Clustered Filesystem Node Scaling with Preauthorized Singleton Clones

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional systems face inefficiencies in adding and removing nodes from clustered filesystems, leading to idle compute resources and poor performance due to membership state checks, consensus protocol participation, and static node formation, which are exacerbated by the increasing data complexity and scale from IoT devices and machine learning models.

Innovation Solution

Implementing a method that emulates preauthorized singleton clusters to efficiently add and remove nodes by cloning data structures and fencing off failed nodes, utilizing template images and token management to improve filesystem performance and reduce latency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional node addition and removal procedures are used in clustered filesystems, then system reliability is maintained through membership state checks and consensus protocols, but node operation time increases leading to idle compute resources and poor performance

Engineering Contradiction:

Improvefilesystem consistencyVSAvoidnode operation time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-authorizing singleton clusters and preparing template images before nodes need to be added or removed. This allows the metadata server to have configuration details ready in advance, eliminating the need for time-consuming membership state checks and consensus protocols during actual node operations.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses copying by creating template images from authorized singleton clusters that contain pre-configured metadata. When adding nodes, the system copies these template images and applies the stored configuration details directly, bypassing the need for real-time consensus protocols and membership state verification.

Inventive Principle:

Principle #26Copying

2Stability of the object's composition

If static node formation is used in clustered filesystems, then system stability is maintained, but system adaptability decreases leading to idle compute resources when nodes cannot be dynamically added or removed

Engineering Contradiction:

Improvecluster stabilityVSAvoidnode scalability

Core Design Contradiction:

Stability of the object's compositionVSAdaptability or versatility

Solution Approach 1:

The system introduces dynamics by enabling flexible addition and removal of nodes through the use of pre-configured template images and stored cluster details. This allows the clustered filesystem to dynamically adapt its composition without compromising stability, as the metadata server maintains authorized singleton cluster configurations that can be rapidly deployed.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system achieves universality by creating a universal template image that can be copied and applied to multiple nodes. The stored cluster details from authorized singleton clusters serve as a universal configuration that can be reused across different nodes, enabling both horizontal scaling and flexible node management while maintaining cluster stability.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Productivity

If cloud bursting is implemented to handle increased data capacity, then processing capacity increases, but network bandwidth consumption increases and latency issues worsen

Engineering Contradiction:

Improvedata processing capacityVSAvoidnetwork bandwidth consumption

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The system applies local quality by enabling edge nodes to process data locally using the clustered filesystem, reducing the need to send all data to centralized cloud locations. The pre-configured nodes can be distributed at edge locations, allowing data processing to occur closer to data sources and reducing network bandwidth consumption and latency.

Inventive Principle:

Principle #3Local quality

4Productivity

If conventional workload managers are used that schedule workloads when cloud resources become available, then resource utilization improves, but application scaling latency increases

Engineering Contradiction:

Improveresource utilizationVSAvoidscaling latency

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The system performs preliminary action by pre-authorizing singleton clusters and storing their configuration details in the metadata server before scaling is needed. When nodes need to be added or removed, the system can rapidly deploy using these pre-prepared configurations, eliminating the latency associated with waiting for cloud resources and scheduling workloads.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12561283B2Adding and/or removing nodes from clustered filesystems

Publication Date: 2026.02.24 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

A computer-implemented method, according to one approach, includes: receiving a request to create a new node in a clustered filesystem. In response to receiving the request, a copy of details is retrieved, the copy of details being associated with an established singleton cluster preconfigured to access the clustered filesystem. The retrieved copy of details are further applied to a new node in the clustered filesystem. Moreover, data structures of the new node are modified according to the received request. Other systems, methods, and computer program products are described in additional approaches herein.