Mapped RAID Drive Extent Allocation in Partnership Groups

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Traditional RAID data storage systems face limitations in adding new disks and suffer from long rebuild times due to write bandwidth bottlenecks, especially as individual disk capacity increases, leading to impractical inflexibility and increased risk of data loss during concurrent failures.

Innovation Solution

The mapped RAID technology allows for the addition of individual storage drives and reduces rebuild time by spreading rebuild operations across multiple drives, splitting the drive group into partnership groups to manage drive extent allocations efficiently and minimize data loss risks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If traditional RAID systems are used to add new storage capacity, then data redundancy and fault tolerance are maintained, but new disks can only be added in increments equal to the RAID stripe width (e.g., 5 disks for 4D+1P RAID-5), resulting in inflexibility and impracticality

Engineering Contradiction:

Improveflexibility in adding storage capacityVSAvoidRAID configuration constraints

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the traditional monolithic RAID group into multiple independent partnership groups, where each partnership group is a self-contained RAID group. This allows individual drives to be added to specific partnership groups without requiring addition to all groups, enabling flexible capacity expansion in increments of single drives rather than full stripe widths

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system dynamically creates new partnership groups when existing groups reach their maximum drive capacity. This dynamic adaptation allows the storage system to continuously accept individual drive additions regardless of the original RAID configuration constraints, transforming the static RAID structure into a flexible, growing architecture

Inventive Principle:

Principle #15Dynamics

2Quantity of substance

If the number of storage drives in a RAID group is increased to enhance storage capacity, then storage capacity increases, but rebuild time increases due to write bandwidth bottlenecks when writing to a single dedicated spare disk

Engineering Contradiction:

Improvestorage capacityVSAvoidrebuild time

Core Design Contradiction:

Quantity of substanceVSLoss of time

Solution Approach 1:

The patent segments the rebuild operation across multiple independent partnership groups, each with its own spare drive. Instead of writing rebuilt data to a single dedicated spare disk, the system distributes rebuild operations across multiple spare drives in different partnership groups, enabling parallel rebuild processing and eliminating the write bandwidth bottleneck

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent combines multiple partnership groups into a single logical storage system while maintaining physical independence of each group. This allows the system to leverage the combined storage capacity of all groups while simultaneously utilizing the combined rebuild capability of all spare drives across groups, achieving both high capacity and fast rebuild times

Inventive Principle:

Principle #5Merging (Combining)

3Quantity of substance

If the number of storage drives is increased to enhance storage capacity, then storage capacity increases, but the risk of data loss increases due to higher probability of concurrent drive failures

Engineering Contradiction:

Improvestorage capacityVSAvoiddata loss risk

Core Design Contradiction:

Quantity of substanceVSReliability

Solution Approach 1:

The patent segments the storage system into multiple independent partnership groups, where failure isolation is achieved at the group level. When a drive fails, only its specific partnership group is affected, while other groups continue operating normally. This segmentation prevents cascading failures across the entire system, maintaining high reliability even as total system capacity increases

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The partnership group structure acts as an intermediary layer between individual drives and the overall storage system. This intermediary architecture provides failure containment, where the partnership group boundary prevents failure propagation from one group to others, thereby protecting the majority of the storage system from single-point failures

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS10296252B1Reducing drive extent allocation changes while splitting a group of storage drives into partnership groups in response to addition of a storage drive to an array of storage drives in a data storage system that uses mapped RAID (redundant array of independent disks) technology

Publication Date: 2019.05.21 EMC IP HLDG CO LLC

AI summary

A mapped RAID system that reduces drive extent allocation changes when splitting a group of storage drives into a first partnership group and a second partnership group. A first drive is assigned to the first partnership group, and then additional drives are assigned to the first partnership group from the unassigned drives, such that each drive assigned to the first partnership group has a lowest movement count of all unassigned drives. A drive's movement count represents a count of drive extent allocations that must be changed if the drive is added to the first partnership group. After each drive is assigned to the first partnership group, the indication of the drive is removed from the list of unassigned drives. A group of RAID extent entries for the first partnership group consists of RAID extent entries indicating relatively higher total numbers of drive extents located in the first partnership group.