Parallel Retry Scheduling for Non-Converging Data Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data processing systems face increased latency and inefficiency when dealing with non-converging data sets, as existing methods do not effectively manage the processing of varying data quality and convergence issues, leading to delayed reporting and reduced throughput.
Innovation Solution
The implementation of priority-based data processing systems that utilize a data detector circuit, data decoder circuit, and scheduling circuit to apply data detection and decode algorithms, allowing for on-the-fly retry processing and out-of-order processing, where higher quality data sets are prioritized and processed in parallel with lower quality sets, using an input buffer to manage and reprocess non-converging data sets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data sets are processed in sequential order with retry processing, then processing completeness is improved, but average latency increases significantly
Solution Approach 1:
The system dynamically adjusts processing order based on convergence status. Data sets that converge successfully are processed and reported immediately, while non-converging data sets are identified and reprocessed in parallel retry operations. This dynamic reordering eliminates the fixed sequential bottleneck, reducing average latency while maintaining processing completeness through systematic retry handling.
Solution Approach 2:
The processing system segments data sets into two distinct groups: converging data sets and non-converging data sets. This segmentation allows independent processing paths - converging sets proceed through normal processing while non-converging sets are routed to retry processing. The segmentation enables parallel execution of different processing strategies, improving overall throughput and reducing latency.
2Reliability
If retry processing is performed sequentially for non-converging data sets, then processing thoroughness is improved, but system throughput decreases
Solution Approach 1:
The system maintains continuous useful action by overlapping retry processing with normal processing operations. While retry processing handles non-converging data sets, the data detector and decoder continue processing new incoming data sets without interruption. This continuity ensures that processing thoroughness is maintained for all data sets while system throughput remains high due to uninterrupted processing streams.
Solution Approach 2:
The system adds a temporal dimension to processing by implementing parallel retry operations. Instead of processing retries in a single sequential dimension, the system schedules retry processing to occur in parallel with normal processing operations. This dimensional change allows multiple processing streams to coexist, improving throughput while maintaining thoroughness through dedicated retry handling.
3Reliability
If all data sets are processed through complete detection and decoding iterations, then processing completeness is improved, but processing time for individual data sets increases
Solution Approach 1:
The system applies partial processing to data sets that converge successfully, performing only the necessary detection and decoding iterations required for successful convergence rather than forcing complete fixed iterations. This partial action approach reduces processing time for converging data sets while maintaining completeness. Non-converging data sets receive excessive action through multiple retry processing attempts to ensure eventual completeness.
Data Source
AI summary
Systems, circuits, devices and/or methods related to systems and methods for data processing, and more particularly to systems and methods for priority based data processing.