3D NAND Read Retry via Metadata-Driven Voltage Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
3D NAND flash memory devices face reliability challenges due to aggressive scaling, leading to read operation failures, which can be exacerbated by factors like temperature changes and read disturb.
Innovation Solution
A memory system that selects from a plurality of read retry routines based on metadata, such as time stamps, temperatures, and read counts, to enhance read operations by adjusting the read voltage accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If 3D NAND flash memory is aggressively scaled to increase storage density, then storage density is improved, but read operation reliability deteriorates
Solution Approach 1:
The system performs preliminary sensing of effector values (temperature, time stamp, read count) before executing read operations. Based on these sensed values, the system pre-selects appropriate read retry routines from a library of routines, each optimized for specific effector conditions. This preliminary preparation allows the system to quickly switch to the correct retry routine if a read failure occurs, thereby maintaining high reliability despite aggressive scaling.
Solution Approach 2:
The system changes read operation parameters dynamically based on sensed effector values. Different read retry routines use different read voltages, pulse widths, and timing parameters optimized for specific temperature ranges, time since programming, and read count conditions. By adjusting these parameters according to the current effector state, the system compensates for reliability degradation caused by scaling effects.
2Reliability
If multiple read retry routines are implemented to handle different failure conditions, then read operation reliability is improved, but device complexity increases
Solution Approach 1:
Instead of using a single complex retry mechanism for all failure conditions, the system creates multiple specialized read retry routines, each optimized for specific local conditions (temperature ranges, time since programming, read count thresholds). The sensing module measures local effector values and selects the appropriate routine, thereby distributing complexity across multiple simple, specialized routines rather than one complex universal routine.
Solution Approach 2:
The system introduces an intermediary sensing module that measures effector values (temperature, time stamp, read count) and an intermediary selection logic that maps these values to appropriate read retry routines. This intermediary layer decouples the complexity of handling multiple failure conditions from the core read operation, making the system more manageable and maintainable while improving reliability.
3Reliability
If effector values are sensed and analyzed before read operations, then read operation reliability is improved, but operation time increases
Solution Approach 1:
The system performs sensing of effector values and selection of read retry routines as a preliminary action before the actual read operation. The effector values (temperature, time stamp, read count) are sensed once and stored, then used to pre-select the appropriate read retry routine. This preliminary preparation eliminates the need for complex real-time analysis during the read operation itself, thereby minimizing the time penalty while maintaining improved reliability.
Solution Approach 2:
The system senses only the most critical effector values (temperature, time stamp, read count) that have the greatest impact on read reliability, rather than measuring all possible parameters. This partial sensing approach provides sufficient information for reliable routine selection without the time overhead of comprehensive measurement, achieving a balance between reliability improvement and operation time.
Data Source
AI summary
The present disclosure provides a memory system for selecting from among a plurality of read retry routines based on metadata. The memory system can include one or more memory devices and a memory controller. The memory controller can also detect a failure of a read operation. The memory controller can also analyze a set of values that correspond to a set of effectors of the read operation. The memory controller can select one or more read retry routines from a plurality of read retry routines based on the analyzing. Each of the plurality of read retry routines can associated with a different effector from the set of effectors and a read voltage that corresponds to the different effector. The memory controller can also perform the selected one or more read retry routines at the portion of the one or more memory devices to negate the failure of the read operation.