Information aggregation and analytics for ATLAS Frontier

CERN is a European Research Organization that operates the largest particle physics laboratory in the world. The Large Hadron Collider (LHC) at CERN has many detectors where the particles collide, the biggest detector called ATLAS generates a large amount of raw data per second, about 1 petabyte/second, and several Oracle databases are used to manage the processing of this data.

One of the most challenging ATLAS databases is the COOL database which stores the Conditions data in neutral schemas, it contains information about the calibrations, data-taking conditions and detector status.

More than 150 computing sites access this Conditions database using the squid-Frontier system which is a squid cache hierarchy used to cache the data at the computing site and make the access easier and faster, besides a Frontier web service to take charge of the different queries requiring data from the COOL database.

A decrease in the system performance has been noticed for some data processing workflows, that’s why the development of a monitoring and performance evaluation tool is necessary in order to study the system components in detail and detect possible
anomalies. This report presents a CERN openlab project that consists of setting up an information aggregation and analytics tool in studying the breaking points of the squid-Frontier system.