Distributed Control Path Partitioning for Active-Active Resource Utilization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current data storage systems face inefficiencies in utilizing system resources for control path processing, particularly in active-passive configurations where reserved resources on passive nodes remain idle, leading to suboptimal use of dual node appliances.

Innovation Solution

The proposed solution involves partitioning control path components across both nodes of each appliance, allowing both nodes to service control path commands in an active-active configuration, with one node running the management database server in primary mode and the other as a standby, enabling simultaneous processing and resource utilization.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If control path components are concentrated on a single primary node in active-passive configuration, then system simplicity and clear command routing are improved, but resource utilization deteriorates as passive nodes remain idle

Engineering Contradiction:

Improvecontrol path configurationVSAvoidresource utilization

Core Design Contradiction:

Device complexityVSProductivity

Solution Approach 1:

The control path components are segmented and distributed across multiple nodes instead of being concentrated on a single primary node. Each node runs specific portions of control path components, enabling both nodes to actively process control path commands simultaneously, thus improving resource utilization while maintaining manageable system complexity through modular architecture

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Both primary and secondary nodes are configured with control path components, enabling them to perform control path processing functions. The secondary node transitions from a purely passive standby role to an active participant that can service control path commands, making the system more versatile and improving overall resource utilization

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Productivity

If control path components are distributed across both nodes in active-active configuration, then resource utilization is improved, but system complexity and coordination overhead increase

Engineering Contradiction:

Improveresource utilizationVSAvoidcontrol path configuration

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

Both nodes continuously perform control path processing functions simultaneously in an active-active configuration. The primary node services control path commands during normal operation, and the secondary node is prepared to immediately take over if the primary node fails, ensuring continuous useful action without interruption and maximizing resource utilization

Inventive Principle:

Principle #20Continuity of useful action

Solution Approach 2:

The secondary node is pre-configured with control path components and maintains readiness to service control path commands before any failure occurs. This preliminary preparation ensures that the secondary node can immediately assume the primary role upon failure, reducing coordination overhead during actual failover events while maintaining active-active resource utilization

Inventive Principle:

Principle #10Preliminary action

3Reliability

If passive nodes are reserved for control path processing, then control path availability is improved, but data path processing capacity deteriorates as resources are reserved and idle

Engineering Contradiction:

Improvecontrol path availabilityVSAvoiddata path processing capacity

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

Control path components are segmented and distributed across both primary and secondary nodes, allowing each node to independently process control path commands. This segmentation enables both nodes to actively contribute to control path availability while simultaneously utilizing their full processing capacity, eliminating the reservation-idle cycle and improving data path processing capacity

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS11609711B2Distributed control path

Publication Date: 2023.03.21 EMC IP HLDG CO LLC

AI summary

Techniques for configuring and processing control path commands may include: partitioning control path components of a control path into a plurality of portions; performing first processing that configures a federation of a plurality of appliances, wherein each of the plurality of appliances includes a plurality of processing nodes, and wherein the first processing includes: for each of the plurality of appliances, configuring each of the plurality of processing nodes of said each appliance to run one portion of the plurality of portions of control path components; and selecting one of the plurality of appliances as a primary appliance of the federation; receiving a first management command at the primary appliance of the federation; and servicing the first management command by one or more of the plurality of appliances of the federation.