NVDIMM Metadata Storage for SSD Cache Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing SSD cache solutions waste expensive high-endurance SSDs due to excessive write requirements, leading to increased user costs and potential impact on response speed.
Innovation Solution
Implementing a hybrid SSD cache that stores metadata in a Non-Volatile Dual-In-Line Memory Module (NVDIMM) instead of SSDs, allowing for the use of cheaper medium- and low-endurance SSDs, with high-endurance SSDs in RAID groups for write caching and low-endurance SSDs as read caches.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If high-endurance SSDs are used for metadata storage in SSD cache, then reliability is improved, but device cost increases and SSD waste occurs
Solution Approach 1:
The patent introduces NVDIMM as an intermediary component between DRAM and SSD to store metadata. NVDIMM provides non-volatile storage with sufficient durability for metadata without requiring the high endurance of expensive SSDs, thus resolving the contradiction between reliability and cost.
Solution Approach 2:
The patent applies different storage media to different parts of the cache system based on their specific requirements: NVDIMM for metadata (requiring non-volatility and moderate durability), SSD for data blocks (requiring high capacity and speed), and DRAM for frequently accessed data (requiring highest speed). This local optimization resolves the contradiction by matching storage characteristics to functional requirements.
2Productivity
If SSD cache capacity is increased, then productivity is improved, but device complexity increases due to hardware limitations
Solution Approach 1:
The patent segments the cache system into multiple layers with different characteristics: DRAM cache for hot data, SSD cache for warm data, and NVDIMM for metadata. This segmentation allows each component to operate within its optimal capacity range, avoiding the hardware limitations that would prevent a single large-capacity cache from achieving the same performance.
Data Source
AI summary
Embodiments of the present disclosure provide a method of storage management, a storage system and a computer program product. The method comprises determining whether a number of I/O requests for a first page in a disk of a storage system exceeds a first threshold. The method further comprises: in response to determining that the number exceeds the first threshold, caching data in the first page to a first cache of the storage system; and storing metadata associated with the first page in a Non-Volatile Dual-In-Line Memory Module (NVDIMM) of the storage system.