Journal Highly available systems №3 for 2018 г.
Article in number:
Memory access address localization problems in some signal processing systems and variants of their solution
Type of article: scientific article
DOI: 10.18127/j20729472-201803-06
UDC: 004
Authors:

A.V. Zaitsev – Head of Sector, JSC «All-Russia SRI of Radio Engineering» (Moscow)

E-mail: alex.zaitsev@gmail.com

L.K. Eisymont – Ph.D.(Phys.-Math.), The Federal Register of Scientific and Technical Experts  of the Russia Ministry of Education and Science (Moscow) E-mail: verger-lk@yandex.ru

Abstract:

The task of studying the processes taking place in the signal processing system in order to determine the position of the observed bodies in a certain space, and then – the choice of computing devices for the effective implementation of these processes. As the main feature of the processes, the modes of spatio-temporal localization of memory accesses are distinguished. The main processes of the system are determined, the most suitable vector and multithreaded processors, software emulation of multithreaded architecture on a multi-core processor are offered. The objects of study are processors Elbrus and NeuroMatrix NM6408. As a result, it is planned to select promising devices of the system processes. The planned study is consistent with the strategy of import substitution in the field of elemental design base of supercomputers aimed at developing a set of problem-oriented specialized VLSI rather than creating direct analogs of the best samples of foreign elemental design base.

Pages: 35-39
References
  1. Ediger D., McColl R., Riedy J., Bader D.A. STINGER: High Performance Data Structure for Streaming Graphs // 2012 IEEE Conference «High Performance Extreme Computing (HPEC)». 10−12 Sept. 2012. 5 p.
  2. Strohmaier E., Shan H. Architecture Independent Performance Characterization and Benchmarking for Scientific Applications. Volendam. The Netherlands. Oct. 2004. URL = https://ftg.lbl.gov/ApeX/mascots.pdf.
  3. Nelson J. et al. Crunching Large Graphs with Commodity Processors. HotPar 2011. 6 p.
  4. Zajcev A.V., Bikonov D.V., Andryushin D.V. HPGAS: opyt ehmulyacii massovo-mul'titredovyh sistem na mnogoyadernyh klasternyh superkomp'yuterah // Tezisy stendovogo doklada. URL = https://www.osp.ru/netcat_files/userfiles/MSKF_2015/MSCF_2015_Fin3.pdf.
  5. Mitrofanov V.V., Sluckin A.I., Eisymont L.K. Superkomp'yuternye tekhnologii dlya strategicheski vazhnyh zadach // Zhurnal «Ehlektronika: NTB». 2008. № 7. S. 66−79.
  6. 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.
  7. Chernikov V.M., Viksne P.E. Perspektivy povysheniya harakteristik i rasshireniya oblastej primeneniya transteraflopsnyh SBIS semejstva NeroMatrix // Sb. dokladov VI Mezhdunar. nauchno-praktich. konf. «Upravlenie informacionnoj bezopasnost'yu v sovremennom obshchestve». 5−7 iyunya 2018. Moskva.
  8. Gorbunov V.S., Eisymont L.K. Kompleksnaya metodika testirovaniya proizvoditel'nosti superkomp'yuterov // Vychislitel'nye metody i programmirovanie. 2013. T. 14. S. 115−121. URL = http://num-meth.srcc.msu.ru/zhurnal/tom_2013/pdf/v14r216.pdf.
  9. Eisymont L.K. Gibridnaya strategiya razvitiya ehlementnoj bazy // Otkrytye sistemy. 2017. № 2. URL = https://www.osp.ru/os/2017/ 02/13052216/.
  10. Eisymont L.K. Nastraivaemye specializirovannye SBIS – real'naya osnova sozdaniya budushchih ehkzamasshtabnyh superkomp'yuterov, zarubezhnyj i otechestvennyj opyt // Sb. dokladov VI Mezhdunar. nauchno-praktich. konf. «Upravlenie informacionnoj bezopasnost'yu v sovremennom obshchestve». 5−7 iyunya 2018. Moskva.
Date of receipt: 3 августа 2018 г.