CERN, located near Geneva, Switzerland, is the European Organization for Nuclear Research. At CERN physicists and engineers are working to determine the fundamental structure of the universe. They do so by studying how particles interact when smashing them together or against stationary things at the speed of light with their Large Hadron Collider (LHC).
CERN wanted to upgrade the data monitoring system for one of its Large Hadron Collider experiments called ALICE (A Large Ion Collider Experiment) to ensure the experiment’s high efficiency. They needed to constantly monitor their 2000 nodes processing data at 3.4 TB/s which leads to an incredible ?600 kHz metric rate. These metrics are collected and aggregated by Flume and Spark, and CERN chose InfluxDB to store these metrics.
InfluxDays presentation
CERN presented at InfluxDays London 2018. In this talk, Adam Wegrzynek provided an overview of InfluxDB at CERN. He demonstrated how InfluxDB is critical for various CERN projects including monitoring its accelerator systems, experiments and data centers.
Additional resources
- View Presentation: Adam Wegrzynek's talk at CHEP 2018 Conference
- Watch Talk: Taming Billions of Metrics and Logs at Scale: Two Years with Kafka as a Central Data Hub for Monitoring at CERN
800 PB
Data stored by the Worldwide LHC Computing Grid in 42 countries
600,000
Number of metrics collected per second arriving from 100,000 sources
2 million
Amount of queries per day
“There are many projects at CERN that are using InfluxDB. The largest installation is with the monitoring of one of our data centers where we are using 31 instances of InfluxDB, ingesting over 1.6 TB of metrics a day.”
CJC
CJC’s big data visualization platform uses InfluxDB to store metrics from servers, networking and middleware systems.
MuleSoft
MuleSoft uses InfluxDB to power its Anypoint Monitoring solution to gain visibility into APIs and integrations.
Storj Labs
Storj Labs uses InfluxDB to collect metrics from its decentralized cloud object storage product.
