METHOD FOR GENERATING TOPOLOGICAL CONSTRAINTS OF COMPUTATIONAL STRUCTURES FOR RECONFIGURABLE COMPUTING SYSTEMS
Abstract
For reconfigurable computing systems based on FPGAs, efficient application programs are parallel-pipeline programs that achieve real performance exceeding 50% of the peak. This article addresses the problem of reducing the development time of such programs. The computational structures of these programs utilize a large volume of FPGA resources operating at high clock frequencies. However, simultaneously maximizing both the amount of FPGA resources used and the clock frequency presents a certain contradiction: as resource utilization increases, the placement flexibility of the functional units of the computational structures decreases, and the FPGA switching matrix fails to provide the required signal propagation characteristics when routing information channels between them. Moreover, in modern CAD tools, placement and routing algorithms consider only the architectural and geometric features of the FPGA. Therefore, when a large number of specialized primitives with very limited placement flexibility are used, achieving high clock frequencies in automatic synthesis mode becomes virtually impossible. To address this problem, it is also necessary to consider the information dependencies between the functional units of the computational structures, but the nature of these dependencies in tasks from different subject areas can vary significantly. As a result, developers are often forced to manually place the functional units of the computational structures on the FPGA by creating script-based instructions for topological constraints. In earlier generations of FPGAs, the time required to generate topological constraints was acceptable, as they typically contained only a few hundred specialized primitives. However, in modern FPGAs, the number of such primitives reaches several thousand or even tens of thousands, significantly increasing the development time of efficient application programs. The proposed method makes it possible to automate the process of developing topological constraints for computational structures. The research was carried out during the development of application programs for solving a range of problems based on FFT, AES, and LU decomposition algorithms for the reconfigurable computer “Tertius-2.” As a result of significantly reducing the time required for optimization iterations of computational structures, the total synthesis time was reduced by up to three times.
References
1. Kalyaev I.A., Dordopulo A.I., Levin I.I., Fedorov A.M. Razvitie otechestvennykh mnogokristal'nykh rekonfiguriruemykh vychislitel'nykh sistem: ot vozdushnogo k zhidkostnomu okhlazhdeniyu [Develop-ment of domestic multi-crystal reconfigurable computing systems: from air to liquid cooling ], Tr. SPIIRAN [Proceedings of SPIIRAS]. St. Petersburg: Izd-vo SPIIRAN FGBUN Sankt-Peterburgskiy institut informatiki i avtomatizatsii RAN, 2017, No. 1 (50), pp. 5-31. DOI: 10.15622/sp.50.1.
2. Kalyaev A.V., Levin I.I. Modul'no-narashchivaemye mnogoprotsessornye sistemy so strukturno-protsedurnoy organizatsiey vychisleniy [Modularly scalable multiprocessor systems with structural-procedural organization of computations]. Moscow: Yanus-K, 2003, 380 p.
3. Kalyaev I.A., Levin I.I. Rekonfiguriruemye vychislitel'nye sistemy na osnove PLIS [Reconfigurable computing systems based on FPGAs]. Rostov-on-Donu: Izd-vo YuNTS RAN, 2022, 506 p. ISBN 978-5-4358-0232-0.
4. Vivado Overview. Available at: https://www.amd.com/en/products/software/adaptive-socs-and-fpgas/vivado.html (accessed 08 November 2024).
5. FPGA Design Software – Quartus Prime. Available at: https://www.intel.com/content/www/us/en/ products/details/fpga/development-tools/quartus-prime.html (accessed 08 November 2024).
6. Alekseev K.N., Sorokin D.A., Leont'ev A.L. Metodika sozdaniya topologicheskikh ogranicheniy pri vysokoy utilizatsii resursov PLIS [Methodology for creating topological constraints with high utilization of FPGA resources], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2022, No. 4, pp. 200-212.
