Managed NAND Compression for Cold Data Offload and Lower Bus Traffic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory systems face inefficiencies in managing inactive or cold data, leading to excess wear on storage systems and reduced performance due to the need for frequent data transfers and compression, which burdens the host processor and increases latency.
Innovation Solution
A method for identifying and managing cold data by writing it to storage systems in contiguous blocks or offload units, utilizing the storage system's processing capabilities to reduce host processor load and optimize data transfer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data is frequently written to non-volatile storage, then storage capacity is utilized, but wear on storage system increases
Solution Approach 1:
The patent creates compressed copies of cold data in non-volatile storage instead of writing original uncompressed data. The compression algorithm generates a smaller representation of the data that can be decompressed when needed, reducing the number of write operations and wear on storage media while maintaining full data capacity utilization
Solution Approach 2:
The patent changes the physical state of data by applying compression algorithms that transform uncompressed data into compressed formats. This parameter change reduces data size by 50-80%, allowing more data to be stored in the same physical space without increasing write operations, thereby improving storage utilization while reducing wear
2Volume of stationary object
If data is compressed before storage, then storage space is reduced, but processing time increases
Solution Approach 1:
The patent performs compression of cold data during idle periods or background processing before the data is actually needed. By preparing compressed data in advance during low-utilization periods, the system reduces storage space requirements without impacting user-perceived processing time when the data is accessed
Solution Approach 2:
The patent implements periodic compression of cold data based on access patterns and system idle periods. The compression operation is performed periodically rather than continuously, allowing the system to balance compression overhead with storage space benefits by compressing data during natural low-activity periods
3Productivity
If cold data is identified and tagged, then storage efficiency is improved, but system complexity increases
Solution Approach 1:
The patent implements automatic identification and tagging of cold data through monitoring access patterns and usage statistics. The system self-manages the identification process by tracking which data has not been accessed recently, eliminating the need for manual intervention or complex external management systems while improving storage efficiency
4Speed
If data is transferred as contiguous blocks, then transfer speed increases, but data organization complexity increases
Solution Approach 1:
The patent merges multiple small data transfers into single contiguous block transfers by identifying and grouping cold data that can be compressed and written together. This merging approach increases transfer speed by utilizing sequential write operations while the compression and grouping logic manages the organizational complexity
Data Source
AI summary
Apparatus and methods are disclosed, including providing available data operations for the storage system processor to a host processor, identifying data operations to be performed by the storage system processor, and assigning identified data operations to the storage system processor to reduce bus traffic between the host processor and the storage system processor, to improve host processor performance, and to reduce energy use by the host processor.