Distributed Data Unit Allocation for Balanced Processing Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In distributed processing systems, the redundancy arrangement technique leads to prolonged processing times and decreased throughput due to the concentration of data units on fewer work units when one work unit fails, as other units without identical data units cannot perform processing in its stead.
Innovation Solution
An information processing apparatus that allocates copied data units to different work unit processors, ensuring a balanced distribution of identical data units across processors to minimize the variation in the number of data units each processor handles, thereby enabling efficient processing even if a failure occurs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data units are arranged in a distributed manner only in view of redundancy, then data reliability is improved, but processing throughput decreases
Solution Approach 1:
The patent segments data units into groups and distributes them across multiple work units in a structured manner. Each work unit holds specific data unit groups, and when a failure occurs, other work units can take over processing of failed work unit's data units. This segmentation approach maintains redundancy while enabling parallel processing across multiple work units, thus improving throughput without sacrificing reliability.
Solution Approach 2:
The patent implements local quality by having different work units hold different sets of data unit groups with specific redundancy relationships. Each work unit has a localized role in the overall system, knowing which specific data units it can take over. This localized organization allows for efficient failure recovery while maintaining high throughput, as only relevant work units need to be involved in takeover operations rather than all work units.
2Reliability
If many work units are used to maintain reliability, then system redundancy is improved, but communication overhead increases
Solution Approach 1:
The patent applies preliminary action by pre-organizing data units into groups and pre-determining which work units can take over which data units before failures occur. This advance organization eliminates the need for complex real-time communication and decision-making when failures happen. The takeover relationships are established in advance, reducing communication overhead during operational phases while maintaining high system redundancy.
3Reliability
If copied data units are concentrated on fewer work units, then redundancy is maintained, but processing time increases
Solution Approach 1:
The patent segments copied data units across multiple work units rather than concentrating them on fewer units. Each work unit holds a portion of the copied data units, enabling parallel processing of multiple data units simultaneously. This segmentation maintains redundancy through distributed copying while reducing processing time by utilizing the computational capacity of multiple work units in parallel, rather than sequential processing on fewer units.
Data Source
AI summary
An information processing apparatus includes data unit output controller configured to output, to a plurality of data unit processors, copied data units obtained by respectively copying a plurality of data units for a number that is set in advance, configured to respectively allocate output destinations of a plurality of identical copied data units to different data unit processors, configured to, for each combination of at least two data unit processors that is set in advance, obtain a number of the copied data unit groups in which the identical copied data units are respectively allocated to the data unit processors made into a combination, and configured to respectively allocate the output destinations of the plurality of copied data units to the data unit processors, so that the degree of variation of the number of copied data unit groups obtained for each combination of the data unit processors becomes smaller.