SIMULATION OF SERVERS WITH INTERRUPTS IN LARGE MULTIPROCESSOR SYSTEMS
Annotation
The problem of managing a specialized cloud computing system performing heterogeneous resource-intensive tasks is considered when methods of the task execution can be represented as an arbitrary sorting out of a large number of options. Data-based parallelization of tasks under uncertainty is analyzed. To solve the problems of scheduling incoming input streams of tasks, the simulation is performed using the SimEvent/Simulink/MatLab software package. The functioning of the system and especially servers, represented as finite state machines, is analyzed. The peculiarity of the proposed model is the possibility of interrupting the server when an external signal appears. Using the developed server model makes it possible to improve system performance by saving time for each server to complete tasks and more efficiently distributing tasks between the servers of the entire system.
Keywords
Постоянный URL
Articles in current issue
- METHODOLOGY FOR ASSESSING THE NOISE IMMUNITY OF DATA EXCHANGE RADIO LINES WITH TIME DIVISION OF CHANNELS
- SIMULATION OF SERVERS WITH INTERRUPTS IN LARGE MULTIPROCESSOR SYSTEMS
- COMBINED CONTROL IN DIGITAL TRACKING SYSTEMS IN THE PRESENCE OF DELAY IN THE DIRECT CIRCUIT
- ROBUST CONTROL SYSTEM FOR HEAT EXCHANGER IN HOP EXTRACT PRODUCTION
- METHOD FOR DETERMINING A SPECIAL OPTIMAL CONTROL
- COMBINED OPTICAL METHOD FOR MEASURING THE THICKNESS OF LAYERS OF TRANSPARENT COATINGS AND FILMS
- ANALYSIS OF THE EFFECTIVENESS OF SCHEMATIC DIAGRAMS OF A TUNABLE SOURCE BASED ON AN ACOUSTO-OPTIC FILTER
- SILICON PHOTOELECTRONIC MULTIPLIERS AS A BASIS FOR CREATING COMBINED SENSORS
- INFLUENCE OF STRUCTURAL FEATURES OF CONSTRUCTION POLYMER-COMPOSITE MATERIALS ON THE PHYSICAL AND MECHANICAL PROPERTIES OF PRODUCTS