CERN Accélérateur de science

ATLAS Slides
Report number	ATL-SOFT-SLIDE-2023-165
Title	Query Service for the new ATLAS EventIndex system
Author(s)	Rybkin, Grigori (Université Paris-Saclay (FR))
Corporate author(s)	The ATLAS collaboration
Collaboration	ATLAS Collaboration
Submitted to	26th International Conference on Computing in High Energy & Nuclear Physics, Norfolk, Virginia, Us, 8 - 12 May 2023
Submitted by	grigori.rybkine@cern.ch on 05 May 2023
Subject category	Particle Physics - Experiment
Accelerator/Facility, Experiment	CERN LHC ; ATLAS
Free keywords	ATLAS experiment ; Event catalogue ; BigData catalogue ; Hadoop ; HBase
Abstract	The ATLAS EventIndex system consists of the catalogue of all events collected, processed or generated by the ATLAS experiment at the CERN LHC accelerator, and all associated software tools. The new system, developed for LHC Run 3, makes use of Apache HBase - the Hadoop database - and Apache Phoenix - an SQL/relational database layer for HBase - to store and access all the events metadata. The Query Service allows users to retrieve the information of interest. Its design is based on a server-client architecture with the sides communicating via a RESTful API. The OpenAPI Specification is used to describe the API and generate parts of the source code of the server and client. Based on selection criteria received from the client, the server executes SQL queries against the Phoenix tables, retrieves the data, and returns it to the client. The client - a command line utility following the UNIX/POSIX-conforming program design style - can output results in various formats including plain text and JSON. The server and the client are developed in Java 8, with the use of the Jersey 2 JAX-RS API implementation and the Spring framework. The service performance is suitable for interactive use with queries for up to 1000 events at a time.