V.M. Chernikov – Ph.D.(Eng.), Main Designer – Head of Department, RC «Module» (Moscow)

E-mail: tchern@module.ru

P.E. Viksne – Head of Sector, RC «Module» (Moscow) E-mail: pvixne@module.ru

This paper represents the approaches to build multicore heterogeneous SoC of over teraflop performance with use of NeroMatrix family SoC NM6408MP as prototype. To achieve more performance and power efficiency it is proposed as follows: usage of new technology (14 nm and bellow), increasing of processor cores on chip, modification of these processor cores to take more computing power with help of new hardware adding and more problem-oriented specialization. The development of computing systems with a capacity of up to several TFLOPS will enable the domestic industry to produce competitive digital signal processing equipment for radar and hydroacoustic systems. In addition, this equipment can effectively solve other important tasks, such as the implementation of neural networks of in-depth training.

1. Henessy J.L., Patterson D.A. Computer Architecture: A Quantitative Approach. Fifth Edition. Elvister Science. 2012. 856 p.
2. The Future of Computing Performance: Game Over or Next Level / Ed. by S.H. Fuller, L.I. Millett. Committee on Sustanding Growth in Computing Performance. National Research Council. National Academy of Science. 2010. 180 p.
3. Multiprocessor System-on-Chip / Ed. by M. Hübner, J. Becker. Hardware Design and Tool Integration. Springer Science + Business Media. 2011. 270 p.
4. Shahbahrami A., Juurlink B., Vassiliadis S. A Comparison Between Processor Architectures for Multimedia Applications // Proc. of 15th Annual Workshop on Circuits, System and Signal Processing (ProRISC 2004). The Netherlands. November 2004. P. 138−152.
5. Talla D., John L.K., Burger D. Bottlenecks in Multimedia Processing with SIMD Style Extensions and Architectural Enhancements // IEEE Transactions on Computers. August 2003. V. 52. № 8. P. 1015−1031.
6. Eisymont L.K. Gibridnaya strategiya razvitiya ehlementnoj bazy // Otkrytye sistemy. 2017. № 2. URL = https://www.osp.ru/os/2017/ 02/13052216/.
7. Ejsymont A.L., Chernikov A.V., Kosorukov D.E., Nasonov I.I., Komlev A.A. Geterogennaya mnogoprocessornaya sistema na kristalle s proizvoditel'nost'yu 512 Gflops // Tekst doklada. Mikroehlektronika-2017. Alushta. 11 s. URL = https://module.ru/upload/images/ 1507808903Гетерогенная многопроцессорная СнК с производительностью 512 Gflops.pdf.
8. Chernikov A.V., Chernikov V.M., Viksne P.E., SHeluhin A.M. Vysokoproizvoditel'noe processornoe yadro NMC4 dlya obrabotki vektornyh dannyh v formatah s plavayushchej i fiksirovannoj tochkami // Tekst doklada. Mikroehlektronika-2017. Alushta. 8 s. URL = https://module.ru/upload/images/1507871419Высокопроизводительное процессорное ядро NMC4 для обработки векторных данных в форматах с плавающей и фиксированной точками.pdf.