Storage Class Memory Persistent Cache for Host Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage systems face challenges in maintaining performance due to limited volatile memory resources close to the host server processor, leading to potential data loss during reboots or power outages and restricted cache sizes due to high RAM costs.
Innovation Solution
Implementing a non-volatile cache in host computers for dedicated host application data caching, with a destaging program that manages data movement based on a destaging plan, including sending dirty page messages, starting aging timers, and responding to storage array messages for controlled destaging and caching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If volatile memory (RAM) is used for host application data caching, then data access speed is improved, but data loss risk increases during reboots or power outages
Solution Approach 1:
The patent changes the physical state parameter of the cache from volatile (RAM) to non-volatile (SCM), allowing the cache to maintain data persistence while preserving fast access characteristics. This parameter change resolves the contradiction by enabling the cache to provide both rapid data access and protection against data loss during system reboots or power failures.
Solution Approach 2:
The patent introduces SCM as an intermediary layer between the host application and the storage array, functioning as a persistent cache that mediates between the speed requirements of the host and the reliability requirements of persistent storage. This intermediary provides both fast access similar to RAM and data persistence similar to storage.
2Quantity of substance
If volatile memory capacity is increased for larger cache, then caching capability is improved, but cost increases due to high RAM prices
Solution Approach 1:
The patent replaces expensive volatile memory (RAM) with more cost-effective non-volatile memory (SCM) for caching purposes. While SCM has different characteristics than RAM, it provides sufficient performance for caching while being more cost-effective, allowing larger cache capacities without proportionally increasing costs.
Solution Approach 2:
The patent changes the memory type parameter from volatile RAM to non-volatile SCM, which alters both the cost structure and the capacity available for caching. This parameter change enables larger cache sizes at lower cost while maintaining acceptable performance characteristics for host application data caching.
3Speed
If cache proximity to host processor is increased, then data access speed is improved, but memory resource limitations worsen
Solution Approach 1:
The patent changes the volatility parameter of the cache memory from volatile to non-volatile, which fundamentally alters the resource characteristics. This allows the system to allocate larger cache capacities without the same resource constraints that apply to volatile memory, as SCM does not require the same power and resource management as RAM.
Solution Approach 2:
The patent transitions from a two-level memory hierarchy (volatile cache and persistent storage) to a three-level hierarchy by inserting SCM as an intermediate layer. This dimensional change in the memory hierarchy allows for larger cache capacities while maintaining fast access paths, effectively resolving the resource limitation constraint.
Data Source
AI summary
A host server in a server cluster has a memory allocator that creates a dedicated host application data cache in storage class memory. A background routine destages host application data from the dedicated cache in accordance with a destaging plan. For example, a newly written extent may be destaged based on aging. All extents may be flushed from the dedicated cache following host server reboot. All extents associated with a particular production volume may be flushed from the dedicated cache in response to a sync message from a storage array.