Block-Based ECKD Data Protection via Alignment and Trailer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current FICON and HPF protocols are limited in adopting block-based end-to-end data protection standards for variable-length ECKD data, leading to inefficiencies and reduced productivity in computing storage environments.
Innovation Solution
Aligning information units (IUs) in a block boundary format and adding a SCSI standard trailer with protection data to each block, while transitioning from IU-based to block-based T10-DIF protection schemes, ensuring true end-to-end data protection and avoiding costly data transformation between data-in-flight and data-at-rest.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If FICON/HPF protocols are used for data transfer, then compatibility with existing systems is maintained, but block-based end-to-end data protection cannot be implemented
Solution Approach 1:
The patent embeds block-based protection structures within the existing FICON/HPF information unit framework. Protection data is nested inside the IU structure, allowing block-level security features to coexist with the legacy protocol without requiring complete protocol replacement, thus resolving the contradiction between implementing modern protection and maintaining compatibility.
Solution Approach 2:
The invention modifies the structural parameters of the information unit by introducing block boundary alignment and protection trailer fields. By changing the data organization parameters from pure IU-based to block-based structure, the system achieves end-to-end protection while still using the FICON/HPF transport mechanism.
2Productivity
If IU-based protection is used, then existing FICON/HPF protocols are utilized, but data transformation costs increase and productivity decreases
Solution Approach 1:
The patent segments the data protection approach into two independent layers: the FICON/HPF information unit layer for transport and the block-based T10-DIF protection layer for security. This segmentation allows each layer to operate independently, eliminating the need for costly transformations between different protection schemes and improving transfer efficiency.
Solution Approach 2:
Instead of transforming protection data between different formats, the invention uses block copying where the same block-based protection structure is applied consistently throughout the data path. This eliminates transformation overhead and reduces energy consumption associated with data conversion.
3Adaptability or versatility
If variable-length ECKD data is transferred, then data flexibility is maintained, but block alignment becomes complex and protection implementation is difficult
Solution Approach 1:
The patent performs preliminary block alignment and protection trailer addition at the source before data transfer. By pre-processing the variable-length ECKD data into properly aligned blocks with protection structures, the complexity of alignment is shifted to the preparation phase rather than the transfer phase, simplifying the overall implementation.
Solution Approach 2:
The invention introduces block boundary alignment as an intermediary structure between the variable-length ECKD data and the fixed protection mechanism. This intermediary layer absorbs the complexity of variable length handling while presenting a uniform block-based interface to the protection system, thereby simplifying protection implementation.
Data Source
AI summary
For block based end-to-end data protection for extended count key data (ECKD) in a computing environment, information units (IU's) are aligned in a block boundary format. Block protection trailer data is added to each one of the IU's.