Aleksandr Drozd

Aleksandr Drozd

Auto-generated profile, based on publicly available information

Aleksandr Drozd

St. Petersburg State University of Veterinary Medicine

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 May 2021
Verified
Conference Paper
Open access
JDJens Domke
Jens Domke
EVEmil Vatai
Emil Vatai
ADAleksandr Drozd
Aleksandr Drozd
PCPeng ChenT
Peng ChenT
YOYosuke Oyama
Yosuke Oyama
SMSatoshi Matsuoka
Satoshi Matsuoka
Matrix engines or units, in different forms and affinities, are becoming a reality in modern processors; CPUs and otherwise. The current and dominant algorithmic approach to Deep Learning merits the commercial investments in these units, and deduced from the No. 1 benchmark in supercomputing, namely High Performance Linpack, one would expect an awakened enthusiasm by the HPC community, too. Hence, our goal is to identify the practical added benefits for HPC and machine learning applications by having access to matrix engines. For this purpose, we perform an in-depth survey of software stacks, proxy applications and benchmarks, and historical batch job records. We provide a cost-benefit analysis of matrix engines, both asymptotically and in conjunction with state-of-the-art processors. While our empirical data will temper the enthusiasm, we also outline opportunities to “misuse” these dense matrix-multiplication engines if they come for free.

24 April 2023 in The International Journal of High Performance Computing Applications
Verified
Journal Article
Closed access
SMSatoshi Matsuoka
Satoshi Matsuoka
JDJens Domke
Jens Domke
MWMohamed Wahib
Mohamed Wahib
ADAleksandr Drozd
Aleksandr Drozd
THTorsten Hoefler
Torsten Hoefler
In this thought-provoking article, we discuss certain myths and legends that are folklore among members of the high-performance computing community. We gathered these myths from conversations at conferences and meetings, product advertisements, papers, and other communications such as tweets, blogs, and news articles within and beyond our community. We believe they represent the zeitgeist of the current era of massive change, driven by the end of many scaling laws such as Dennard scaling and Moore’s law. While some laws end, new directions are emerging, such as algorithmic scaling or novel architecture research. Nevertheless, these myths are rarely based on scientific facts, but rather on some evidence or argumentation. In fact, we believe that this is the very reason for the existence of many myths and why they cannot be answered clearly. While it feels like there should be clear answers for each, some may remain endless philosophical debates, such as whether Beethoven was better than Mozart. We would like to see our collection of myths as a discussion of possible new directions for research and industry investment.

14 December 2023 in ACM Transactions on Architecture and Code Optimization
Verified
Journal Article
Open access
JDJens Domke
Jens Domke
EVEmil Vatai
Emil Vatai
BGBalazs Gerofi
Balazs Gerofi
YKYuetsu Kodama
Yuetsu Kodama
MWMohamed Wahib
Mohamed Wahib
SMSatoshi Matsuoka
Satoshi Matsuoka
Over the last three decades, innovations in the memory subsystem were primarily targeted at overcoming the data movement bottleneck. In this paper, we focus on a specific market trend in memory technology: 3D-stacked memory and caches. We investigate the impact of extending the on-chip memory capabilities in future HPC-focused processors, particularly by 3D-stacked SRAM. First, we propose a method oblivious to the memory subsystem to gauge the upper-bound in performance improvements when data movement costs are eliminated. Then, using the gem5 simulator, we model two variants of a hypothetical LARge Cache processor (LARC), fabricated in 1.5 nm and enriched with high-capacity 3D-stacked cache. With a volume of experiments involving a broad set of proxy-applications and benchmarks, we aim to reveal how HPC CPU performance will evolve, and conclude an average boost of 9.56× for cache-sensitive HPC applications, on a per-chip basis. Additionally, we exhaustively document our methodological exploration to motivate HPC centers to drive their own technological agenda through enhanced co-design.

