Segmented Data Flow Graph for Concurrent Node Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data flow graphs are limited in parallelization, as they only allow concurrent evaluation of independent nodes identified by declared dependencies, restricting the system's efficiency in processing data flow schemes.
Innovation Solution
A modified data flow graph is introduced that identifies independent segments of data elements, allowing nodes to be evaluated concurrently by determining dependencies between segments rather than relying solely on declared dependencies, thereby enabling further parallelization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the system relies on declared dependencies for parallelization, then the evaluation order is determined correctly, but the parallelization efficiency is limited
Solution Approach 1:
The patent segments data elements into independent segments and divides node invocations into segment-specific invocations. This allows the system to identify independent segments that can be evaluated in parallel without requiring comprehensive declared dependencies between all nodes, thereby improving evaluation efficiency while managing complexity through structured segmentation.
2Speed
If nodes are evaluated sequentially based on declared dependencies, then dependency correctness is maintained, but processing speed is reduced
Solution Approach 1:
The patent performs preliminary actions by dividing data elements into segments and generating segment-specific node invocations before evaluation. This pre-segmentation allows the system to identify independent segments that can be evaluated concurrently, reducing overall evaluation time while maintaining dependency correctness through the segment-based invocation structure.
3Adaptability or versatility
If the system uses traditional data flow graph parallelization, then implementation simplicity is maintained, but the degree of parallelization is insufficient
Solution Approach 1:
The patent introduces a new dimension to traditional data flow graph parallelization by segmenting data elements and creating segment-specific node invocations. This transforms the parallelization approach from node-level to segment-level, enabling finer-grained parallelization and improving adaptability without significantly increasing overall system complexity.
Data Source
AI summary
Provided herein are methods, systems, and computer products for evaluating nodes concurrently using a modified data flow graph. The modified data flow graph can identify independent nodes that can run as separate tasks. However, rather than relying on declared dependencies, embodiments herein can determine dependencies between segments of data elements in a data flow graph, and modify the data flow graph to take advantage of the determined dependencies. In such embodiments, the data elements can be divided into segments. By separating data elements into segments, nodes that previously depended on each other can be evaluated concurrently when independent segments are identified.