RESEARCH OF APPLICABILITY LIMITATIONS FOR ELBRUS MICROPROCESSORS FOR SOLVING TASKS OF TECHNICAL VISION
Abstract
One of the key areas in the artificial intelligence is technical vision. For resource-intensive
tasks of technical vision high-performance, computing systems are created with use of specialized
accelerators. The use of such accelerators is necessary due to the inability of general-purpose
microprocessors (GPM) to solve such problems in a given time due to a high computational load.
However, the microprocessors of Elbrus series are successfully used to solve technical vision
problems in both server and on-board modes, and the appearance of the sixth-generation Elbrus
microprocessors should further improve performance on such tasks. Due to the high cost, greater
complexity and limitations in the use of systems with specialized accelerators, the question arises
of determining the conditions under which, it is sufficient to use CPU’s to solve the tasks of technical
vision, for example, with the microprocessors of the Elbrus series without special accelerators.
One of the most resource-intensive tasks in the field of technical vision are detection and
classification of objects. For the detection of objects one of the popular methods is the Viola-Jones
method. Convolutional neural networks are usually used to solve the classification problem.Mathematical models of computations have been developed for VGG16 and VGG19 neural networks
in relation to the actual microprocessors of the Elbrus series. Using the developed models,
the theoretical sufficiency of the performance of Elbrus microprocessors for technical vision tasks
is substantiated. Also, based on these methods, programs for modeling detection and classifications
objects in the image and video stream have been developed. The programs are written in
C++ using the OpenCV library, OPO Elbrus, the GNS Platform library and the ImageNet competition
database. Using the implemented programs, comparative testing was carried out on a number
of high-performance computing systems with Elbrus and Intel CPU’s and NVidia video card.
Based on the results obtained, it is shown that the Elbrus-8S is sufficient to solve the problem of
searching for objects in the image for input resolutions up to 1920 x 1080, where the processing
speed of the video stream is more than 20 frames per second.
References
robots], Peredovye innovatsionnye razrabotki. Perspektivy i opyt ispol'zovaniya, problemy
vnedreniya v proizvodstvo [Advanced innovative developments. Prospects and experience of
use, problems of introduction into production], 2019, pp. 73-75.
2. Gondimalla A. [et al.]. Sparten: A sparse tensor accelerator for convolutional neural networks,
Proceedings of the 52nd Annual IEEE/ACM International Symposium on Microarchitecture,
2019, pp. 151-165.
3. Deng L. [et al.]. Model compression and hardware acceleration for neural networks: A comprehensive
survey, Proceedings of the IEEE, 2020, Vol. 108, No. 4, pp. 485-532.
4. Auten A., Tomei M., Kumar R. Hardware acceleration of graph neural networks, 2020 57th
ACM/IEEE Design Automation Conference (DAC). IEEE, 2020, pp. 1-6.
5. Kim A.K. [i dr.]. Mikroprotsessory i vychislitel'nye kompleksy semeystva" El'brus" [Microprocessors
and computing complexes of the Elbrus family]. Saint Petersburg: Izdatel'skiy
dom" Piter", 2013.
6. Limonova E.E. [i dr.]. Otsenka bystrodeystviya sistemy raspoznavaniya na VLIW arkhitekture
na primere platformy El'brus [Evaluation of the performance of the recognition system on the
VLIW architecture on the example of the Elbrus platform], Programmirovanie [Programming],
2019, No. 1, pp. 15-21.
7. Bocharov N.A., Zuev A.G., Slavin O.A. Proizvoditel'nost' mikroprotsessora El'brus-8SV dlya
resheniya zadach tekhnicheskogo zreniya v usloviyakh ogranicheniy energopotrebleniya [The
performance of the Elbrus-8SV microprocessor for solving problems of technical vision in
conditions of limited energy consumption], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya
SFedU. Engineering Sciences], 2021, No. 1 (218), pp. 259-271.
