Memory Pipeline Predictive Execution for Idle Time Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semiconductor memory systems face challenges in programming data efficiently, leading to prolonged wait times for host devices due to limitations in data transfer and processing speeds within the memory system.
Innovation Solution
The implementation of an integrated memory system with a controller and memory die configuration, utilizing NAND flash memory and error correction circuits, along with advanced signal pathways and control circuits, enables concurrent programming and verification of memory cells, optimizing data transfer and reducing idle times through predictive data unavailability detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data programming is performed sequentially with verification, then programming accuracy is improved, but programming time increases
Solution Approach 1:
The patent applies preliminary action by performing verification operations in parallel with programming operations. The verification circuit begins verifying programmed data before the programming process is fully complete, eliminating the need to wait for sequential completion. This allows the system to prepare verification activities in advance and execute them concurrently, thus maintaining programming accuracy while reducing total programming time.
Solution Approach 2:
The patent implements continuity of useful action by ensuring that verification operations continue without interruption alongside programming operations. The system maintains both programming and verification activities running simultaneously through dedicated circuits and pipelines, eliminating idle time and ensuring continuous productive work throughout the data storage process.
2Reliability
If the controller waits for data availability from the host, then data transfer accuracy is improved, but controller idle time increases
Solution Approach 1:
The patent applies preliminary action by implementing a data buffer that pre-stores incoming data from the host before the programming operation begins. This allows the controller to prepare data for programming in advance, eliminating waiting time while ensuring data integrity through buffered storage. The buffer acts as a preliminary preparation mechanism that decouples data reception from programming execution.
Solution Approach 2:
The patent uses a data buffer as an intermediary between the host and the programming circuit. This intermediary component receives and holds data from the host, allowing the controller to proceed with programming operations without waiting for host data availability. The buffer mediates the timing mismatch between data arrival and programming readiness, eliminating controller idle time while maintaining data transfer accuracy.
3Measurement precision
If the memory system processes data step-by-step, then processing precision is improved, but overall system productivity decreases
Solution Approach 1:
The patent applies segmentation by dividing the memory system into separate functional units: a programming circuit, a verification circuit, and a data buffer. Each segment handles specific tasks independently and simultaneously. The programming circuit programs data while the verification circuit verifies it in parallel, and the buffer manages data flow. This segmentation enables multiple precision-critical operations to occur concurrently, maintaining processing precision while dramatically improving system productivity.
Solution Approach 2:
The patent transitions from sequential one-dimensional processing to parallel multi-dimensional processing. Instead of processing data in a single sequential step, the system creates multiple parallel processing dimensions where programming, verification, and data management occur simultaneously in different operational dimensions. This dimensional expansion allows precision operations to run in parallel without interfering with each other, boosting productivity while maintaining precision.
Data Source
AI summary
An apparatus is provided that includes a memory die including a pipeline circuit coupled to a memory structure. The memory die is configured to execute a first command by receiving in the pipeline circuit data to be written to the memory structure, processing the received data in the pipeline circuit and providing the processed data to the memory structure, predicting that the pipeline circuit has completed processing the received data, and ending execution of the first command based on the prediction.