PERFORMANCE OF THE ELBRUS-8SV MICROPROCESSOR FOR TECHNICAL VISION TASKS UNDER POWER CONSTRAINTS
Abstract
Research in the field of creating specialized computing systems for robots is conducted in
many world scientific centers, including our country. The development of capabilities of sensor
systems, global navigation systems, growth of computing power and improvement of algorithms
allow creating onboard computing systems with broad intellectual capabilities. An important, butunsolved problem remains the equipping of such computing systems with domestically produced
microprocessors. The emergence of domestic computing systems and software of the new generation,
such as microprocessor "Elbrus-8SV" and OS “Elbrus” opens up new opportunities for developers
of robotic complexes. The peak performance of "Elbrus-8SV" microprocessor is more
than 0.25 TFLOPS of double precision, which allows solving computationally complex tasks, such
as technical vision tasks, on the microprocessor. Another important requirement of onboard computing
technology, in addition to computational power, is low power consumption. As a rule, on
general-purpose microprocessors, high performance is impossible with low power consumption,
and specialized processors, such as vector or neuroprocessors, are used to solve computationally
complex technical vision tasks. To reduce the power consumption of general-purpose microprocessors,
there are special methods, among which the authors considered: switching off the physical
cores, reducing the clock frequency, switching off the pipeline, switching off synchronous pulses
in the idle state. The authors reviewed typical technical vision tasks solved by onboard computing
systems. An experiment was conducted to estimate power consumption and execution time of
vision algorithms when the clock frequency is reduced and the microprocessor cores are switched
off. The experiments showed the possibility to decrease the power consumption of the Elbrus-8SV
microprocessor cores by 36-46% with an increase in the program execution time. Based on the
results of the experiment, conclusions were made about the applicability of the Elbrus-8SV microprocessor
for creating advanced onboard computing systems with the ability to operate both in
high performance mode and with reduced power consumption. The results obtained by the authors
indicate the prospects of import substitution in the field of robotics.
References
protsessora «El'brus-8SV» [Solutions for computer equipment based on the processor "Elbrus-
8SV"], Nanoindustriya [Nanoindustria], 2019, No. S89, pp. 57-59.
2. Volkonskiy V.Yu., Ivanov M.A., Kim A.K., Semenikhin S.V. Kompleks programmnykh sredstv
antivirusnoy zashchity komp'yuternykh sistem, funktsioniruyushchikh pod upravleniem os
semeystva El'brus [A set of software tools for anti-virus protection of computer systems operating
under the control of the “Elbrus” OS family], REDS: Telekommunikatsionnye ustroystva
i sistemy [REDS: Telecommunications devices and systems], 2015, Vol. 5, No. 3, pp. 282-285.
3. Kim A.K., Perekatov V.I., Ermakov S.G. Mikroprotsessory i vychislitel'nye kompleksy
semeystva ”El'brus” [Microprocessors and computing systems of the ”Elbrus " family]. Saint
Petersburg: Piter, 2013, 272 p.
4. Renan Augusto Starke, Andreu Carminati, Rymulo Silva de Oliveira, Evaluation of a low
overhead predication system for a deterministic VLIW architecture targeting real-time applications,
Microprocessors and Microsystems, 2017, Vol. 49, pp. 1-8.
5. Neyman-zade M.I., Volkonskiy V.Yu. Sredy programmirovaniya i optimiziruyushchie
kompilyatory dlya komp'yuterov s mikroprotsessorami arkhitektury «El'brus» [Programming
environments and optimizing compilers for computers with microprocessors of the "ELBRUS"
architecture], Pribory [Devices], 2018, No. 8, pp. 21-29.
6. Volkonskiy V.Yu., Breger A.V., Buchnev A.Yu., Grabezhnoy A.V., Ermolitskiy A.V., Mukhanov
L.E., Neyman-zade M.I., Stepanov P.A., Chetverina O.A. Metody rasparallelivaniya programm
v optimiziruyushchem kompilyatore [Methods of program parallelization in an optimizing
compiler], Voprosy radioelektroniki. Ser. EVT [Questions of radio electronics. EVT series],
2012, Issue 3, pp. 63-88.
7. Ermolitskiy A.E., Neyman-zade M.I., Chetveri-na O.A., Markin A.L., Volkonskiy V.Yu. Agressivnaya
inlayn-podstanovka funktsiy dlya VLIW-arkhitektur [Aggressive inline function substitution for
VLIW architectures], Tr. ISP RAN [Proceedings of ISP RAS], 2015, Vol. 27, Issue 6.
8. Volkonskiy V.Yu., Grabezhnoy A.V., Mukhanov L.E., Neyman-zade M.I. Issledovanie vliyaniya
podsistemy pamyati na proizvoditel'nost' rasparallelennykh programm [Investigation of the effect of
the memory subsystem on the performance of parallelized programs], Voprosy radioelektroniki. Ser.
EVT [Questions of radio electronics. EVT series], 2011, Issue 3, pp. 22-37.
9. Ivanov D.S. Raspredelenie registrov pri planirovanii instruktsiy dlya VLIW-arkhitektur [Register
allocation when planning instructions for VLIW architectures], Programmirovanie [Programming],
2010, No. 6, pp. 74-80.
10. Bukin A.G., Lychagov A.S., Sadekov R.N., Slavin O.A. Apparatno-programmnyy kompleks
komp'yuternogo zreniya dlya resheniya zadach navigatsii nazemnykh podvizhnykh ob"ektov
[Hardware and software complex of computer vision for solving problems of navigation of ground
mobile objects], Giroskopiya i navigatsiya [Gyroscopy and navigation], 2015, No. 2, pp. 58-66.
