Volume-Level Erasure Coding for Low-Overhead Sequential Storage

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing data storage systems face challenges in optimizing data performance and integrity on network computing and storage systems, particularly in managing large volumes of data with redundancy coding techniques that ensure availability and durability while minimizing storage overhead.

Innovation Solution

The use of redundancy codes, such as erasure codes, to store original data as shards across multiple volumes, accompanied by sparse indices that facilitate efficient retrieval and regeneration of data, even in the event of shard corruption or unavailability, by leveraging a generator matrix to generate redundancy coded shards.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If redundancy coding techniques are applied to ensure data availability and durability, then data reliability is improved, but storage overhead increases

Engineering Contradiction:
Improvedata availability and durabilityVSAvoidstorage overhead
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent segments data into multiple shards and distributes them across different storage volumes. By applying redundancy coding at the shard level rather than treating data as a single unit, the system achieves improved reliability while optimizing storage overhead. Each shard can be independently stored and retrieved, allowing efficient use of redundancy codes.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent dynamically adjusts redundancy coding parameters based on data characteristics and storage conditions. By changing the degree of redundancy applied to different data types or storage scenarios, the system optimizes the balance between data reliability and storage overhead, avoiding uniform application of redundancy codes to all data.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If data is stored as shards across multiple volumes, then data durability is improved, but retrieval complexity increases

Engineering Contradiction:
Improvedata durabilityVSAvoidretrieval process complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent pre-computes and stores metadata information about shard locations, relationships, and reconstruction requirements before data retrieval is needed. This preliminary organization of information allows the retrieval process to simply follow pre-determined paths and procedures, significantly reducing retrieval complexity despite the distributed shard architecture.

Inventive Principle:
Principle #10Preliminary action

3Speed

If sparse indices are used to facilitate efficient retrieval, then data access speed is improved, but storage overhead increases

Engineering Contradiction:
Improvedata retrieval speedVSAvoidindex storage overhead
Core Design Contradiction:
SpeedVSQuantity of substance

Solution Approach 1:

The patent applies sparse indexing selectively to specific data regions or access patterns rather than uniformly across all stored data. By identifying which data segments benefit most from rapid retrieval and applying indices only to those portions, the system achieves improved access speed for critical operations while minimizing the overall storage overhead consumed by index structures.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3764241B1Volume-level redundancy coding techniques for sequential transfer optimized storage devices
Publication Date: 2025.08.27 AMAZON TECH INC
  • EP3764241B1 patent drawingFigure 1
  • EP3764241B1 patent drawingFigure 2
  • EP3764241B1 patent drawingFigure 3

AI summary

Techniques described and suggested herein include systems and methods for storing, indexing, and retrieving original data of data archives on data storage systems using redundancy coding techniques. For example, redundancy codes, such as erasure codes, may be applied to archives (such as those received from a customer of a computing resource service provider) so as allow the storage of original data of the individual archives available on a minimum of volumes, such as those of a data storage system, while retaining availability, durability, and other guarantees imparted by the application of the redundancy code. Sparse indexing techniques may be implemented so as to reduce the footprint of indexes used to locate the original data, once stored.