ATLAS

Abstract:

The ATLAS Eventlndex is the global catalogue of all ATLAS real and simulated events. During the LHC long shutdown between Run 2 (20152018) and Run 3 (2022-2025) all its components were substantially revised and a new system was deployed for the start of Run 3 in Spring 2022. The new core storage system, based on HBase tables with a SQL interface provided by Phoenix, allows much faster data ingestion rates and scales much better than the old one to the data rates expected for the end of Run 3 and beyond. All user interfaces were also revised and a new command-line interface and web services were also deployed. The new system was initially populated with all existing data relative to Run 1 and Run 2 datasets, and then put online to receive Run 3 data in real time. After extensive testing, the old system, which ran in parallel to the new one for a few months, was finally switched off in October 2022. This paper describes the new system, the move of all existing data from the old to the new storage schemas and the operational experience gathered so far

Abstract:

The ATLAS EventIndex is a global catalogue of the events collected, processed or generated by the ATLAS experiment. The system was upgraded in advance of LHC Run 3, with a migration of the Run 1 and Run 2 data from HDFS MapFiles to HBase tables with a Phoenix interface. Two frameworks for testing functionality and performance of the new system have been developed. There are two types of tests running. First, the functional test that must check the correct functioning of the import chain. These tests run event picking over a random set of recently imported data to see if the data have been imported correctly, and can be accessed by both the CLI and the PanDA client. The second, the performance test, generates event lookup queries on sets of the EventIndex data and measures the response times. These tests enable studies of the response time dependence on the amount of requested data, and data sample type and size. Both types of tests run regularly on the existing system. The results of the regular tests as well as the statuses of the main EventIndex subsystems (services health, loaders status, filesystem usage, etc.) are sent to InfluxDB in JSON format via HTTP requests and are displayed on Grafana monitoring dashboards. In case (part of) the system misbehaves or becomes unresponsive, alarms are raised by the monitoring system