|
Library
|
Your profile |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.|
Software systems and computational methods
Reference:
Knyazkov V.S., Volchenskaya T.V.
Markov models for performance evaluation of parallel computing
// Software systems and computational methods.
2013. ¹ 1.
P. 10-27.
URL: https://en.nbpublish.com/library_read_article.php?id=62445
Knyazkov V.S., Volchenskaya T.V. Markov models for performance evaluation of parallel computingAbstract: the article discusses a hierarchical approach to the analysis of the complexity of parallel computing process implementation. The authors introduce the concept of elementary homogeneous and elementary heterogeneous computational processes on which the analysis of the heterogeneous branching parallel processes and heterogeneous computational processes is based. The article shows that complex parallel process and heterogeneous computational processes can be represented in the form of discrete Markov processes of a special type. This approach allows to transform such processes to a classical discrete Markov processes with a consecutive change of states. Keywords: Software, efficiency, parallel computations, computing, computational structures, Markov models, organization of computations, conveyor computations, computations with parallel objects, optimization of computations
This article can be downloaded freely in PDF format for reading. Download article
This article written in Russian. You can find original text of the article here . References
1. Osnovy teorii vychislitel'nykh sistem. Pod redaktsiey Mayorova S.A. M., «Vysshaya shkola», 1978, 407 s.
2. Knyaz'kov V.S., Volchenskaya T.V. Konveyernye vychisleniya i struktury: formal'-nye modeli i analiticheskie otsenki slozhnosti vychisleniy. V kn.: Materialy 12 Mezhdunarodnoy nauchno-tekhnicheskoy konferentsii «Problemy peredachi i obra-botki informatsii v setyakh i sistemakh telekommunikatsiy», Ryazan', 2004, s. 128-131. 3. Knyaz'kov V.S., Potapov A.A. Metodika otsenki trudoemkosti realizatsii matrichnykh mul'tiprotsessornykh sistem. Trudy mezhdunarodnogo simpoziuma «Aktual'nye pro-blemy nauki i obrazovaniya», Penza, 2003, s. 400-402. 4. Knyaz'kov V.S. Sposoby organizatsii i slozhnost' massivnykh vychisleniy v odnomernykh iterativno-bitovykh protsessornykh sredakh // Vychislitel'naya tekhnika v avtomatizi-rovannykh sistemakh kontrolya i upravleniya: Mezhvuzovskiy sbornik nauchnykh trudov – Penza: Izdatel'stvo Penzenskogo gosudarstvennogo universiteta, 1999 g.– Vyp. 26.– s. 3-9. 5. Knyaz'kov V.S. Arifmetiko-konveyernye vychisleniya v dvumernykh odnorodnykh sredakh // Materialy 3 Mezhdunarodnoy nauchno-tekhnicheskoy konferentsii «Novye informa-tsionnye tekhnologii i sistemy» 10-11 dekabrya 1998 g., g. Penza, – Penza,1998, S. 53-54. 6. Knyaz'kov V.S. Obshchaya otsenka slozhnosti realizatsii massivno-kletochnykh vychisleniy v iteratsionno-bitovykh vychislitel'nykh strukturakh s mnogomernoy organizatsiey. //Izvestiya TRTU, N 3,1997,c.218. 7. Knyaz'kov V.S. Dvumernye iterativno-bitovye protsessory: vremennaya i prostran-stvennaya slozhnost' posledovatel'no-massivnykh iteratsionno-razryadnykh vychisle-niy //Materialy 2-oy Mezhdunar. n/t konf. «Novye informatsionnye tekhnologii i sistemy,ch.1,1996, Penza,S.108-109. 8. Knyaz'kov V.S., Biktashev R.A. Arkhitektura parallel'nykh vychislitel'nykh sistem.-Penza,1993.-s.166. 9. Knyaz'kov V, S., Volchenskaya T. V. Sposoby postroeniya konveyernykh vychislitel'-nykh struktur s upravleniem kommutatsii potokov dannykh. Dep. ruk., VINITI, N 5581V9D ot 31. 10. 90, Minvuz RSFSR, Penza, Penzenskiy politekhnicheskiy i – nt, 1990, 10 s |
© 1998 – 2024 Nota Bene. Publishing Technologies. NB-Media Ltd.
Eng
