Lukasz Szustak
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Lukasz Szustak has a total of 53 co-authors. Here are the top co-authors by number of shared publications:
- Roman Wyrzykowski
Częstochowa University of Technology
17 shared publications - Krzysztof Rojek
Częstochowa University of Technology
11 shared publications - Marcin Lawenda
Poznań Supercomputing and Networking Center
6 shared publications - Pawel Bratek
Czȩstochowa University of Technology, Poland
6 shared publications - Kamil Halbiniak
Częstochowa University of Technology
5 shared publications - Tomasz Olas
Częstochowa University of Technology
5 shared publications
Latest Publications
- 2 July 2026
- Published by Springer Nature
- p. 624-630
Boosting Performance of Counting Queries in Machine Learning Applications with a ccNUMA-Aware Implementation
- 2 July 2026
- Published by Springer Nature
- p. 624-630
Boosting Performance of Counting Queries in Machine Learning Applications with a ccNUMA-Aware Implementation
- Open Access
- 17 November 2025
- Published by International Conference on Information Systems Development
Optimizing the DFS-based Strategy for Efficient Execution of Counting Queries in Machine Learning Applications
Counting queries are fundamental operations widely used in machine learning applications. This paper focuses on optimizing their execution by introducing algorithmic enhancements to the bitmap-based counting query strategy that relies on a Depth-First Search (DFS) traversal. The proposed approach is evaluated through a benchmark involving the execution of random query streams across multiple test datasets. The experimental results demonstrate a significant speedup, with execution times reduced by factors ranging from 1.26× to 2.25×. Furthermore, potential directions for further improving the performance of counting queries on modern high-performance computing (HPC) systems are discussed.
- 17 November 2025
- Published by International Conference on Information Systems Development
Open AccessOptimizing the DFS-based Strategy for Efficient Execution of Counting Queries in Machine Learning Applications
Counting queries are fundamental operations widely used in machine learning applications. This paper focuses on optimizing their execution by introducing algorithmic enhancements to the bitmap-based counting query strategy that relies on a Depth-First Search (DFS) traversal. The proposed approach is evaluated through a benchmark involving the execution of random query streams across multiple test datasets. The experimental results demonstrate a significant speedup, with execution times reduced by factors ranging from 1.26× to 2.25×. Furthermore, potential directions for further improving the performance of counting queries on modern high-performance computing (HPC) systems are discussed.
- 8 November 2025
- Published by Elsevier in Future Generation Computer Systems
- Vol. 177
Evaluating AMD EPYC CPU architectures on CFD applications
- 8 November 2025
- Published by Elsevier in Future Generation Computer Systems
- Vol. 177
Evaluating AMD EPYC CPU architectures on CFD applications
- 5 July 2025
- Published by Springer Nature
- p. 151-166
Advances in Adapting Memory-Bound CFD Computations to RISC-V Multicore Architecture
- 5 July 2025
- Published by Springer Nature
- p. 151-166
Advances in Adapting Memory-Bound CFD Computations to RISC-V Multicore Architecture
- 1 April 2025
- Published by Springer Nature
- p. 63-78
Efficient Allocation of Image Recognition and LLM Tasks on Multi-GPU System
- 1 April 2025
- Published by Springer Nature
- p. 63-78
Efficient Allocation of Image Recognition and LLM Tasks on Multi-GPU System
- 21 March 2025
- Published by Elsevier in Future Generation Computer Systems
- Vol. 169
Prediction model of performance–energy trade-off for CFD codes on AMD-based cluster
- 21 March 2025
- Published by Elsevier in Future Generation Computer Systems
- Vol. 169
Prediction model of performance–energy trade-off for CFD codes on AMD-based cluster
- Open Access
Efficient allocation of image recognition and LLM tasks on multi-GPU system
This work is concerned with the evaluation of the performance of parallelization of learning and tuning processes for image classification and large language models. For machine learning model in image recognition, various parallelization methods are developed based on different hardware and software scenarios: simple data parallelism, distributed data parallelism, and distributed processing. A detailed description of presented strategies is given, highlighting the challenges and benefits of their application. Furthermore, the impact of different dataset types on the tuning process of large language models is investigated. Experiments show to what extent the task type affects the iteration time in a multi-GPU environment, offering valuable insights into the optimal data utilization strategies to improve model performance. Furthermore, this study leverages the built-in parallelization mechanisms of PyTorch that can facilitate these tasks. Furthermore, performance profiling is incorporated into the study to thoroughly evaluate the impact of memory and communication operations during the training/tuning procedure. Test scenarios are developed and tested with numerous benchmarks on the NVIDIA H100 architecture showing efficiency through selected metrics.