1 September 2018
Verified
Conference Paper
Closed access
SSShweta Salaria
Shweta Salaria
ADAleksandr Drozd
Aleksandr Drozd
APArtur Podobas
Artur Podobas
SMSatoshi Matsuoka
Satoshi Matsuoka
Performance prediction of parallel applications across systems becomes increasingly important in today's diverse computing environments. A wide range of choices in execution platforms pose new challenges to researchers in choosing a system which best fits their workloads and administrators in scheduling applications to the best performing systems. While previous studies have employed simulation-or profile-based prediction approaches, such solutions are time-consuming to be deployed on multiple platforms. To address this problem, we use two collaborative filtering techniques to build analytical models which can quickly and accurately predict the performance of workloads across different multicore systems. The first technique leverages information gained from performance observed for certain applications on a subset of systems and use it to discover similarities among applications as well as systems. The second collaborative filtering based model learns latent features of systems and workloads automatically and use these features to characterize the performance of applications on different platforms. We evaluated both the methods using 30 workloads chosen from NAS Parallel Benchmarks, BOTS and Rodinia benchmarking suites on ten different systems. Our results show that such collaborative filtering methods can make predictions with RMSE as low as 0.6 and with an average RMSE of 1.6.

28 June 2024 in International Journal of Veterinary Medicine
Verified
Journal Article
Open access
PDF file
TKT. V. Kalyuzhnaya
T. V. Kalyuzhnaya
ADA. V. Drozd
A. V. Drozd
Existing methods for the determination of melamine in milk and its processed products are time-consuming to reproduce, require special equipment, training of personnel to work on this equipment, the use of chemical reagents and auxiliary equipment, often expensive. A separate area in the analysis of food products is spectroscopy in the middle infrared range, used to identify substances. The aim of the work was to record and analyze the infrared spectra of a standard sample of melamine in the middle spectral range using an NSAID prefix, a collapsible liquid cuvette and vaseline, as well as in tablets with KBr. The research was carried out in stages on the basis of the educational and research center for the examination of food and Animal Feed of the St. Petersburg State University of Veterinary Medicine in 2024. A standard sample of melamine was used as the research materials. At the first stage, the spectra of a standard sample of melamine were recorded using an infrared Fourier spectrometer "InfraLUM FT-08" and an NPVO prefix with a ZnSe crystal. At the next stage, the spectra of a standard sample of melamine in petroleum jelly were taken using a collapsible liquid cuvette, and then melamine pressed into tablets with KBr. As a result of the conducted studies, the presence of absorption bands of varying degrees of intensity in the IR spectra of the standard melamine sample was established, characterizing various types of fluctuations in the bonds of NH and CH groups. When comparing the IR spectra of melamine obtained at different stages of the study, a decrease in the intensity of the recorded absorption bands in different regions of the middle spectral range was revealed when using a liquid cuvette with vaseline and tablets with KBr. In addition, it was found that the IR spectrum of melamine obtained using the NSAID prefix exceeded the intensity of the absorption bands of similar spectra taken using KBr tablets and a liquid cuvette with vaseline.

28 June 2024 in International Journal of Veterinary Medicine
Verified
Journal Article
Open access
PDF file
ADA. V. Drozd
A. V. Drozd
PZP. S. Zhmurkina
P. S. Zhmurkina
The development and intensification of the dairy industry currently often leads to the appearance of counterfeit food products in circulation. Component adulteration of milk is widespread in relation to the most valuable protein component. To date, the problem is that the estimation of the amount of protein in milk and its processed products by arbitration methods is based on the determination of nitrogen and is currently imperfect, since when adulterated with nitrogencontaining substances such as melamine, urea gives false inflated values and does not allow for operational input and production control. We have analyzed the regulatory framework, scientific publications of domestic and foreign authors, review materials from open sources over the past 10 years on the issues of falsification of the protein component of milk and its processed products, methods of identification of adulterating components. As a result of the analysis of regulatory documents and scientific literature, it was established that the existing methods approved in regulatory documents are laborious to reproduce, require special equipment, training of personnel to work on this equipment, the use of chemical reagents and auxiliary equipment, often expensive. The modern method of identifying melamine using infrared analyzers will allow to establish its quantitative values with great accuracy in the presence of appropriate calibration dependencies, in many ways accelerate and simplify the procedure for delivery and acceptance of raw milk while maintaining quality and safety indicators at a high level in accordance with regulatory documents.