Getting Started with LHCb Open Data

LHCb offers two means for exploring open data:

the LHCb Ntupling Service: a web service for on-demand production and publishing of custom LHCb open data. This amounts to over 4 PB of data to explore! For this option, users interact with ntuples, and no experiment specific software is needed.
The Run 1 $pp$ data collected at LHCb is available on the CERN Open Data portal in DST format. For this option, users interact with DST files, and need to use LHCb specific software to create ntuples from the DST files.

The LHCb Open Data Guide provides a usage guide with examples for both of these options.

LHCb Ntupling Service

▶ Try the new LHCb Ntupling Service, allowing for custom ntuple creation and exploration of more than 4 PB of pp data!

In 2026, LHCb released the LHCb Ntupling Service, providing the public access to both Run 1, and for the first time, Run 2 $pp$ data collected by LHCb. This release marks a significant advancement in the LHCb open data infrastructure, and is an exciting step for the larger open data community, where the barrier for entry of data analysis is lowered without the need for any experiment specific software. Instead, custom ntuples are produced containing a collection of physics objects and quantities based on user specifications. The Ntupling Service serves as an all in one place for users to request custom ntuples with LHCb data, track the request process, communicate with the LHCb open data team, and download the resulting ntuples.

The Ntupling Service chapter of the LHCb Open Data Guide contains a detailed walkthrough for request submission and tracking, in addition to a growing list of analysis example workflows with the resulting ntuples.

Run 1 DSTs

By the end of 2023, LHCb released all of its Run I data, via CERN Open Data Portal, to the general public. The data comes in .DST and .MDST format which is the same format used by LHCb internally.

Every data set released is narrated by an "Open Data Record" accessible through the CERN Open Data portal. Open Data Records contain various bits of information about the selected data set (this is called metadata). An example of the types of metadata provided in the record is:

Number of events in the dataset
Number of files in the dataset
Combined size in TB of the dataset
Production ID
Production Type
Detector conditions (condb, dddb tags)
List of Trigger Configuration Keys (TCKs)
Scripts used for each production step
List of Logical File Names (LFNs) on LHCb DIRAC

The Analyzing Run 1 LHCb Open Data chapter of the LHCb Open Data Guide and the metadata provided on the Open Data Portal should help the user to navigate, select, and work with LHCb Open Data.