Multi-Plane Pre-Read Buffering for Sequential Memory Reads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Memory systems with larger quantities of planes per memory die experience misalignment with data chunk sizes, leading to decreased performance in sequential read operations due to the need for multiple read operations on different subsets of planes.
Innovation Solution
Implementing a buffer within the memory system to store data read from planes not associated with the current sequential read command, allowing subsequent output without additional read operations on the memory die.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If larger quantities of planes per memory die are used, then storage capacity and parallel read capability are improved, but misalignment with data chunk sizes occurs leading to decreased sequential read performance
Solution Approach 1:
The patent applies preliminary action by performing a pre-read operation that reads data from multiple planes into a buffer before the actual sequential read command is executed. This pre-positioning of data in the buffer allows subsequent read commands to be served without additional plane read operations, thereby resolving the misalignment issue between plane quantity and data chunk size while maintaining high sequential read performance
2Reliability
If multiple read operations are performed on different subsets of planes, then complete data retrieval is achieved, but latency and response time increase
Solution Approach 1:
The patent performs a preliminary multi-plane read operation that loads data into a buffer before the actual read command is completed. This pre-loading ensures that all necessary data is available in the buffer for subsequent read operations, eliminating the need for additional read operations and reducing latency while ensuring complete data retrieval
Solution Approach 2:
The patent introduces a buffer as an intermediary component between the memory planes and the read command interface. This buffer acts as a mediator that decouples the fixed plane structure from the variable data chunk size requirements, allowing efficient data retrieval without requiring multiple read operations on different plane subsets
Data Source
AI summary
Methods, systems, and devices for supporting improved sequential read performance using plane pre-read are described. A memory system may include a buffer configured to store data read from planes of a memory device. A host system may issue a read command for first data stored sequentially at a set of memory devices, and a controller may perform multi-plane read operations in response to the read command. For a last multi-plane read operation, data read from a first subset of planes from the memory device may be output to the host device, while data read from a second subset of planes from the memory device may be stored at the buffer. The host device may issue another read command for second data, and the controller may output the data stored at the buffer in response to the read command, without performing another read operation on the memory device.