8. Bocharov N.A. i dr. Proizvoditel'nost' vychislitel'noy tekhniki s protsessorom «El'brus-8S» na
zadachakh robototekhnicheskogo kompleksa [The performance of computing equipment with
the Elbrus-8C processor on the tasks of the robotic complex], Nanoindustriya [Nanoindustry],
2018, No. 82, pp. 79.
9. Kozhin A.S. snovnye proektnye resheniya dlya protsessora «El'brus-16S» [Basic design solutions
for the processor "Elbrus-16S"], Nanoindustriya [Nanoindustry], 2020, Vol. 13, No. S4,
pp. 74-75.
10. Bychkov I.N., Lobanov I.N., Molchanov I.A. Vychislitel'naya tekhnika na osnove apparatnoprogrammnoy
platformy «El'brus» dlya perspektivnykh informatsionnykh sistem [ omputer
technology based on the hardware and software platform "Elbrus" for advanced information
systems], Pribory [Devices], 2018, No. 8, pp. 14.
11. Kozhin A.S., Neyman-zade M.I., Tikhorskiy V.V. Vliyanie podsistemy pamyati
vos'miyadernogo mikroprotsessora «El'brus-8S» na ego proizvoditel'nost' [The influence of the
memory subsystem of the eight-core microprocessor "Elbrus-8S" on its performance], Voprosy
radioelektroniki [Questions of radio electronics], 2019, No. 3, pp. 13-21.
12. Vizil'ter Yu.V. i dr. Obrabotka i analiz izobrazheniy v zadachakh mashinnogo zreniya [Image
processing and analysis in machine vision tasks], 2010.
13. Vizil'ter Yu.V., Gorbatsevich V.S., Zheltov S.Yu. Strukturno-funktsional'nyy analiz i sintez
glubokikh konvolyutsionnykh neyronnykh setey [Structural and functional analysis and synthesis
of deep convolution neural networks], Komp'yuternaya optika [Computer optics], 2019,
Vol. 43, No. 5, pp. 886-900.
14. Albawi S., Mohammed T.A., Al-Zawi S. Understanding of a convolutional neural network,
2017 international conference on engineering and technology (ICET). Ieee, 2017, pp. 1-6.
15. O'Shea K., Nash R. An introduction to convolutional neural networks, arXiv preprint
arXiv:1511.08458, 2015.
16. Krizhevsky A., Sutskever I., Hinton G.E. Imagenet classification with deep convolutional neural
networks, Advances in neural information processing systems, 2012, Vol. 25.
17. Simonyan K., Zisserman A. Very deep convolutional networks for large-scale image recognition,
arXiv preprint arXiv: 1409.1556, 2014.
18. Neyman-zade M.I., Korolev S.D. Rukovodstvo po effektivnomu programmirovaniyu na
platforme" El'brus" [Guide to effective programming on the Elbrus platform]. Moscow:
A « TSST, 2020.
19. Viola P. [et al.]. Robust real-time object detection, International journal of computer vision,
2001, Vol. 4, No. 34-47, pp. 4.
20. Biryukov A.A., Taranin M.V., Taranin S.V. Protsessor 1879VM6YA. Realizatsiya glubokikh
svertochnykh neyronnykh setey [Processor 1879VM6YA. Implementation of deep convolutional
neural networks], DSPA: Voprosy primeneniya tsifrovoy obrabotki signalov [DSPA: Issues
of application of digital signal processing], 2018, Vol. 8, No. 4, pp. 191-195.
21. Petrichkovich Ya. i dr. RoboDeus-50-yadernaya geterogennaya SnK dlya vstraivaemykh
sistem i robototekhniki [RoboDeus-50-nuclear heterogeneous SnC for embedded systems and
robotics], Elektronika: Nauka, tekhnologiya, biznes [Electronics: Science, Technology, Business],
2020, No. 7, pp. 52-63.
