Dispatching method and system of distributed system
A technology of distributed system and scheduling method, applied in the field of scheduling method and system of distributed system, can solve problems such as resource waste and waste, and achieve the effects of avoiding resource waste, improving processing speed, and reducing network overhead.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0052] According to an embodiment of the present invention, a scheduling method for a distributed system is provided, image 3 It is a flowchart of a scheduling method for a distributed system according to an embodiment of the present invention, such as image 3 As shown, this embodiment includes:
[0053] Step S102: the data scheduler receives a scheduling request, the scheduling request includes a read operation of data, wherein the data is stored in at least one entity providing data services in the distributed system;
[0054] Step S104: The data scheduler determines the status of each data service corresponding to the scheduling request, and the status information in this embodiment includes: read, write, idle;
[0055] Step S105: Judging whether it is in an idle or reading state, if yes, execute step S106, otherwise execute step S107;
[0056] Step S106: Send a scheduling response agreeing to read each data service; each data service is at least one same data service, ...
Embodiment 2
[0060] Figure 4 It is a schematic diagram for analyzing the scheduling method of the distributed system according to Embodiment 2 of the present invention, such as Figure 4 As shown, this embodiment takes the file scheduler as a data scheduler and the distributed environment system as an example for description. The system includes a task scheduler and a file scheduler, and node 1 stores files CF1 (including CF1.1, CF1.2 two parts), node 2 and node 3 save the backup of the file, that is, each file has 3 copies (the master-slave backup can not be distinguished), which are stored on node 1, node 2 and node 3 respectively, and 3 copies The status information of the replica's data service needs to be consistent, that is, the status of the data service includes: read, write, and idle.
[0061] Assuming that the task scheduler receives a scheduling request for a task (also called an application) and needs to schedule a file CF1 (including CF1.1 and CF1.2), it sends the scheduling...
Embodiment 3
[0072] Figure 5 It is a flow chart of the dispatching method of the distributed system according to Embodiment 3 of the present invention. In this embodiment, the data scheduler includes a small table scheduler and a file scheduler, and the distributed environment system has a three-layer data structure as an example. The system includes task scheduler, small table scheduler, file scheduler, such as Figure 5 As shown, this embodiment includes:
[0073] Step S202: After receiving the task, the task scheduler determines the operation type of the task and the data service corresponding to the task, such as the name of the small table for input and output, whether it is a read or write operation, etc., and sends the scheduling request to the small table scheduler;
[0074] Step S204: After receiving the scheduling request, the small table scheduler determines the small table where each data service corresponding to the scheduling request is located. If the small table scheduler...
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