350 rub
Journal Information-measuring and Control Systems №6 for 2015 г.
Article in number:
MISD-method and hardware semaphore of final result
Authors:
E.A. Titenko - Ph.D. (Eng.), Associate Professor, Chair of Information Systems and Technologies, South-Western State University (Kursk). E-mail: johntit@mail.ru
Abstract:
This paper proposes an organization MISD- calculations. The essence of the calculations is to divide the total computing process into separate local processes. The local processes have their own starting points. Parallel computing on local data fragments is achieved by simultaneously calculating the starting point for the structure of the original data. The result is a variable time duration of the operation. It is determined by the duration of most of the local computing process. Thus, the distinguishing feature of MISD-computing method is depending on the time of the operation of the structure of the data being processed.
Create specialized additional scheme to monitor of the readiness result. It monitors the current state of parallel processes performed locally and sets the flags of their readiness. The scheme is a hardware semaphore of readiness results. It is created on the basis of the shift register. In this case input multiplexers. Multiplexers selected data source as the previous value of an trigger or external starting point.
MISD-method and the hardware semaphore readiness result allows to reduce the overall average computation time. Reduction of time is achieved by processing statistically large number of small local area along the length of the processes.
Pages: 68-72
References
- Titenko E.A. Organizacija rekonfiguriruemogo multiprocessora potoka dannykh // Vestnik kompjuternykh i informacionnykh tekhnologijj. 2012. № 2. S. 24-29.
- Titenko E.A. Semenikhin E.A., Petrik E.A. Strukturno-funkcionalnaja organizacija arbitra parallelnojj obrabotki zaprosov // Informacionno-izmeritelnye i upravljajushhie sistemy. 2010. № 11. T. 8. S. 30-35.
- Titenko E.A., Skornjakov K.S., Busygin K.N.Metody i summatory s parallelnymi gruppovymi processami // Izvestija JUgo-Zapadnogo gosudarstvennogo universiteta. Ser. Upravlenie, vychislitelnaja tekhnika, informatika. Medicinskoe priborostroenie. 2013. № 2. CH. 1 S. 161-165.
- Zotov I.V. i dr. Apparatnaja realizacija barernojj sinkhronizacii v matrichnykh multikompjuterakh // Informacionno-izmeritelnye i upravljajushhie sistemy. 2013. № 8. S. 41-45.
- Titenko E.A., Evsjukov V.S., Semenikhin E.A.Apparatnyjj semafor dlja asinkhronnykh ustrojjstv // Teoreticheskie i prikladnye voprosy sovremennykh informacionnykh tekhnologijj: Materialy Vseros. nauch.-tekhn. konf. Ulan-Udeh: Izd-vo VSGTU. 2006. CH. I. S. 196-200.
- Titenko E.A., Emeljanov S.G., Zerin I.S. Odnorodnye vychislitelnye struktury dlja parallelnykh simvolnykh vychislenijj // Izvestija JUgo-Zapadnogo gosudarstvennogo universiteta. № 6(39). CH. 2. S. 77?82.