Tailor made monitoring solutions
The CMS experiment is one of four particle physics detectors built at the Large Hadron Collider (LHC). Located deep underground at the border of Switzerland and France, the project is currently focused on experiments investigating Standard Model physics, extra dimensions and dark matter.
The computing infrastructure to deal with the multi-petabyte datasets produced by CMS requires best-in-class systems to monitor workload and data management, data transfers, and submission of production requests.
The CMS experiment has long relied on scalable, open source solutions to satisfy real-time and historical monitoring needs. However, after encountering storage and scalability issues with long-term monitoring solutions such as Prometheus and InfluxDB, the CMS monitoring team began the search for alternatives.
Edging out existing technology
The CMS monitoring team has engaged VictoriaMetrics following a post by CTO and Co-Founder Aliaksandr Valialkin on Medium, which benchmarked VictoriaMetrics against other popular monitoring systems, and were won over by the rigorous, scientific detail on display.
"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." said V. Kuznetsov from Cornell University (member of CMS collaboration).
Initially implementing VictoriaMetrics as backend storage for Prometheus, the CMS monitoring team progressed to using the solution as front end storage to replace InfluxDB and Prometheus. This had the added impact of removing cardinality issues with Influx.
Since installing VictoriaMetrics, the CMS monitoring team had zero issues with cardinality, or using the software on the operational side. The CMS monitoring team gained added confidence in the open source flexibility of VictoriaMetrics after seamlessly implementing new features for vmalert, the solution's alerting system.
"Working with CMS to monitor the experiment computing infrastructure is a great honor for the team here. The number of use cases for monitoring and observability is growing exponentially, and seeing our tech applied to cutting-edge science is testament to how critical monitoring has become. Our open source, community driven model is and will be at the core of our offering, granting us the flexibility to serve projects as complex as CMS infrastructure in the future", said Roman Khavronenko, Co-Founder of VictoriaMetrics.
Milestones gather momentum
The announcement marks the latest in a series of recent milestones for the company. Founded in 2018, by former engineers from Google, Cloudflare, and Lyft looking for ways to measure the growth of data in their organisations, VictoriaMetrics now counts Ably, Roblox, and Semrush as part of its thriving open source community.
In 2022, VictoriaMetrics announced surpassing 50 million docker pulls, 1M GitHub downloads, and 6,000+ GitHub Stars.