Snapline Metadata for Storage Snapshot Efficiency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing snapshot solutions in data storage systems are limited by high storage requirements, long creation times, and the need to pause writing operations, with space-efficient solutions still relying on significant metadata storage and being unsuitable for low-endurance storage drives.

Innovation Solution

The method involves creating snaplines with metadata including timestamps and status markers for offset ranges, using a global counter for timestamp uniformity across compute nodes, and maintaining a prefix tree for dependency management, allowing for efficient snapshot creation and storage without duplicating metadata, and enabling snapshot reconstruction based on previous states.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If full copies of data structures are used for snapshots, then snapshot reliability is improved, but storage capacity requirements increase significantly

Engineering Contradiction:
Improvesnapshot reliabilityVSAvoidstorage capacity
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent divides the data structure into multiple segments or blocks, and the snapshot only records the state information (pointers or metadata) of these segments at a specific time point, rather than copying the entire data structure. This segmentation allows the system to maintain snapshot reliability while significantly reducing storage capacity requirements.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts only the essential state information (such as pointers, offsets, or metadata) needed to reconstruct the data structure state at a given time, rather than storing complete copies. This extraction approach maintains the ability to recover full snapshot states while using minimal storage space.

Inventive Principle:
Principle #2Taking out (Extraction)

2Quantity of substance

If space-efficient snapshots using metadata pointers are used, then storage capacity is reduced, but metadata storage requirements increase significantly

Engineering Contradiction:
Improvestorage capacityVSAvoidmetadata storage
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent merges the metadata storage with the existing data structure storage system by integrating snapshot state information into the same storage infrastructure. This merging eliminates the need for separate metadata storage systems, reducing overall device complexity while maintaining space efficiency.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent designs the storage system to serve multiple functions: it stores both current data and snapshot state information using the same storage mechanisms. This multi-functionality reduces the need for dedicated metadata storage structures, simplifying the overall system architecture.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Loss of information

If snapshots are created frequently to record all previous states, then snapshot completeness is improved, but system performance deteriorates due to writing pauses

Engineering Contradiction:
Improvesnapshot completenessVSAvoidsystem performance
Core Design Contradiction:
Loss of informationVSProductivity

Solution Approach 1:

The patent prepares snapshot state information in advance by continuously maintaining state metadata as data changes occur. When a snapshot is requested, the pre-prepared state information is immediately available for retrieval, eliminating the need to pause writing operations to capture the current state.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent enables continuous data writing operations to proceed uninterrupted while snapshot state information is being collected and stored in the background. The snapshot creation process does not require pausing or stopping the data flow, maintaining system productivity while ensuring snapshot completeness.

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentUS11061569B2Method and system for providing improved efficiency snapshots using snaplines
Publication Date: 2021.07.13 VAST DATA LTD
  • US11061569B2 patent drawing
  • US11061569B2 patent drawing
  • US11061569B2 patent drawing

AI summary

A system and method for efficiently providing snapshots in storage systems. The method includes creating a plurality of snaplines for a storage system, each snapline having metadata including a timestamp and a status marker for each of a plurality of offset ranges, wherein each offset range corresponds to data stored in the storage system, the storage system including a plurality of compute nodes, wherein the timestamp of each snapline is a value of a global counter at a time of creation of the snapline, wherein the global counter is uniform among the plurality of compute nodes; and creating a snapshot based on the plurality of snaplines and a time value, wherein the snapshot indicates a snapline of the plurality of snaplines that is not to be deleted, wherein the timestamp of the snapline that is not to be deleted corresponds to the time value.