Storage System Replication Lag Threshold Management

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing storage systems face challenges in efficiently managing data storage and retrieval across multiple storage nodes, particularly in ensuring high availability and performance in distributed environments.

Innovation Solution

The proposed solution involves a storage system architecture that utilizes a cluster of storage nodes with non-volatile solid state storage units, each equipped with a processor and flash memory. This architecture employs erasure coding and redundant metadata to distribute data across multiple storage nodes, ensuring data availability and performance even in the event of node failures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If data is distributed across multiple storage nodes using erasure coding, then data availability and reliability are improved, but system complexity increases

Engineering Contradiction:
Improvedata availabilityVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments data into multiple fragments and distributes them across different storage nodes using erasure coding. This allows the system to maintain high availability by reconstructing data from any sufficient subset of nodes, directly resolving the contradiction by improving reliability through segmentation while managing complexity through standardized segmentation protocols.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a storage manager as an intermediary component that handles the complexity of erasure coding, data distribution, and reconstruction operations. This mediator abstracts the complex underlying mechanisms from the user, improving reliability through sophisticated data management while hiding system complexity from end users.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If redundant metadata is used to track data locations, then data retrieval efficiency is improved, but storage overhead increases

Engineering Contradiction:
Improvedata retrieval efficiencyVSAvoidstorage overhead
Core Design Contradiction:
ProductivityVSQuantity of substance

Solution Approach 1:

The patent pre-computes and stores metadata including data location information, fragmentation patterns, and reconstruction requirements before actual data retrieval operations. This preliminary action enables fast data retrieval by having all necessary indexing information ready in advance, while the metadata is optimized to minimize storage overhead through efficient data structures.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent dynamically adjusts metadata storage parameters such as the level of redundancy and detail based on system conditions and access patterns. This allows the system to optimize the balance between retrieval efficiency and storage overhead by changing metadata parameters adaptively rather than using fixed overhead levels.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If proactive data rebuilding is implemented after node failure, then system reliability is maintained, but processing time increases

Engineering Contradiction:
Improvesystem reliabilityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent implements proactive data rebuilding that begins immediately upon detecting node failure, before the system becomes unavailable. By initiating reconstruction operations promptly and using pre-computed metadata to guide the process, the system maintains reliability by restoring data redundancy while minimizing the time the system operates in a degraded state.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent ensures continuous data protection by implementing overlapping reconstruction operations that can proceed in parallel across multiple nodes. This continuity approach maintains system reliability by ensuring data is constantly being reconstructed and protected, while parallel processing reduces the total time required compared to sequential rebuilding.

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentUS20250036537A1Application Management Based on Replication Performance of a Storage System
Publication Date: 2025.01.30 PURE STORAGE INC
  • US20250036537A1 patent drawing
  • US20250036537A1 patent drawing
  • US20250036537A1 patent drawing

AI summary

An illustrative method includes a multi-site storage system determining performance data associated with the multi-site storage system; determining, based on the performance data, a replication lag threshold value; and providing the replication lag threshold value to an application. In some embodiments, the application uses the replication lag threshold value to set a configuration parameter for the application, such as a replication timeout value of the application.