Memory Die Write Routing During Data Folding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Concurrent host write operations with data folding in memory systems often result in suspended or delayed host writes due to busy memory dies, leading to reduced performance and buffer overflow.
Innovation Solution
Data intended for busy memory dies is written to accessible dies, allowing concurrent host writes and data folding operations, with a buffer allocation for temporary storage and data reordering to accommodate sequential reads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data folding operations are performed on memory dies, then storage efficiency is improved, but host write operations are suspended or delayed due to busy dies
Solution Approach 1:
The memory system is divided into multiple independent memory dies that can operate concurrently. Data folding operations are performed on a first subset of dies while host write operations are performed on a second subset of dies, allowing parallel execution and eliminating the suspension issue.
Solution Approach 2:
A buffer is introduced as an intermediary component to temporarily store data during transitions. The buffer allocation allows data to be staged for folding operations without blocking host write operations, and data reordering in the buffer accommodates sequential read requirements.
2Productivity
If host write operations are performed concurrently with data folding, then host write performance is improved, but buffer overflow occurs due to insufficient temporary storage
Solution Approach 1:
The buffer management system dynamically allocates and manages buffer spaces based on operational needs. Buffer spaces are allocated for data folding operations and data reordering operations separately, allowing flexible utilization of buffer capacity without fixed reservations, thus preventing overflow while maintaining concurrent operations.
3Loss of time
If data is written to accessible memory dies during folding operations, then host write latency is reduced, but data reordering is required to accommodate sequential reads
Solution Approach 1:
Data reordering is performed in advance during the buffer staging phase before data is transferred to the final destination. By reordering data in the buffer to match sequential read patterns before the folding operation completes, the system eliminates the need for complex post-processing reordering operations.
Data Source
AI summary
Methods, systems, and devices for techniques for improving host write performance during data transferring (e.g., folding) are described. A memory system may determine to transfer first data from one or more source data blocks of the memory system to one or more destination data blocks, where the source data blocks and the destination data blocks are associated with one or more memory dies of a set of memory dies of the memory system. The memory system may also receive, from a host system, a command to write second data to a first memory die of the one or more memory dies, and write, concurrent with transferring the first data, the second data to a second memory die of the set of memory dies different than the one or more memory dies based on the transfer of the first data being associated with the first memory die.