Hybrid Flash Memory Decoding for Throughput and Power Balance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional flash memory storage systems face challenges in implementing soft iterative decodable codes due to increased power usage and decreased throughput, as they are optimized for hard decision decoding and struggle to obtain necessary soft data for decoding.
Innovation Solution
A hybrid decoding approach is employed using a primary decoder of lower complexity and higher throughput, which activates an auxiliary decoder when the primary decoder fails, allowing for efficient soft iterative decoding in storage systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If soft iterative decoding techniques are employed to maintain consistent corrected bit error rates, then reliability is improved, but device complexity increases substantially
Solution Approach 1:
The decoder is segmented into two distinct components: a primary decoder using hard decision decoding for fast, low-complexity operation, and an auxiliary decoder using soft iterative decoding for high-reliability correction. This segmentation allows each decoder to be optimized for its specific function, resolving the contradiction between reliability and complexity.
Solution Approach 2:
The system dynamically switches between hard and soft decision decoding modes based on the error characteristics of the received data. The controller determines whether to activate the primary or auxiliary decoder based on error rates and other parameters, allowing the system to adapt its complexity level to the actual reliability needs.
2Reliability
If soft iterative decoding is used to achieve rigorous reliability requirements, then corrected bit error rate consistency is improved, but power consumption increases
Solution Approach 1:
The power consumption issue is addressed by segmenting the decoding function into two parts with different power profiles. The primary hard decision decoder consumes less power and handles most decoding tasks, while the power-intensive auxiliary soft iterative decoder is activated only when needed for difficult-to-decode data, thus resolving the contradiction between reliability and power consumption.
Solution Approach 2:
Instead of always applying full soft iterative decoding, the system applies partial action by using hard decision decoding for the majority of cases and reserving soft iterative decoding only for the minority of cases where hard decoding fails. This partial application of the more complex technique reduces overall power consumption while maintaining reliability.
3Reliability
If soft iterative decoding is implemented to maintain consistent corrected bit error rates, then reliability is improved, but throughput decreases
Solution Approach 1:
The throughput problem is resolved by segmenting the decoding workload. The primary hard decision decoder provides high throughput for most data, while the auxiliary soft iterative decoder handles only the subset of data that requires more intensive processing. This segmentation maintains overall system throughput while achieving the required reliability.
Solution Approach 2:
The system dynamically adjusts its decoding approach based on the characteristics of the received data. By controlling the switch between hard and soft decision decoding modes, the system optimizes throughput by avoiding unnecessary soft iterative decoding while ensuring reliability when needed.
Data Source
AI summary
Systems and methods for decoding data using a decoder that includes a primary decoder and an auxiliary decoder are provided. A codeword is retrieved from a storage device. A primary decoder attempts to decode the codeword using hard data associated with the codeword. If the primary decoder fails, an indication of the failure may be received by a decoder controller, which activates an auxiliary decoder. The auxiliary decoder attempts to decode the codeword using either hard data or soft data associated with the codeword. The primary decoder is designed to consume less power, consume less silicon area, and have a higher throughput than the auxiliary decoder. The primary decoder is configured to have a higher probability of successfully decoding a codeword, stored in the storage device, in the first attempt to decode the codeword, than failing and requiring the auxiliary decoder to decode the codeword.