Physical Extent Manager for Decoupled Storage Scaling

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current data storage solutions, such as Data Domain's scale-up architecture, require large and complex disk farms to meet capacity and performance needs, limiting flexibility and scalability, especially in virtual environments.

Innovation Solution

The method involves reassembling local disk managers and array groups using a Physical Extent Manager (PEM) that manages virtual disks across nodes, allowing for dynamic scaling and resilience by decoupling array groups from specific nodes, enabling flexible data storage and migration.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If scale-up architecture is used to meet capacity and performance requirements, then storage capacity increases, but system complexity and disk farm size increase

Engineering Contradiction:

Improvestorage capacityVSAvoidsystem complexity

Core Design Contradiction:

Quantity of substanceVSDevice complexity

Solution Approach 1:

The system is segmented into independent storage nodes that can be individually managed and scaled. Each node contains local disk managers and array groups that are decoupled from specific hardware nodes, allowing the storage system to be divided into manageable units that can be added incrementally without increasing overall system complexity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Storage nodes are designed with universal functionality to perform multiple roles within the cluster. The decoupling of array groups from specific nodes allows any node to assume different storage roles, making the system more versatile and reducing the need for specialized components that would increase complexity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Ease of operation

If array groups are tied to specific nodes, then data placement is simplified, but scalability and flexibility are reduced

Engineering Contradiction:

Improvedata placement simplicityVSAvoidscalability

Core Design Contradiction:

Ease of operationVSAdaptability or versatility

Solution Approach 1:

The system implements dynamic data placement where array groups are not statically bound to specific nodes. Instead, the mapping between array groups and nodes is flexible and can change based on cluster conditions, enabling both simple operation through automated management and high scalability through adaptive reconfiguration.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

An intermediary layer is introduced between array groups and physical nodes to manage the mapping relationship. This intermediary abstraction allows data placement to remain simple from the user perspective while enabling complex scalability and flexibility in the underlying infrastructure through the decoupling mechanism.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Device complexity

If single node appliances are used, then system simplicity is maintained, but performance and capacity are limited

Engineering Contradiction:

Improvesystem simplicityVSAvoidperformance

Core Design Contradiction:

Device complexityVSProductivity

Solution Approach 1:

Multiple storage nodes are merged into a unified cluster that presents a single logical storage system. This combining approach maintains simplicity from the user perspective while achieving enhanced performance and capacity through the aggregated resources of multiple nodes working together in the cluster.

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS11704039B2Method for reassembling local disk managers and array groups

Publication Date: 2023.07.18 EMC IP HLDG CO LLC

AI summary

A method of reassembling a local disk manager (LDM) and array group (AGRP) includes starting a physical extent manager (PEM) configured to run on a number of nodes. The PEM on each node is configured to manage an AGRP running on the same node. A number of LDMs are reassembled, and each LDM is configured to manage virtual disks on each of the nodes. Once enough LDMs are reassembled, an AGRP can be reassembled.