7. Dichenko A.A., Sorokin D.A., Levin I.I. Printsipy razmeshcheniya vychislitel'nykh struktur na PLIS rekonfiguriruemykh vychislitel'nykh sistem [Principles of placing computing structures on FPGAs of reconfigurable computing systems], Deutsche Internationale Zeitschrift für Zeitgenössische Wissen-schaft, 2024, No. 79, pp. 54-63. DOI: 10.5281/zenodo.11127391. EDN TBKWEJ.
8. Alekseev K.N., Sorokin D.A. Optimizatsiya vychislitel'nykh struktur pod arkhitekturu PLIS XILINX [Optimization of computing structures for the XILINX FPGA architecture], FPGA-Systems Magazine: FSM: № ALFA (state_0), 2023, 166 p., pp. 75-82. Available at: https://fpga-systems.ru/fs_fsm/ state_0/fsm_state_0.pdf (accessed 17 September 2024).
9. Vivado Design Suite User Guide: Using Constraints (UG903). Available at: https://docs.amd.com /r/en-US/ug903-vivado-using-constraints (accessed 20 November 2024).
10. Vivado Design Suite User Guide: Design Analysis and Closure Techniques (UG906). Available at: https://docs.amd.com/r/en-US/ug906-vivado-design-analysis (accessed 20 November 2024).
11. Intel Quartus Prime Standard Edition User Guide: Design Constraints. Available at: https://www.intel.com/content/www/us/en/docs/programmable/683492/18-1/constraining-designs.html (accessed 21 November 2024).
12. Speed Grade. Available at: https://www.intel.com/content/www/us/en/docs/programmable/683703/17-1/speed-grade.html (accessed 03 December 2024).
13. Xilinx WP380 Xilinx Stacked Silicon Interconnect Technology Delivers Breakthrough FPGA Capacity, Bandwidth, and Power Efficiency (WP380). Available at: https://docs.amd.com/v/u/en-US/wp380_Stacked_Silicon_Interconnect_Technology (accessed 03 December 2024).
14. Virtex-6 Family Overview (DS150). Available at: https://docs.amd.com/v/u/en-US/ds150 (accessed 10 December 2024).
15. Kalyaev I.A., Levin I.I., Semernikov E.A., Dordopulo A.I. Rekonfiguriruemye vychislitel'nye sistemy na osnove PLIS semeystva Virtex-6 [Reconfigurable computing systems based on FPGAs of the Virtex-6 family], Parallel'nye vychislitel'nye tekhnologii (PaVT'2011): Tr. mezhdunarodnoy nauchnoy konfer-entsii, Moskva, 28 marta – 01 2011 goda [Proceedings of the international scientific conference, Mos-cow, March 28 – January 2011], Responsible for the release: L.B. Sokolinskiy, K.S. Pan. Moscow: Iz-datel'skiy tsentr YuUrGU, 2011, pp. 203-210. EDN TBLWND.
16. 7 Series FPGAs Data Sheet: Overview (DS180). Available at: https://docs.amd.com/v/u/en-US/ds180_7Series_Overview (accessed 10 December 2024).
17. UltraScale Architecture and Product Data Sheet: Overview (DS890). Available at: https://docs.amd.com/v/u/en-US/ds890-ultrascale-overview (accessed 10 December 2024).
18. Versal Architecture and Product Data Sheet: Overview (DS950). Available at: https://docs.amd.com/v/u/en-US/ds950-versal-overview (accessed 11 December 2024).
19. Vivado Design Suite (WP416). Available at: https://docs.amd.com/v/u/en-US/wp416-Vivado-Design-Suite (accessed 10 December 2024).
20. Defining Implementation Strategies • Vivado Design Suite User Guide: Implementa-tion (UG904). Available at: https://docs.amd.com/r/en-US/ug904-vivado-implementation/Defining-Implementation-Strategies (accessed 10 December 2024).
21. Tertsius-2 | NITS super-EVM i neyrokomp'yuterov [Tertsius-2 | Research Center of Supercomputers and Neurocomputers]. Available at: https://superevm.ru/index.php?page=tertsius-2 (accessed 14 Janu-ary 2025).
