Impact of Data-Oriented and Object-Oriented Design on Performance and Cache Utilization with Artificial Intelligence Algorithms in Multi-Threaded CPUs

Impact of Data-Oriented and Object-Oriented Design on Performance and Cache Utilization with Artificial Intelligence Algorithms in Multi-Threaded CPUs Gabriel M. Arantes ∗, Richard F. Pinto ∗, Bruno L. Dalmazo ∗, Eduardo N. Borges ∗, Viviane L. D. de Mattos ∗, Rafael A. Berri ∗ ∗Federal University of Rio Grande (FURG), Rio Grande, Brazil Giancarlo Lucca † †Catholic University of Pelotas (UCPel), Pelotas, Brazil Fabian C. Cardoso ‡ ‡University of Rio Verde (UniRV), Rio Verde, Brazil Abstract—The growing performance gap between multi-core CPUs and main memory necessitates hardware-aware software design paradigms. This study provides a comprehensive per- formance analysis of Data-Oriented Design (DOD) versus the traditional Object-Oriented Design (OOD), focusing on cache utilization and efficiency in multi-threaded environments. We de- veloped and compared four distinct versions of the A* search al- gorithm: single-threaded OOD (ST-OOD), single-threaded DOD (ST-DOD), multi-threaded OOD (MT-OOD), and multi-threaded DOD (MT-DOD). The evaluation was based on metrics including execution time, memory usage, and CPU cache misses. In multi-threaded tests, the DOD implementation demonstrated considerable performance gains, with faster execution times and a lower number of raw system calls and cache misses. While OOD occasionally showed marginal advantages in memory usage or percentage-based cache miss rates, DOD’s efficiency in data-intensive operations was more evident. Furthermore, our findings reveal that for a fine-grained task like the A* algorithm, the overhead associated with thread management led to single-threaded versions significantly outperforming their multi-threaded counterparts in both paradigms. We conclude that even when performance differences appear subtle in simple algorithms, the consistent advantages of DOD in critical metrics highlight its foundational architectural superiority, suggesting it is a more effective approach for maximizing hardware efficiency in complex, large-scale AI and parallel computing tasks. Index Terms—data-oriented design, object-oriented design, multi-threading, performance optimization, cache efficiency I. INTRODUCTION In the modern world, computers are extremely important and are used in nearly every aspect of daily life. They are continuously advancing, with constant improvements in their performance and capabilities, making their efficient use increasingly necessary and important [1]. The authors would like to thank FAPERGS (24/2551-0001396-2, 23/2551-0000773-8), CNPq (305805/2021-5) and FAPERGS/CNPq (23/2551- 0000126-8). In particular, the CPU (Central Processing Unit) is evolving at a rapid pace, with annual improvements in processing speed and memory storage, known in the CPU context as cache. The cache has different levels of speed and size, whereas other components have not evolved at the same rate [2]. Storage speed outside the CPU is significantly slower, and in this paper, we will focus on RAM, which is approximately 100 times slower than the CPU [3]. A solution to mitigate this performance gap may lie in the more effective use of the cache through the adoption of Data- Oriented Design/Data-Oriented Programming (DOD/DOP) patterns, as opposed to the current industry standard of Object- Oriented Design/Object-Oriented Programming (OOD/OOP). OOP is not efficient in terms of cache usage, as it has a greater impact on performance and scalability, whereas DOD does not exhibit these issues [4]. OOP consists of using object classes for data manipulation and well-encapsulated functions with multiple abstraction layers [5]. In contrast, DOD makes better use of data by separating it from the code, thus improving data access and allowing for greater scalability and easier code maintenance [6]. The remainder of the paper is organized as follows. Section II presents a literature review. The methodology is presented in Section III, and details of its results in Section IV. Finally, in Section V, some final remarks are made, and directions for future research are indicated. II. LITERATURE REVIEW To develop this paper, a study on the subject was conducted by reviewing scientific literature, and this section presents the results of that study. A. Articles and Books Computer architecture is a vast and constantly evolving field, with numerous significant contributions over the decades. arXiv:2512.07841v1 [cs.AI] 22 Nov 2025 In this context, the book ”Computer Architecture” by Hen- nessy and Patterson [7] highlights the importance of memory in computing, particularly the role of cache in enhancing overall system performance. They argue that systems with more efficient caches tend to perform better overall, even though they do not provide a study specifically focused on cache, supporting their assertion through a broad analysis of computer architectures. The relevance of cache is corroborated by various other studies and publications. For example, Culler et al. [8], in ”Parallel C

…(Full text truncated)…

This content is AI-processed based on ArXiv data.