Events

MLIT results of the year at the solemn seminar

On 27 December, a solemn meeting of MLIT staff members was held at the Meshcheryakov Laboratory of Information Technologies with the participation of the JINR Directorate. At the meeting, MLIT Director Sergei Shmatov and MLIT Scientific Leader Vladimir Korenkov presented the main results of the Laboratory’s activities for 2024 and plans for the coming year. The winners of youth grants and prizes were announced.

JINR Director Grigory Trubnikov welcomed the audience: “I believe that the currency of this decade is data and information. In this sense, it is symbolic that today we have met in our Laboratory of Information Technologies. You are our major information platform. You promote and support science thanks to the telecommunication channels created, as well as absolutely fantastic opportunities for data analysis, storage and transfer. Through data, through Digital JINR, you bring us closer together. I congratulate you all on the upcoming New Year. I am grateful to you for the past year, for the results, for your work, for your talents. I wish the Laboratory to grow steadily, gain specialists from the Member States and be at the forefront of the Institute’s development.”

JINR Scientific Leader Victor Matveev and JINR Vice-Director Vladimir Kekelidze expressed congratulations and warm wishes for scientific success in the coming year.

“MLIT is actively developing, now it is a sought-after leader in the field, one of the organizers of the Institute’s activities. I would like to note the great contribution of the MLIT Directorate in how they maintain an internal atmosphere of friendly interaction in the name of the objectives that the Laboratory sets for itself,” Victor Matveev pointed out.

“With NICA’s upcoming launch, next year will be special for the entire Institute and for your Laboratory, and we expect even more power from you. This is another challenge that awaits you in 2025, and there are many such challenges, but we are confident that you will overcome them all,” Vladimir Kekelidze underlined.

MLIT Scientific Leader Vladimir Korenkov opened the summing up of the Laboratory’s output. First of all, in his talk, he noted the large research infrastructure project of the Laboratory, the Multifunctional Information and Computing Complex (MICC). In 2024, a multitude of work to enhance the MICC engineering infrastructure was done. The necessary upgrade of the equipment that ensures the power supply and cooling of the computing complex was carried out. The modernization was performed without stopping the operation of all active equipment and at the same time promptly, with little loss of time and performance. “This work is not yet complete, but MLIT has done a significant part of the task, thereby ensuring the possibility of actively updating the system so that we can install new equipment and provide the even more reliable operation of the MICC,” the Laboratory’s Scientific Leader noted.

Active work on the automation and monitoring of the MICC engineering infrastructure started. This is a sophisticated task due to the fact that the computing complex is equipped with a large number of various equipment. “Somewhere there are ready-made monitoring and control tools, somewhere there is nothing at all. This entailed the installation of individual modules, sensors and other necessary equipment to gradually collect all this information for the timely monitoring of the operability of our entire engineering infrastructure,” Vladimir Korenkov said.

During the year, work to install new servers in the MICC was performed. 12 Asus servers and two Huawei servers were installed. In December, 30 SILA CP2-1627 servers and six SILA CP1-162 servers, two SILA CK3 630A-32Q network switches were placed in server cabinets 404 and 416. Eight ASUS RS720-E10-RS12 servers were installed in cabinet 405 within the development of the disk storage of the MICC cloud component for the JUNO experiment.

In 2024, the Laboratory successfully fulfilled its obligations to ensure the reliable and sustainable operation of the major systems, namely, Tier1, Tier2, the cloud infrastructure and the “Govorun” supercomputer. In total, within the year, the MICC ensured the execution of over 10 million tasks on the Tier1 and Tier2 grid infrastructure, as well as about 3.8 million tasks on the “Govorun” supercomputer. At the same time, in 2024, JINR Tier1 for CMS was ranked first among all sites of the experiment worldwide in terms of the CPU time for data processed. The JINR Tier2 output is the highest in the Russian grid segment (Russian Data Intensive Grid, RDIG).

Speaking about objectives for the next year, Vladimir Korenkov noted the provision of high-speed network communications between Russia and the outside world. As known, Russia’s participation in CERN was not prolonged. Previously, the leading role in ensuring the operation of external networks belonged to the National Research Center “Kurchatov Institute” (NRC KI). Now the management of this infrastructure, as well as the majority of funding, should be taken over by the Laboratory, with the further participation of NRC KI and the National Research Computer Network of Russia. The RDIG-M consortium is being transformed to develop the distributed computer infrastructure that integrates key centers participating in Russian megascience projects. JINR, together with NRC KI and the Institute for System Programming of the Russian Academy of Sciences, should become the core of this integration.

