Abstract

In modern Simultaneous Multi-Threading (SMT) processors, efficient management of shared resources like the rename register file is crucial for optimizing system throughput and utilization. The register file, being a critical shared resource, significantly influences performance by affecting how multiple threads access and use the available registers. When a few threads monopolize the register file, it can lead to a degradation in overall performance. Register capping is a method of limiting the number of rename registers each thread is allowed to use. The method has been used to improve performance. However, to understand the behavior of multiple threads sharing a register file and the effect that capping has on different combinations of programs, an analytical tool to observe the register file’s dynamics is necessary. To address this, this paper develops a theoretical model utilizing Markov Chain to analyze rename register utilization in SMT systems. The model proposed incorporates capping to examine the dynamics of register allocation and usage patterns among concurrent threads across different threads. By varying parameters such as cap value and consumption rates, the model reveals insight into the behavior of register sharing and its impact on overall system performance.

References

• H. Hirata, K. Kimura, S. Nagamine, Y. Mochizuki, A. Nishimura, Y. Nakase, T. Nishizawa, An elementary processor architecture with simultaneous instruction issuing from multiple threads, in: Proceedings of the 19th annual international symposium on Computer architecture, 1992, pp. 136–145.
• D. M. Tullsen, S. J. Eggers, H. M. Levy, Simultaneous multithreading: Maximizing on-chip parallelism, in: Proceedings of the 22nd annual international symposium on Computer architecture, 1995, pp. 392–403.
• M. Navarro, J. Feliu, S. Petit, M. E. G´omez, J. Sahuquillo, Synpa: Smt performance analysis and allocation of threads to cores in arm processors (2023). arXiv:2310.12786. URL https://arxiv.org/abs/2310.12786
• D. M. Tullsen, S. J. Eggers, J. S. Emer, H. M. Levy, J. L. Lo, R. L. Stamm, Exploiting choice: Instruction fetch and issue on an implementable simultaneous multithreading processor, in: Proceedings of the 23rd annual international symposium on Computer architecture, 1996, pp. 191–202.
• F. J. Cazorla, A. Ramirez, M. Valero, E. Fern´andez, Dynamically controlled resource allocation in smt processors, in: 37th International Symposium on Microarchitecture (MICRO-37’04), IEEE, 2004, pp. 171–182.
• S. Carroll, W.-M. Lin, Applied on-chip machine learning for dynamic resource control in multithreaded processors, Parallel Processing Letters 29 (03) (2019) 1950013. doi: 10.1142/S0129626419500130.
• Y. Zhang,W.-M. Lin, Efficient physical register file allocation in simultaneous multi-threading cpus, in: 33rd IEEE International Performance Computing and Communications Conference (IPCCC 2014), Austin, Texas, 2014, pp. 5–7.
• D. Kanter, Intel’s sandy bridge microarchitecture, https://www.realworldtech.com/sandy-bridge/ (Sep 2010).
• D. Kanter, Amd’s bulldozer microarchitecture, https://www.realworldtech.com/bulldozer (Aug 2010).
• J. L. Lo, S. S. Parekh, S. J. Eggers, H. M. Levy, D. M. Tullsen, Software-directed register deallocation for simultaneous multithreaded processors, IEEE Transactions on Parallel and Distributed Systems 10 (9) (1999) 922–933.
• T. Monreal, A. Gonz´alez, M. Valero, J. Gonz´alez, V. Vi˜nals, Dynamic register renaming through virtual-physical registers, Journal of Instruction Level Parallelism 2 (2000) 4–16.
• Y. Zhang, W.-M. Lin, Intelligent usage management of shared resources in simultaneous multi-threading processors, in: Proceedings of the International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA), The Steering Committee of The World Congress in Computer Science and Computer . . . , 2015, p. 27.
• H. G¨ung¨orer, G. K¨uc¸ ¨uk, Dynamic capping of physical register files in simultaneous multi-threading processors for performance, in: Computer and Information Sciences, Springer International Publishing, 2018, pp. 41–48.
• H. Zhan, V. S. Sheng, W.-M. Lin, Reinforcement learning-based register renaming policy for simultaneous multithreading cpus, Expert Systems with Applications 186 (2021) 115717. doi:https://doi.org/10.1016/j. eswa.2021.115717. URL https://www.sciencedirect.com/science/ article/pii/S095741742101099X
• A. Papoulis, S. Pillai, Probability, Random Variables and Stochastic Processes, McGraw-Hill series in electrical engineering: Communications and signal processing, Tata McGraw-Hill, 2002. URL https://books.google.com/books?id= g6eUoWOlcQMC
• K. Trivedi, Probability and Statistics with Reliability, Queuing and Computer Science Applications, Wiley, 2016.
• S. Carroll, W.-M. Lin, A queuing model for cpu functional unit and issue queue configuration, ArXiv abs/1807.08586 (2018). URL https://api.semanticscholar.org/CorpusID: 49904986
• J. Sharkey, D. Ponomarev, K. Ghose, M-sim: a flexible, multithreaded architectural simulation environment, Techenical report and Department of Computer Science and State University of New York at Binghamton (2005).
• Standard Performance Evaluation Corporation, Standard Performance Evaluation Corporation (SPEC), https:// www.spec.org, accessed: May 29, 2025 (2025).
• T. Sherwood, E. Perelman, G. Hamerly, B. Calder, Automatically characterizing large scale program behavior, ACM SIGPLAN Notices 37 (10) (2002) 45–57.