VictoriaMetrics Observability Blog

Author
Category

Filter: Performance

Announcing vmestimator: Real-time Cardinality Estimations for VictoriaMetrics and Prometheus

vmestimator is a new open-source tool that estimates metric cardinality in real time. It can alert to high-cardinality spikes before they slow queries and increase resource usage. Works with any Prometheus-compatible setups, Grafana, and vmalert

How Airbnb Built a High-Volume Metrics Pipeline with OpenTelemetry and vmagent

Learn how Airbnb rebuilt its observability pipeline with OpenTelemetry and vmagent to handle over 100 million samples per second, reduce cost by 10x, and simplify high-scale metrics aggregation.

VictoriaMetrics at KubeCon: Optimizing Tail Sampling in OpenTelemetry with Retroactive Sampling

If you’ve been struggling with the high resource overhead of tail sampling, check out retroactive sampling, an approach that significantly reduces sampling overhead for distributed tracing in OpenTelemetry.

Developer's Note: Research on Distributed Tracing, Comparing With Tempo and ClickHouse

When it comes to observability, people often refer to its three pillars: metrics, tracing, and logging. At VictoriaMetrics, we already have solutions for metrics and logging. Throughout these years, we’ve heard a common question from the community: When will VictoriaTraces be released? If you are interested in this topic, then this blog post is for you. We’ll share the results of our research into distributed tracing and explore two key questions: Will there be a VictoriaTraces? And how competitive could it be?

Inside Go's Unique Package: String Interning Simplified

When you’ve got several identical values in your code, you only store one copy. Instead of having several copies of the same thing, they all just point to this one version, which is a lot more efficient. It’s a process often called ‘interning’ in programming circles.

How to reduce expenses on monitoring: be smarter about data

This blog post is the second in the series of the blog posts based on the talk about ‘How to reduce expenses on monitoring’, stackconf 2023. It is about open-source instruments and techniques from the VictoriaMetrics ecosystem for improving cost-efficiency of monitoring.

Comparing Performance and Resource Usage: Grafana Agent vs. Prometheus Agent Mode vs. VictoriaMetrics vmagent

We compared the performance and resource usage of Grafana Agent, Prometheus Agent Mode, and VictoriaMetrics vmagent to help readers make informed decisions when choosing an agent for their monitoring needs.

How to reduce expenses on monitoring: Swapping in VictoriaMetrics for Prometheus

This blog post is the first in the series of the blog posts based on the talk about ‘How to reduce expenses on monitoring’, stackconf 2023. It is about open-source instruments and techniques from theVictoriaMetrics ecosystem for improving cost-efficiency of monitoring.

Performance optimization techniques in time series databases: sync.Pool for CPU-bound operations

This blog post is the fourth in the series of blog posts based on the talk about ‘Performance optimizations in Go’, GopherCon 2023. It is dedicated to various optimization techniques used in VictoriaMetrics for improving performance and resource usage.