Nonvolatile Memory Wraparound Read for Cache Miss Handling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computing systems face challenges in maintaining data integrity and low power consumption while ensuring compatibility with volatile RAM, particularly in using flash memory as a data storage device or working memory, where cache misses lead to inefficiencies in data retrieval.
Innovation Solution
A computing system and read method that utilize a wraparound scheme to fill cache lines with page data from nonvolatile memory, allowing for efficient output of critical words to the processor, even in cases of cache misses, by sequentially reading and error-correcting data from nonvolatile memory and storing it in cache memory.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If flash memory is used as a data storage device or working memory to maintain compatibility with volatile RAM, then data integrity and low-power characteristics are improved, but cache misses lead to inefficiencies in data retrieval
Solution Approach 1:
The patent applies preliminary action by pre-loading page data into the cache memory before the processor actually needs it. When a read request is made, the system checks if the data is in the cache; if not, it pre-loads the entire page containing the requested data. This ensures that subsequent accesses to other data in the same page can be served quickly from cache, improving overall data retrieval efficiency while maintaining the reliability benefits of flash memory.
2Loss of time
If page data is read from nonvolatile memory in a wraparound scheme, then critical words can be output timely to the processor, but the complexity of the read operation increases
Solution Approach 1:
The patent implements dynamics by using a wraparound read scheme where the reading position dynamically wraps around to the beginning of the page when reaching the end. Instead of linear sequential reading, the system can start reading from any position and wrap around to access critical words earlier. This dynamic reading approach allows flexible positioning and reduces the time to output critical words, while the wrapping mechanism manages the complexity through a systematic modular approach.
3Reliability
If the cache line is filled with page data from nonvolatile memory, then data integrity is maintained through error correction, but the time required to fill the cache line increases
Solution Approach 1:
The patent applies continuity of useful action by performing error correction decoding continuously during the cache line filling process rather than as a separate step. As page data is read from nonvolatile memory and transferred to the cache line, error correction is performed in parallel or concurrently, ensuring data integrity is maintained throughout the transfer. This continuous approach prevents idle time and reduces the total time required to fill the cache line with corrected data.
Data Source
AI summary
A read method executed by a computing system includes a processor, at least one nonvolatile memory, and at least one cache memory performing a cache function of the at least one nonvolatile memory. The method includes receiving a read request regarding a critical word from the processor. A determination is made whether a cache miss is generated, through a tag determination operation corresponding to the read request. Page data corresponding to the read request is received from the at least one nonvolatile memory in a wraparound scheme when a result of the tag determination operation indicates that the cache miss is generated. The critical word is output to the processor when the critical word of the page data is received.