Cloud Processing Plan Allocation for Table and Record Parallelization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cloud-based processing systems struggle with inefficient processing time and load management due to the lack of appropriate parallelization techniques, requiring skilled personnel to determine optimal task execution order, especially when handling nonstationary processing tasks on on-demand volume-charged cloud resources.
Innovation Solution
A management computer system that utilizes a processor to generate and execute processing plans, employing table and record parallelization strategies to allocate tasks efficiently across virtual cores, optimizing processing time and reducing overhead costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If task processing is performed in parallel on cloud resources, then processing speed is improved, but processing overhead increases
Solution Approach 1:
The patent segments tasks into units based on table size thresholds. When a table exceeds the threshold, it is divided into multiple records for parallel processing. This segmentation allows the system to achieve parallel processing benefits while controlling overhead by only dividing when necessary, thus resolving the contradiction between processing speed and overhead.
Solution Approach 2:
The patent dynamically adjusts the parallelization level based on table size parameters. By changing the processing unit from table-level to record-level based on the data size parameter, the system optimizes the balance between processing speed and overhead, achieving faster processing only when the data volume justifies the additional parallelization complexity.
2Productivity
If manual determination of parallelization is performed, then processing optimization is achieved, but operational complexity increases
Solution Approach 1:
The patent implements self-service through automatic parallelization determination based on predefined table size thresholds. The system autonomously decides whether to divide tables into records without requiring manual intervention, thereby maintaining processing optimization while significantly reducing operational complexity and the need for skilled personnel.
Solution Approach 2:
By establishing automatic parameter-based decision rules (threshold values for table size), the patent transforms manual optimization into automated parameter-driven processing. This eliminates the need for skilled personnel to manually determine parallelization strategies, reducing operational complexity while maintaining optimized processing outcomes.
3Adaptability or versatility
If cloud resources are used for nonstationary processing, then resource flexibility is improved, but processing time management becomes difficult
Solution Approach 1:
The patent segments processing tasks into manageable units (tables and records) that can be independently allocated to cloud resources. This segmentation enables better time management by allowing granular control over processing duration and resource allocation, thus resolving the contradiction between resource flexibility and processing time management.
Solution Approach 2:
The patent incorporates feedback mechanisms to monitor and adjust processing time based on actual execution characteristics. By continuously feedback on processing progress and resource utilization, the system dynamically manages processing time while maintaining the flexibility benefits of cloud resources, resolving the time management difficulty.
Data Source
AI summary
Table parallelization processing for parallelizing data processing is performed on a plurality of tables in units of allocation tables to a core of a processing execution computer, and record parallelization processing for dividing a table having a large data size into a plurality of records and parallelizing data processing on a plurality of records in units of allocation records to the core of the processing execution computer is performed when the table is larger than a predetermined data size.


