Laboratory of Information Technologies

Date and Time: Tuesday, 4 June 2024, at 11:00 AM

Venue: room 310, Laboratory of Information Technologies

Information about seminar

Seminar of participants of XIII Scientific Conference of Young Scientists and Specialists “Alushta-2024”

1. Seminar topic: “JINR Repository of scientific publications on the Dspace platform (Pilot version)”

   Speaker: Tatiana Zaikina

   Authors: Irina Filozova, Galina Shestakova, Andrey Kondratyev, Aleksey Bondyakov, Tatiana Zaikina, Irina Nekrasova, Yanina Popova

   Abstract:

   The JINR Open Access Archive (repository) on the DSpace platform consist of thematic communities and collections with various types of scientific documents. The repository is organized by Open Access principles as part of the JINR Digital Ecosystem (https://digital.jinr.ru/). The repository supports English and Russian languages. The full texts and other materials stored at the repository comply with the open Creative Commons Attribution CC Attribution (CC BY) license.

   The test prototype of the repository will be presented, including its main capabilities, the structure of communities and collections, submissions, and feedback forms.

2. Seminar topic: “Workload Management System for SPD Online filter”

   Speaker: Nikita Greben

   Abstract:

   One of the key technical features of the SPD (Spin Physics Detector) is the triggerless data acquisition. The data acquisition system (DAQ) aggregates data from the detectors of the facility and organizes them into blocks for further primary processing. This approach allows achieving data arrival rate of up to 20 Gb/sec, with hundreds of petabytes of the annual volume of the collected data. To solve the primary problem of unscrambling events from the aggregated data stream before filtering them out, a specialised computing system «SPD Online filter» is being created.

   «SPD Online filter» will be a hardware-software complex for high-throughput processing of primary data coming from the detection system of the SPD Experiment. The hardware part will consist of a set of multi-core computing nodes, high-performance data storage systems, and a number of control servers; the software part will consist not only of application software, but also of a set of middleware that organize and execute multi-stage data processing steps.

   This talk will present a brief description of the architecture and implementation of a prototype workload management system that performs the generation of computational jobs for processing a block of data, the distribution of jobs to compute nodes, and the control of their execution.