictoriaMetrics/VictoriaMetrics

By using VictoriaMetrics, collecting, storing, and analyzing large-scale time series metrics becomes easier. Through visualization and analysis of the entire system, it is possible to improve performance and ensure system stability.

Use Cases for VictoriaMetrics OSS

VictoriaMetrics OSS is an open-source monitoring platform that supports monitoring and troubleshooting of large-scale time series metrics. Below, we explain in detail how VictoriaMetrics OSS is used in certain software architectures, why it is used in those software architectures.

Use Cases

For example, if a company adopts a microservices architecture or Kubernetes clusters, a huge volume of metrics is sent in real-time from many services and nodes. In such cases, real-time monitoring of the performance of each service and node is necessary, and quick response is required when issues arise. VictoriaMetrics OSS is very useful for companies struggling with the scalability of Prometheus, as it can effectively and rapidly collect and store large-scale metrics in such situations.

Use in Software Architectures

VictoriaMetrics OSS is a monitoring tool based on a Prometheus-compatible time series database. Prometheus-based metric collection has challenges with long-term storage and scalability, but by combining VictoriaMetrics as a remote storage, these challenges can be solved. Therefore, it is suitable for use in systems that utilize microservices architecture, containers like Docker, and Kubernetes, especially environments requiring large-scale monitoring.

Purpose of Use

Companies that adopt microservices architecture or container platforms must continuously monitor metrics generated from many services and nodes, making monitoring crucial. By using VictoriaMetrics OSS, long-term storage of metrics with low cost and high performance becomes possible, and the state of systems can be visualized quickly and accurately using PromQL/MetricsQL. This helps maintain service quality and reduce response times in case of failures.

The VictoriaMetrics source code includes the following key packages:

1. vmagent:

   • This package contains the code of a lightweight agent that scrapes metrics from various endpoints and forwards them to VictoriaMetrics or any Prometheus-compatible storage. It provides functions such as remote data collection and label rewriting.

2. vminsert:

   • This package contains the code of the component responsible for receiving and routing metrics in cluster configurations. It distributes data received from clients across appropriate vmstorage nodes for writing.

3. vmselect:

   • This package contains the code of the component responsible for accepting queries, querying multiple vmstorage nodes, and aggregating results. It processes PromQL/MetricsQL queries and returns results to users or Grafana.

4. vmstorage:

   • This package contains the code of the storage engine responsible for persisting time series data. It is optimized for high compression and fast writes/reads, making it suitable for long-term storage.

5. vmalert:

   • This package contains the code of the component that evaluates alerting and recording rules based on PromQL/MetricsQL. It can integrate with Alertmanager to send alert notifications.

These packages provide the necessary components to realize the key functionalities of the VictoriaMetrics project. Each package is appropriately divided to ensure code maintainability and extensibility. The VictoriaMetrics source code is available at https://github.com/VictoriaMetrics/VictoriaMetrics, so those interested can check for more details.