Open AccessEfficient allocation of image recognition and LLM tasks on multi-GPU system
This work is concerned with the evaluation of the performance of parallelization of learning and tuning processes for image classification and large language models. For machine learning model in image recognition, various parallelization methods are developed based on different hardware and software scenarios: simple data parallelism, distributed data parallelism, and distributed processing. A detailed description of presented strategies is given, highlighting the challenges and benefits of their application. Furthermore, the impact of different dataset types on the tuning process of large language models is investigated. Experiments show to what extent the task type affects the iteration time in a multi-GPU environment, offering valuable insights into the optimal data utilization strategies to improve model performance. Furthermore, this study leverages the built-in parallelization mechanisms of PyTorch that can facilitate these tasks. Furthermore, performance profiling is incorporated into the study to thoroughly evaluate the impact of memory and communication operations during the training/tuning procedure. Test scenarios are developed and tested with numerous benchmarks on the NVIDIA H100 architecture showing efficiency through selected metrics.
- 14 April 2024
- Published by Springer Nature
- p. 327-333
Parallel Auto-Scheduling of Counting Queries in Machine Learning Applications on HPC Systems
- 14 April 2024
- Published by Springer Nature
- p. 327-333
Parallel Auto-Scheduling of Counting Queries in Machine Learning Applications on HPC Systems
- 14 January 2024
- Published by Wiley in Concurrency and Computation: Practice and Experience
- Vol. 36 (10)
Special Issue on the pervasive nature of HPC (PN‐HPC)
Summary: This special issue on the Pervasive Nature of HPC (PN‐HPC) collects an extension of the most valuable works presented at the sixth Workshop on Models, Algorithms and Methodologies for Hybrid Parallelism in New HPC Systems (MAMHYP‐22), held in Gdansk (Poland) in September 2022, jointly with the 14th conference on Parallel Processing and Applied Mathematics (PPAM‐22). New original papers related to the workshop themes are also included. The final aim is to provide a glimpse of the current state of knowledge related to the development of efficient methodologies and algorithms for HPC systems with multiple forms of parallelism.
- 14 January 2024
- Published by Wiley in Concurrency and Computation: Practice and Experience
- Vol. 36 (10)
Special Issue on the pervasive nature of HPC (PN‐HPC)
Summary: This special issue on the Pervasive Nature of HPC (PN‐HPC) collects an extension of the most valuable works presented at the sixth Workshop on Models, Algorithms and Methodologies for Hybrid Parallelism in New HPC Systems (MAMHYP‐22), held in Gdansk (Poland) in September 2022, jointly with the 14th conference on Parallel Processing and Applied Mathematics (PPAM‐22). New original papers related to the workshop themes are also included. The final aim is to provide a glimpse of the current state of knowledge related to the development of efficient methodologies and algorithms for HPC systems with multiple forms of parallelism.
- 3 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 11 (2) , 2135-2146
Large-Scale Parallelization of Human Migration Simulation
Forced displacement of people worldwide, for example, due to violent conflicts, is common in the modern world, and today more than 82 million people are forcibly displaced. This puts the problem of migration at the forefront of the most important problems of humanity. The Flee simulation code is an agent-based modeling tool that can forecast population displacements in civil war settings, but performing accurate simulations requires nonnegligible computational capacity. In this article, we present our approach to Flee parallelization for fast execution on multicore platforms, as well as discuss the computational complexity of the algorithm and its implementation. We benchmark parallelized code using supercomputers equipped with AMD EPYC Rome 7742 and Intel Xeon Platinum 8268 processors and investigate its performance across a range of alternative rule sets, different refinements in the spatial representation, and various numbers of agents representing displaced persons. We find that Flee scales excellently to up to 8192 cores for large cases, although very detailed location graphs can impose a large initialization time overhead.
- 3 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 11 (2) , 2135-2146
Large-Scale Parallelization of Human Migration Simulation
Forced displacement of people worldwide, for example, due to violent conflicts, is common in the modern world, and today more than 82 million people are forcibly displaced. This puts the problem of migration at the forefront of the most important problems of humanity. The Flee simulation code is an agent-based modeling tool that can forecast population displacements in civil war settings, but performing accurate simulations requires nonnegligible computational capacity. In this article, we present our approach to Flee parallelization for fast execution on multicore platforms, as well as discuss the computational complexity of the algorithm and its implementation. We benchmark parallelized code using supercomputers equipped with AMD EPYC Rome 7742 and Intel Xeon Platinum 8268 processors and investigate its performance across a range of alternative rule sets, different refinements in the spatial representation, and various numbers of agents representing displaced persons. We find that Flee scales excellently to up to 8192 cores for large cases, although very detailed location graphs can impose a large initialization time overhead.