Tape Read Error Recovery Using a Reserved Buffer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional tape drive systems face inefficiencies in reading data due to high error rates, leading to prolonged read times and potential media damage, as they require multiple retries and backhitches to correct errors using error correction codes (ECC) and reconstructive error recovery procedures (ERP).
Innovation Solution
Implementing a system with a reserved data buffer that stores correctable data sets using different hardware settings during repeated reads, allowing for aggregation and correction of errors using C1 and C2 ECC, and enabling reconstructive error recovery to assemble complete data sets efficiently, reducing the need for multiple backhitches and improving data recovery time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple ERP retries are performed to correct high error rates, then data integrity is improved, but read time increases and media damage worsens
Solution Approach 1:
The system performs preliminary actions by pre-allocating a reserved buffer specifically for ERP operations and proactively identifying data sets that will require error recovery. This allows the system to prepare recovery resources in advance rather than reacting to errors as they occur, reducing the time penalty of multiple retries.
Solution Approach 2:
The system segments the buffer into a reserved portion specifically dedicated to ERP operations, separating error-prone data processing from normal data flow. This segmentation allows independent optimization of error recovery operations without impacting overall system throughput, enabling parallel processing of good data sets while recovering problematic ones.
2Reliability
If multiple ERP retries are performed to correct high error rates, then data integrity is improved, but media damage worsens
Solution Approach 1:
The system identifies data sets requiring ERP in advance and marks them for recovery before attempting multiple retries. This preliminary identification allows the system to concentrate recovery efforts on specific problematic areas rather than repeatedly scanning entire tapes, minimizing unnecessary media access and reducing cumulative media damage from excessive backhitches.
Solution Approach 2:
The system skips over data sets that are already determined to be recoverable or are currently being processed, rushing through them without performing full ERP sequences. This selective approach avoids redundant operations on already-handled data, reducing unnecessary media access and minimizing further media wear.
3Productivity
If a reserved buffer is allocated for ERP operations, then data recovery efficiency is improved, but device complexity increases
Solution Approach 1:
The reserved buffer serves multiple functions: it stores data sets identified for ERP, holds corrected data during recovery operations, and can be reused across multiple ERP cycles. This multi-functionality maximizes the utility of the allocated buffer space, justifying the added complexity by providing substantial productivity gains through efficient resource utilization.
Solution Approach 2:
The system implements a buffer management strategy where data in the reserved buffer is discarded and recovered in controlled cycles. Once ERP operations complete for a particular data set, the buffer space is reclaimed and reused for the next recovery operation. This systematic discarding and recovering pattern simplifies buffer management compared to maintaining separate permanent storage for each error-prone data set.
Data Source
AI summary
In one embodiment, a system includes a processor configured to execute logic, the logic being configured to read a plurality of data sets, each data set including a plurality of portions which combine together to wholly form the data set, and reread at least a first uncorrected data set using a different setting in an error recovery procedure (ERP) when an error in the first uncorrected data set is not correctable using C2-ECC or an error in any portion of the first uncorrected data set is not correctable using C1-ECC. Other systems and methods for reading data from tape using a reconstructive ERP to reduce backhitches are presented according to more embodiments.