NVRAM Cache Availability via Dynamic Drive Set Reconfiguration

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Data storage systems with non-volatile random-access memory (NVRAM) caches face unavailability when all NVRAM drives in a virtual pair fail, as they cannot be reconfigured to maintain redundancy, leading to system downtime until the failed drives are replaced.

Innovation Solution

Configuring NVRAM drives into initial virtual drive sets that can be reconfigured into new virtual drive sets if all drives in an initial set fail, ensuring continuous operation by maintaining mirroring copies across remaining healthy drives.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If fixed physically paired NVRAM disks are used to form the non-volatile cache, then data redundancy is maintained through mirroring, but the system becomes unavailable when all drives in a pair fail

Engineering Contradiction:
Improvedata redundancyVSAvoidsystem availability
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent implements dynamic drive set reconfiguration where NVRAM drives are not permanently bound to fixed pairs but can be dynamically reassigned. When all drives in a virtual drive set fail, the system dynamically reconfigures remaining drives into new virtual drive sets, transitioning from static fixed pairing to dynamic flexible pairing to maintain system availability

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

Each NVRAM drive is designed to serve multiple functions and belong to multiple potential virtual drive sets. A single NVRAM drive can participate in different virtual drive set configurations depending on which drives are operational, making each drive universal rather than dedicated to a single fixed pair

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Reliability

If fixed disk pairs are used for the non-volatile cache, then redundancy is achieved through mirroring, but the system must remain offline until failed drives are replaced

Engineering Contradiction:
ImproveredundancyVSAvoidsystem downtime
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary reconfiguration actions automatically when drive failures are detected. Rather than waiting for manual intervention and drive replacement, the system proactively reorganizes remaining drives into new virtual drive sets before the failure becomes critical, minimizing downtime through automated preliminary recovery actions

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system performs self-service recovery by automatically detecting failed drives and reconfiguring remaining drives into new virtual drive sets without requiring manual intervention. The system serves itself by maintaining its own availability through automated reconfiguration, eliminating the need for operator action during failure recovery

Inventive Principle:
Principle #25Self-service

3Productivity

If NVRAM drives are reconfigured into new virtual drive sets after failure, then cache availability is maintained, but the system complexity increases

Engineering Contradiction:
Improvecache availabilityVSAvoidreconfiguration complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The system implements continuous feedback monitoring of drive health status and automatically triggers reconfiguration when failures are detected. This closed-loop feedback mechanism manages the complexity of reconfiguration by only activating it when necessary, rather than requiring constant manual management of drive assignments

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system changes the logical parameter of drive grouping from fixed physical pairs to flexible virtual drive sets. By changing how drives are organized and assigned (the grouping parameter) rather than the physical drives themselves, the system maintains simplicity while enabling dynamic reconfiguration for high availability

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS11513956B2Maintaining availability of a non-volatile cache
Publication Date: 2022.11.29 EMC IP HLDG CO LLC
  • US11513956B2 patent drawing
  • US11513956B2 patent drawing
  • US11513956B2 patent drawing

AI summary

A technique maintains availability of a non-volatile cache. The technique involves arranging a plurality of non-volatile random-access memory (NVRAM) drives into initial drive sets that form the non-volatile cache. The technique further involves detecting a failed initial drive set among the initial drive sets. The plurality of NVRAM drives now includes failed NVRAM drives that belong to the failed initial drive set and remaining non-failed NVRAM drives. The technique further involves, in response to detecting the failed initial drive set, re arranging the remaining non-failed NVRAM drives of the plurality of NVRAM drives into new drive sets that form the non-volatile cache.