Channel Adapter Check Codes for Variable-Length Data Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In mixed information processing systems combining main-frame and open systems, existing data storage technologies fail to accurately detect errors during data updates in variable-length data formats, particularly when data is stored in fixed-length formats, leading to undetected faults in updated logical blocks or fields.
Innovation Solution
A method and channel adapter that create and utilize field-checking codes (FCC) and block-checking codes (BCC) to manage and write variable-length data in a fixed-length format, ensuring accurate error detection by calculating and appending these codes to each field and logical block during data transfer and storage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If data is stored in fixed-length format in a disc array unit, then data access speed and fault resistance are improved, but error detection capability for variable-length data updates is deteriorated
Solution Approach 1:
The patent segments the error checking mechanism into two levels: field-checking codes for individual fields and block-checking codes for logical blocks. This segmentation allows the system to maintain fixed-length storage structure while enabling granular error detection at both field and block levels, resolving the contradiction between storage efficiency and error detection capability.
Solution Approach 2:
The patent applies preliminary action by pre-calculating and storing field-checking codes and block-checking codes along with the data. When data is read, these pre-computed codes enable immediate error detection without requiring complex re-calculation, thus maintaining high data access speed while ensuring reliable error detection.
2Reliability
If block-checking code is attached for each logical block, then data integrity is improved, but fault detection for partial updates is deteriorated
Solution Approach 1:
The patent divides the error checking structure into hierarchical segments: field-checking codes for individual fields and block-checking codes for logical blocks. This allows precise fault detection at the field level while maintaining overall block integrity verification, enabling detection of partial updates that a single block-level code cannot detect.
Solution Approach 2:
The patent adds another dimension to error detection by introducing field-level checking codes alongside block-level codes. This creates a two-dimensional error detection framework that can precisely identify whether faults occur at the field level or block level, significantly improving fault detection precision.
3Measurement precision
If field-checking code is created for each field during data update, then error detection precision is improved, but data update time is deteriorated
Solution Approach 1:
The patent applies partial action by creating field-checking codes only for fields that are actually updated, rather than recalculating codes for all fields. This selective approach maintains high error detection precision for modified data while significantly reducing the time overhead compared to full recalculation.
Solution Approach 2:
The system uses the existing field data to self-generate the field-checking codes through deterministic calculations. This self-service mechanism eliminates the need for external verification processes, achieving precise error detection with minimal additional computational time.
4Adaptability or versatility
If data is stored in variable-length format, then data management flexibility is improved, but storage efficiency and access speed are deteriorated
Solution Approach 1:
The patent segments variable-length data into fixed-length logical blocks, with each block containing fields and their associated checking codes. This segmentation allows the system to maintain the flexibility of variable-length data management while achieving the storage efficiency and access speed of fixed-length structures through standardized block handling.
Data Source
AI summary
A method of writing data includes receiving a record of a variable-length data format, creating a field-checking code for each field of the record received, creating a block-checking code in units of the fixed-length data for the data received, and writing data by reading the record, assembling fixed length data that includes the field-checking code and the block-checking code by using the field-checking code and the block-checking code, and transferring the data to a cache memory.