Among the tasks for the coming year, Vladimir Korenkov also indicated the development of approaches in monitoring the computing, engineering and network infrastructures. The operational monitoring of the computing center entails additions to the analytical component, and here the application of digital twin technology, which will allow analyzing the functioning of the whole infrastructure, effectively optimizing and developing it, offers great prospects. This work has already started at the Laboratory and was included in a PhD thesis, which was successfully defended by MLIT researcher Daria Priakhina at St. Petersburg State University at the end of this year.

“A lot has been done to develop the JINR Digital EcoSystem. A major role in this belongs to Sergey Belov, who brilliantly defended his PhD thesis at JINR at the end of December. I would also like to note the deep reworking of the database information system of JINR staff members, PIN-2, which was implemented by Sergei Kuniaev’s team, and it is already operating in test mode. At the beginning of the next year, the JINR scientific publication repository, developed by Irina Filozova’s group, will be put into operation,” Vladimir Korenkov said.

On 6 December 2024, there was received a license for a new educational master’s program at the Dubna branch of Moscow State University in direction 01.04.02 Applied Mathematics and Computer Science, and the first student intake is planned for next year. “This is gratifying and highly important for us, since we get a good opportunity to train high-level specialists already in the master’s program, specialists whom we will know very well, and involve them in solving the Laboratory’s tasks,” MLIT Scientific Leader underlined. A highly good influx of young scientists is provided by the already traditional MLIT IT Schools. In 2024, 38 out of 58 students of the Autumn IT School chose topics for their theses based on MLIT activities. In total, 18 people became JINR employees in two years of the School series.

The Workshop on Mathematical Problems in Quantum Information Technologies (MPQIT-2024) and the International Conference on Mathematical Modeling and Computational Physics (MMCP-2024) were among the highlights of 2024 held by the Laboratory. The following events are planned for next year: anniversary seminar to mark the 95th birthday of Nikolay Nikolaevich Govorun (18 March), 11th International Conference on Distributed Computing and Grid Technologies in Science and Education GRID’2025 (July), as well as workshops in Egypt and South Africa.

The overview of the Laboratory’s achievements in 2024 was continued in the report by MLIT Director Sergei Shmatov, who reminded that for MLIT, as well as for the Institute, this year was the first in the implementation of the new Seven-Year Plan for the Development of JINR, as well as within the updated structure of the JINR Topical Plan. In the TP, MLIT is represented by the large research infrastructure project “Multifunctional Information and Computing Complex”, two projects and two activities within the theme “Information Technology” (06-6-1119). We are talking about the projects “Mathematical Methods, Algorithms and Software for Modeling Physical Processes and Experimental Facilities, Processing and Analyzing Experimental Data” and “Methods of Computational Physics for Studying Complex Systems”, as well as activities within this theme.

The result of work in these directions in 2024 was the publication of over 150 scientific papers, three monographs and about 100 articles within international collaborations. MLIT staff members delivered over 140 reports at international and Russian conferences.

“The priority objective for MLIT is to provide computational support for the elaboration and implementation of the physical program of large-scale research projects conducted at JINR and other scientific centers with the Institute’s participation,” Sergei Shmatov pointed out. He said that to solve the tasks, three directions of work were identified at the Laboratory. These are the modeling of physical processes and facilities, data reconstruction and analysis along with the development of corresponding software, as well as the organization of an information and analytical environment for physical experiments, which enables to analyze and manage data flows, perform visualization, including data models and databases.

In light of the upcoming launch of the NICA collider in 2025 and the start of modeling data accumulation within the SPD experiment, Sergei Shmatov noted among the important results the creation of a prototype of a distributed experimental data processing and analysis system on top of the PANDA system at MLIT. The prototype was successfully tested on the integrated resources of MLIT and Konstantinov Petersburg Nuclear Physics Institute. A system for primary data processing for the SPD experiment, Online Filter, was developed, and a testbed for developing and debugging the data acquisition system of the experiment was created in the Laboratory together with the Dzhelepov Laboratory of Nuclear Problems (DLNP).

