RAID Controller DRAM Pin Cache Offload to SAS Switch Flash

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

RAID systems face data loss and performance issues due to pinned cache data being held in memory when a RAID controller card fails, as other virtual disks must operate in write-through mode, affecting I/O throughput and user-initiated background operations.

Innovation Solution

A method and system that offloads controller DRAM pin cache contents to a flash module on an SAS switch, allowing pinned cache data to be flushed to a new disk upon replacing the failed virtual disk, ensuring data integrity and enabling write-back mode for other virtual disks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If write back cache policy is used in RAID controller, then I/O throughput is improved, but data integrity is compromised when controller fails with pinned cache

Engineering Contradiction:
ImproveI/O throughputVSAvoiddata integrity
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent introduces an intermediary mechanism (SAS switch with flash module or external storage) to store pinned cache data temporarily when controller failure occurs. This intermediary allows the system to maintain write-back mode for normal operations while providing a safety net for data integrity during failures, resolving the contradiction between throughput improvement and reliability concern.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system performs preliminary action by proactively saving pinned cache data to external storage (SAS switch flash module or separate disk) before controller failure occurs. This preliminary data preservation enables the system to recover after failure without data loss, allowing continuous use of write-back cache policy while maintaining data integrity.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If pinned cache data is held in memory during controller failure, then data can be recovered, but other virtual disks must operate in write-through mode reducing I/O throughput

Engineering Contradiction:
Improvedata recovery capabilityVSAvoidI/O throughput
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent extracts the pinned cache data protection function from the RAID controller memory and relocates it to external storage (SAS switch flash module or separate disk). This extraction allows the controller memory to be freed for normal cache operations while the external storage provides dedicated protection for pinned cache data, enabling both reliability and performance to coexist.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system segments the cache protection function into separate components: the RAID controller handles normal I/O operations with write-back mode, while the SAS switch flash module or external storage independently handles pinned cache data preservation. This segmentation allows each component to specialize, maintaining high throughput for normal operations while ensuring data recovery capability.

Inventive Principle:
Principle #1Segmentation

3Reliability

If battery or supercap solution is used to hold pinned cache data, then data integrity is maintained during power failure, but device complexity and cost increase

Engineering Contradiction:
Improvedata integrity during power failureVSAvoidcontroller complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent uses an intermediary external storage device (SAS switch flash module or separate disk) to replace the need for battery or supercap solutions within the controller. This external intermediary handles the power failure protection function, allowing the controller itself to remain simpler without requiring complex power backup mechanisms while still maintaining data integrity.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system employs self-service by using existing external storage resources (SAS switch flash module or available disks) to provide pinned cache protection without requiring additional specialized components like batteries or supercaps in the controller. The external storage serves its own purpose while simultaneously providing cache protection, reducing overall system complexity.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS9026846B2Data recovery in a raid controller by offloading contents of DRAM to a flash module on an SAS switch
Publication Date: 2015.05.05 AVAGO TECHNOLOGIES INTERNATIONAL SALES PTE LTD
  • US9026846B2 patent drawing
  • US9026846B2 patent drawing
  • US9026846B2 patent drawing

AI summary

Disclosed is a system and method for providing data integrity for pinned cache even if a RAID controller card fails while it has pinned cache or a memory module goes bad. A controller is enabled to use complete cache lines even if pinned cache is present, thereby enabling other virtual disks to run in write-back mode when pinned cache is present.