Cloud Monte Carlo state machine system and architecture method
A state machine and state technology, applied in the field of cloud Monte Carlo state machine system, architecture method and storage medium, can solve the problems of high computing pressure and high usage cost, and achieve the effect of balanced operation and improved computing efficiency
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0050] figure 1 and figure 2 A schematic structural diagram of the cloud Monte Carlo state machine system 100 is shown. The cloud Monte Carlo state machine system 100 provided by the present invention can include a main control terminal 160, a data processing center 170 and several computing centers 180, wherein the main control terminal, data processing centers and computing centers are connected by communication; preferred in this embodiment Generally, several computing centers 180 may be included in a cloud Monte Carlo state machine system 100, such as figure 1 As shown, the cloud Monte Carlo state machine system 100 includes 1# computing center, 2# computing center...n# computing center. Preferably, as figure 2As shown, the present invention also provides a cloud Monte Carlo state machine system 200 , the difference between the cloud Monte Carlo state machine system 200 and 100 is that the state machine system includes several data processing centers 170 .
[0051] S...
Embodiment 2
[0071] The present invention also provides a cloud Monte Carlo state machine system architecture method, comprising the following steps:
[0072] in image 3 It is a flowchart of the cloud Monte Carlo state machine system architecture method in this embodiment.
[0073] (1) The main control terminal 160 divides the calculation task into several subtasks, then distributes the subtasks to the computing center 180, and notifies the data processing center about the task assignment status and task content 210 of each computing center 180;
[0074] Wherein, the subtask is one or more of the following: GPU parallel task, CPU parallel task or CPU-GPU mixed parallel task;
[0075] The distribution method of subtasks is to assign a subtask to each computing center;
[0076] Each subtask distribution has at least one random number seed, wherein the random number can be randomly assigned by the main control terminal 160 or generated by the computing center 180, each random number seed d...
Embodiment 3
[0084] The present invention also provides a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions, and the instructions are adapted to be loaded by the memory and execute the image registration method, including the following steps:
[0085] (1) The master control terminal divides the calculation task into several subtasks, then assigns the subtasks to the computing centers, and notifies the data processing center about the task assignment status and task content of each computing center;
[0086] (2) The main control terminal transmits the input data to the data processing center, and the computing center performs calculation tasks after obtaining the data from the data processing center;
[0087] (3) After the calculation tasks of each calculation center are completed, the calculation results are fed back to the main control terminal; after the calculation tasks of each calculation center are completed or reach t...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com