Storage Volume Snapshot Optimization via Indirection Layer Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing storage systems face inefficiencies in operations like garbage collection and snapshot creation, as they often scan the entire address space even when only a small portion has changed, leading to suboptimal performance.

Innovation Solution

The introduction of an indirection layer that maps volumes to anchor mediums, allowing for read-write operations to be targeted specifically to changing portions, with the allocation of new anchor mediums only when necessary, and the use of read optimization operations to reduce the complexity of mappings.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If background operations scan the entire address space to maintain data integrity, then reliability is improved, but productivity deteriorates due to unnecessary operations on unchanged data

Engineering Contradiction:
Improvedata integrityVSAvoidoperation efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent divides the volume address space into multiple segments or regions, each tracked separately. Instead of scanning the entire address space, background operations are performed on individual segments that have changed. The storage controller maintains metadata that identifies which segments contain modified data, allowing selective processing of only those segments rather than the whole volume.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements partial action by performing background operations only on the necessary portion of the address space. When data is modified, only the affected segments are included in subsequent garbage collection and read optimization operations. This eliminates the excessive action of scanning and processing unchanged portions of the address space, improving operational efficiency while maintaining data integrity.

Inventive Principle:
Principle #16Partial or excessive action

2Reliability

If snapshots are created by copying entire volumes, then reliability is improved, but loss of time increases due to unnecessary copying of unchanged data

Engineering Contradiction:
Improvesnapshot integrityVSAvoidsnapshot creation time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent extracts only the changed segments from the volume for snapshot creation. Instead of copying the entire volume, the storage controller identifies which segments have been modified since the last snapshot and copies only those segments. This extraction approach maintains snapshot integrity for changed data while eliminating the time waste of copying unchanged portions.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies partial action by creating snapshots that include only the necessary changed segments rather than the entire volume. The storage controller uses metadata to determine which segments contain modifications and performs copy operations only on those segments, significantly reducing snapshot creation time while ensuring that all changed data is captured in the snapshot.

Inventive Principle:
Principle #16Partial or excessive action

3Measurement precision

If the storage system maintains detailed mappings to track all data changes, then measurement precision is improved, but device complexity increases

Engineering Contradiction:
Improvechange detection accuracyVSAvoidmapping structure complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the address space into manageable regions and maintains separate metadata for each segment. This segmentation allows the system to track changes with high precision at the segment level while avoiding the complexity of maintaining a single large mapping structure for the entire volume. Each segment's metadata is independent and can be processed separately.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a new dimension of organization by structuring mappings hierarchically, with top-level volume metadata and lower-level segment metadata. This dimensional change allows efficient tracking of changes through multiple levels of abstraction, improving measurement precision while managing complexity through hierarchical organization rather than flat structures.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS10482061B1Removing invalid data from a dataset in advance of copying the dataset
Publication Date: 2019.11.19 PURE STORAGE INC
  • US10482061B1 patent drawing
  • US10482061B1 patent drawing
  • US10482061B1 patent drawing

AI summary

Systems, methods, and computer readable storage mediums for optimizing performance in a storage system. A volume may be mapped to an anchor medium through an indirection layer. When a snapshot is taken of the volume, the allocation of a new anchor medium is delayed until a write operation targeting the volume is received. When a write operation targeting the volume is received, a new anchor medium may be allocated for only a portion of the volume corresponding to the portion of the volume which is targeted by the write operation. Also, when a snapshot is taken of the volume, read optimization operations are performed only the portions of the volume which have changed since the previous snapshot.