Open Source & Enterprise Observability for Infrastructure & AI

We were searching for alternative solutions following performance issues with Prometheus and InfluxDB. VictoriaMetrics’ walkthrough of use cases, and concise detail gave us excellent insight into how they could help us. The solution’s backwards compatibility with Prometheus made implementation into the CMS monitoring cluster as smooth and seamless as possible.

Fortunately, with VictoriaMetrics, we were able to sign a contract to get the support we needed to confidently run a monitoring system at our scale. The numbers speak for themselves. Our proof-of-concept trial showed dramatically reduced compute and storage costs, translating into a 10x lower AWS bill.

To effectively manage the vast amounts of data generated by Gemini’s solar panels and battery storage systems, we implemented a solution that could handle millions of metrics per second with exceptional accuracy and efficiency. VictoriaMetrics is an essential part of data management for the Gemini project and is the backbone of IHI Terrasun’s data collection and management system. The platform’s ability to handle massive datasets with ease has been invaluable.

We tested other available monitoring solutions like Prometheus but it never scaled well enough for us, then we discovered VictoriaMetrics. Acting as a drop-in replacement for Prometheus, we were able to set-up our initial test system within only 20 minutes. With the Enterprise features, we will be able to keep our slightly down-sampled metrics for months, or years, instead of throwing everything away after two weeks as we do currently.

We decided to switch to VictoriaMetrics Cloud from Grafana Cloud (Mimir), which, with our large number of active time series, we found very expensive for storing metrics. VictoriaMetrics is compatible with Prometheus, so we didn’t need to change any of our code or queries. Due to VictoriaMetrics’ high optimization features, it handles our volume of time series on relatively modest hardware, reducing our costs of metrics storage by approximately five times.