11. Bocharov N.A., Paramonov N.B., Slavin O.A., Yanko D.V. Otsenka perspektiv ispol'zovaniya
vychislitel'nykh sredstv semeystva «El'brus» pri realizatsii algoritmov raspoznavaniya v
sovremennykh robototekhnicheskikh kompleksakh [Evaluation of the prospects for the use ofcomputing tools of the "Elbrus" family in the implementation of recognition algorithms in
modern robotic complexes], Voprosy radioelektroniki [Questions of radio electronics], 2018,
No. 2, pp. 99-105.
12. Suman Harapanahalli, Niall O Mahony, Gustavo Velasco Hernandez, Sean Campbell, Daniel
Riordan, Joseph Walsh, Autonomous Navigation of mobile robots in factory environment,
Procedia Manufacturing, 2019, Vol. 38, pp. 1524-1531. ISSN 2351-9789. Available at:
https://doi.org/10.1016/j.promfg.2020.01.134.
13. Pileun Kim, Jisoo Park, Yong K. Cho, Junsuk Kang, UAV-assisted autonomous mobile robot
navigation for as-is 3D data collection and registration in cluttered environments, Automation
in Construction, 2019, Vol, 106, pp. 102918. ISSN 0926-5805. Available at: https://doi.org/
10.1016/j.autcon.2019.102918.
14. Bulatov K., Arlazarov V., Chernov T., Slavin O., Nikolaev D. Smart IDReader: Document
Recognition in Video Stream, The 14th IAPR International Conference on Document Analysis
and Recognition (ICDAR 2017), CBDAR 2017: November 9–12, Kyoto, Japan, 2017, pp. 39-44.
15. Arlazarov V.V., Arlazarov V.L., Bulatov K., Nikolaev D.P., Polevoy D., Slavin O. Sistema
raspoznavaniya dokumentov v videoposledovatel'nosti [Document recognition system in video
sequences], Gosudarstvennyy reestr poleznykh modeley Rossiyskoy Federatsii [State Register
of Utility Models of the Russian Federation].
16. Limonova E.E., Bocharov N.A., Paramonov N.B., Bogdanov D.S., Arlazarov V.V., Slavin O.A.,
Nikolaev D.P. Performance Evaluation of a Recognition System on the VLIW Architecture by
the Example of the Elbrus Platform, Programming and Computer Software, 2019, Vol. 45 (1),
pp. 12-17. Doi: 10.1134/S0361768819010055.
17. Neyman-zade M.I., Korolev S.D. Rukovodstvo po effektivnomu programmirovaniyu na
platforme «El'brus» [Guide to effective programming on the “Elbrus” platform], 2020.
18. Intel® Threading Building Blocks (Intel TBB). Available at: https:// www.threadingbuilding
blocks.org/ (accessed 18 February 2021).
19. ra tsuno E. ., i a lo .S., Semenikhin S.V. Ispol'zovanie preryvani sistemnogo
kontrolya SCI dlya upravleniya energopotrebleniem mikroprotsessorov seme stva «El'brus»
[Use of interrupts of the SCI system control for power consumption control of the "Elbrus"
family of microprocessors], Voprosy radioelektroniki. Ser. EVT [Questions of radio
electronics. EVT series], 2015, Issue 1.
20. Ablakatov M.A. Upravlenie energopotrebleniem servera semeystva «El'brus», osnovannoe na
otklyuchenii protsessornykh yader po dannym podsistemy balansirovki zagruzki [Power management
of the server of the "Elbrus" family, based on the shutdown of processor cores according
to the load balancing subsystem], Voprosy radioelektroniki [Questions of radio electronics],
2017, No. 3, pp. 48-51.
21. David Meisner, Brian T. Gold, Thomas F. Wenisch. PowerNap: Eliminating Server Idle Power,
ASPLOS’09, March 7-11 2009, Washingtonm DC, USA.
22. Advanced Configuration and Power Interface Specification, Hewlett-Packard Corporation,
Intel Corporation, Microsoft Corporation, Phoenix Technologies Ltd, Toshiba Corporation,
Revi-sion 4.0a, April 5, 2010.
23. Kravtsunov E.M., Semenikhin S.V. Upravlenie energopotrebleniem SNK «El'brus-2S+» v
sostoyanii prostoya protsessornogo yadra [Power consumption management of the SNK "Elbrus-
2S+" in the idle state of the processor core], Voprosy radioelektroniki [Questions of radio
electronics], 2013, Vol. 4, No. 3, pp. 143-157.
24. Thomas Gleixner, Douglas Niehaus. Hrtimers and Beyond: Transforming the Linux Time
Subsystems, Proceedings of the Linux Symposium, 2006, Vol. One, pp. 333.
25. Venkatesh Pallipadi, Shaohua Li, Adam Belay. cpuidle – Do nothing, efficiently… , Proceedings
of the Linux Symposium, 2007, Vol. Two. pp. 119.
26. TPS53667 6-Phase, D-CAP+, Step-Down, Buck Controller with NVM and
PMBusTMInterface for ASIC Power and High-Current Point-of-Load. Available at:
https://www.ti.com/lit/ds/symlink/tps53667.pdf (accessed 18 February 2021).
27. TPS53819A3-Vto28-VInput, 40-A, Eco-ModeTM, D-CAP2TMSynchronous Buck Controller
With PMBusTM. Available at: https://www.ti.com/lit/ds/symlink/tps53819a.pdf (accessed 18
February 2021).
