File Block Variant Tracking for Multi-Disk Array Parallelism

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing file systems face performance degradation due to asynchronous data input and output operations across processor cores, leading to congestion when writing to centralized file servers, and there is a need for quicker provisioning and release of resources in multi-disk arrays to leverage parallelism effectively.

Innovation Solution

A computer-implemented method and apparatus that utilizes a file block tracker and librarian system to manage file block variants across multiple processor cores and storage resources, allowing for efficient aggregation of file writes, redirection of read requests, and release of resources upon acknowledgement, thereby improving throughput and reducing congestion.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If files are written to a centralized file server, then data integrity is maintained, but congestion occurs and performance degrades

Engineering Contradiction:

Improvedata integrityVSAvoidwrite throughput

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent segments the centralized file server architecture into a distributed system with multiple file servers. Each file server handles a portion of the write operations independently, allowing parallel processing of file writes across multiple nodes. This segmentation eliminates the single-point congestion bottleneck while maintaining data integrity through coordinated distributed file block variant tracking.

Inventive Principle:

Principle #1Segmentation

2Reliability

If resource holding time is extended to ensure data integrity, then reliability improves, but throughput decreases

Engineering Contradiction:

Improvedata integrityVSAvoidresource holding time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent implements preliminary action by pre-tracking file block variants and maintaining version information before write operations complete. The system proactively manages file block metadata and variant information in advance, allowing resources to be released sooner while data integrity is preserved through pre-established tracking mechanisms.

Inventive Principle:

Principle #10Preliminary action

3Productivity

If parallelism is increased across multiple processor cores, then throughput improves, but congestion occurs at centralized storage

Engineering Contradiction:

ImprovethroughputVSAvoidcongestion management

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary file block tracker that sits between multiple processor cores and the storage system. This intermediary component coordinates file block variant tracking and manages the complexity of parallel operations, allowing multiple cores to write concurrently without creating congestion at the centralized storage point.

Inventive Principle:

Principle #24Intermediary (Mediator)

4Use of energy by moving object

If file block operations are performed asynchronously across processor cores, then resource utilization improves, but performance degrades due to random access patterns

Engineering Contradiction:

Improveresource utilizationVSAvoidfile system performance

Core Design Contradiction:

Use of energy by moving objectVSProductivity

Solution Approach 1:

The patent implements dynamic adaptation by adjusting file block operation scheduling based on system state. The system dynamically coordinates asynchronous operations from multiple processor cores, adapting the timing and sequencing of file block accesses to optimize performance while maintaining high resource utilization. This dynamic coordination reduces random access patterns while preserving asynchronous processing benefits.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS12182429B2Hierarchical file block variant tracking for performance in parallelism at multi-disk arrays

Publication Date: 2024.12.31 IC MANAGE INC

AI summary

A computer implemented method implemented by at least one processor performing executable instructions improves the performance of multi-disk arrays by aggregating several file block variant commitments into a fewer number of parallel file write operations. Using the method, a system and apparatus intermediates between a central file service and distributed virtual machine data center servers. Requests for file block variants are redirected to the location of the freshest retention. Each data center has a non-transitory file block variant store locally attached. Each file block librarian provides a file system view to each virtual machine. Several Virtual Machine Data Centers are coordinated by a tracker which locates, redirects, and retrieves any file block variant within its name space or a name subspace delegated to a librarian. A threshold of parallelism related to dimensions of a multi-disk array determines scheduling of write operations.