Article

Article title USE OF STRUCTURAL REDUNDANCY FOR INCREASE OF RELIABILITY OF ARITHMETIC UNIT OF THE COMPUTING ELEMENT OF BIMODULAR ARITHMETIC
Authors V.M. Amerbaev, E.S. Balaka, A.N. Schelokov
Section SECTION VI. COMPUTER COMPLEXES OF NEW GENERATION AND NEUROCOMPUTERS
Month, Year 07, 2014 @en
Index UDC 004.272.2
DOI
Abstract In this work a study of hardware methods for improving reliability characteristics of arithmetic unit of computing element implemented basing on bimodular arithmetic was carried out. RNS computing element is a minimal computing unit included in RNS processor which is structured as а latticed multilayer network. Advantage of such a structural solution is the elimination of redundant hardware needed to perform non-RNS operations and their parallel execution on computing elements, and also usage of error- and fault-tolerant calculations not only for RNS operations, but for non-RNS operations as well. Choice of bimodular arithmetic as the basis for CE construction was preferred due to more economic construction of its main units regarding hardware expenditures. The most well-known types of control were considered: modulo control, hardware duplication, triplicated redundancy with majority elements. For experiments special scripts were developed, generating Verilog descriptions of arithmetic unit together with the control circuits under consideration, the results were compared with respect to additional control hardware area and the loss in performance relative to unprotected circuits.

Download PDF

Keywords RNS-based prosessor; bimodular residue number system; methods of hardware control.
References 1. Akushskiy I.Ya., Yuditskiy D.M. Mashinnaya arifmetika v ostatochnihkh klassakh [Machine arithmetic in residual classes]. Moscow: Sov. radio, 1968, 440 p.
2. Shchelkunova Yu.O., Kalmykov M.I., Goncharov P.S., Petrova E.V. Razrabotka metodiki postroeniya nepozitsionnogo protsessora s degradiruemoy strukturoy [Development of a method of a non-positional processor with degradiruiut structure], Uspekhi sovremennogo estestvoznaniya [The successes of modern natural Sciences], 2014, No. 3, pp. 124-130.
3. Matutino P., Pettenghi H., Chaves R. and Sousa L. RNS arithmetic units for modulo {2n ± k}," in Digital System Design (DSD), 2012 15th Euromicro Conference on, September 2012, pp. 795-802.
4. Stempkovskiy A.L., Amerbaev V.M., Kornilov A.I. Modulyarnaya logarifmetika – novye vozmozhnosti dlya proektirovaniya modulyarnykh vychisliteley i preobrazovateley (kratkiy obzor) [The modular logaritmica - new opportunities for designing modular calculators and converters (brief overview)], IV Vserossiyskaya nauchno-tekhnicheskaya konferentsiya «Problemy razrabotki perspektivnykh mikro- i nanoelektronnykh sistem – 2010»: Sb. nauchn. tr. [IV all-Russian scientific-technical conference "problems of development of micro- and nanoelectronic systems - 2010": Collection of scientific works]. Moscow: IPPM RAN, 2010.
5. Geden B. Understand and Avoid Electromigration (EM) & IR-drop in Custom IP Blocks, Synopsys Webinars. 2011.
6. Stempkovskiy A.L., Amerbaev V.M., Solovev R.A. Vychislitelnyy element modulyarnoy arifmetiki [Computing element modular arithmetic]. Patent RF No 123995 ot 10 yanvarya 2013.
7. Amerbaev V.M., Balaka E.S. Arifmeticheskoe ustroystvo bimodulnoy arifmetiki konechnogo polya GF(P) [The arithmetic device bimodules arithmetic of finite field GF(P)], Research Journal of International Studies XX(17), pp. 2; 5-9.
8. Pospelov D.A. Arifmeticheskie osnovy vychislitelnykh mashin diskretnogo deystviya [Arithmetic basics of computers discrete steps]. Moscow: Vihsshaya shkola, 1970, 308 p.
9. Kuzmin I.V., Burnazyan R.G., Kovergin A.A. Apparatnyy kontrol elektronnykh tsifrovykh vychislitelnykh mashin [Hardware control of electronic digital computing machines]. Moscow: Ehnergiya, 1974.
10. Amosov V.V. Skhemotekhnika i sredstva proektirovaniya tsifrovykh ustroystv [Schematic and design tools of digital devices]. Saint-Petersburg: BKhV-Peterburg, 2007, 542 p.
11. Kornilov A.I., Semenov M.Yu., Lastochkin O.V., Kalashnikov V.S. Principy postroeniya specializirovannykh vyhchisliteley s primeneniem modulyarnoy arifmetiki [Principles of construction of specialized solvers using modular arithmetic], Problemy razrabotki perspektivnykh mikroelektronnykh sistem – 2005. Sbornik nauchnykh trudov [Problems of development of advanced microelectronic systems - 2005. Collection of scientific works]. Moscow: IPPM RAN, 2005, pp. 346-351.

Comments are closed.