The data processing and analysis system based on the DIRAC distributed computing platform is successfully operating, working with BM@N experiment data, MPD and SPD collider experiment modeling data. The main users of the distributed information and computing environment (DICE) based on DIRAC, which integrates the clouds of the JINR Member States’ organizations, are currently the Baikal-GVD, JUNO and NOvA neutrino experiments. Both the Laboratory and the Institute have accumulated significant experience in creating information systems and databases for key experiments at the LHC at CERN. “In light of the suspension of Russia’s participation at CERN, through cooperation in the format of JINR associate personnel, we managed to preserve a group of people who worked in Russian institutes and made a key contribution to ensuring our work with CERN. Now we will be able to transfer their knowledge, experience and developments to the tasks of the NICA megascience project,” MLIT Director remarked.

The Laboratory continues to actively participate in the collaborations of LHC experiments. Here, two results in experimental data analysis can be highlighted. Together with BLTP and VBLHEP, new constraints on the cross sections of interaction of dark matter particles with Standard Model particles were obtained in the CMS experiment when processing LHC RUN2 data at 13 TeV. In cooperation with SINP MSU and ITEP, promising results were obtained in searching for a resonance with a mass of 28 GeV in the two-muon decay mode using CMS data in RUN2.

At the same time, Sergei Shmatov noted the Laboratory’s ongoing work in the field of LHC open data analysis, the development of systems for processing and analyzing Baikal-GVD, OLVE-HERO and TAIGA data, the elaboration of a prototype model of a digital tracking calorimeter (GATE package) on top of GEANT4 on the HybriLIT platform, algorithms for searching clusters in SPD, the adjustment of the GEM detectors of the BM@N facility and tracking for CMS.

A whole range of outstanding works on developing computational physics methods to study complex physical systems were performed by MLIT specialists under the supervision of Jan Busa. Several of them were included in the JINR official report to the Russian Academy of Sciences. First of all, these are works on calculating various applications of science, which are aimed at finding stable energy states in the Ising model, as well as at investigating collective effects in the atomic nucleus model. Together with colleagues from South Africa, a numerical study of spherically symmetric standing waves in a ball, considered as an approximation of weakly radiating oscillons in the ϕ4 theory, was carried out.

In the field of life sciences, MLIT focuses on the application of artificial intelligence technologies to solve various tasks in agriculture, and a working application for detecting plant diseases has already been created. Jointly with LRB, web services of the BIOHLIT project are developed on the ML/DL/HPC ecosystem of the HybriLIT platform. Work on the application of AI technologies and Earth remote sensing data to predict the state of the environment is also underway.

In cooperation with BLTP, Josephson junctions are modeled on the ML/DL/HPC ecosystem of the HybriLIT platform. Within the interlaboratory working group on modeling radiation conditions at the NICA accelerator complex, work to assess the radiation situation in the premises of the NICA temporary control room (a joint project of MLIT, LRB, VBLHEP, FLNP and DRS) is carried out. To ensure high-energy neutron dosimetry, two methods for reconstructing the neutron spectrum based on readings from the Bonner multisphere spectrometer were elaborated.

In total, in 2024, the Laboratory conducted scientific and technical cooperation with 20 countries and 43 scientific and educational organizations, as well as with 52 Russian centers. Scientific findings were presented at 36 joint laboratory seminars (74 reports) and 15 seminars of the Scientific Department of Computational Physics (20 reports). Currently, 318 employees, including 62 candidates of science and 21 doctors of science, work at MLIT. In 2024, 14 people were hired to the Laboratory.

Speaking about plans for 2025, first of all, Sergei Shmatov indicated preparations for processing MPD experiment data, modernizing the MICC and attracting new personnel. The concept of developing the network infrastructure, integrating resources within the RDIG-M consortium and cooperating with Russian centers and foreign partners will be worked out. It is planned to develop data processing and analysis systems (NICA and LHC experiments, neutrino program), information and analytical systems, data processing and analysis methods and algorithms (especially promising ones based on AI), computational physics methods to study complex systems, as well as the JINR Digital EcoSystem. In addition, in 2025, preparations for celebrating the 60th anniversary of the Laboratory will start.

The results of MLIT activities were concluded by a ceremonial presentation of the winners of grants from the JINR Association of Young Scientists and Specialists and of the M.G. Meshcheryakov and N.N. Govorun prizes. Andrey Kondratyev, Igor Pelevanyuk, Ivan Sokolov received AYSS grants for young researchers, and Andrey Evlanov and Anna Ilina received AYSS grants for young specialists. Dina Badreeva, Nikita Greben, Alexey Didorenko, Ilyas Satyshev, Alexander Khmelev and Egor Tsamtsurov became the winners of the Meshcheryakov prize. Anastasia Anikina, Natalya Belyakova, Ilya Kalagin, Ivan Kashunin and Daria Stankus received the Govorun prize.