File Recipe Generation for Fast Copy Overwrite Workloads

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Fast copy overwrite (FCOW) workloads in deduplication storage systems do not create file recipes, missing out on benefits like differencing, tiering, and replication efficiencies due to the lack of synthetic API usage.

Innovation Solution

A method to create file recipes for FCOW workloads by detecting the workload and tracking write operations to store recipe information, enabling the use of extended attributes and Merkle tree structures for efficient data replication and storage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If FCOW workload is used without synthetic API, then I/O amplification is reduced through fixed sized segmentation, but recipe information is not created and replication efficiencies are lost

Engineering Contradiction:

Improvereplication efficiencyVSAvoidworkload complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary mechanism (write tracking system) that monitors FCOW operations and automatically generates recipe information without requiring changes to the FCOW workload itself. This mediator captures write operations and creates recipe metadata, enabling replication efficiencies while maintaining the simplicity of fixed-sized segmentation.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system enables self-service by having the FCOW workload automatically generate recipe information through write tracking. The workload serves itself by capturing its own write operations and creating the necessary metadata for optimization, without requiring external intervention or synthetic API calls.

Inventive Principle:

Principle #25Self-service

2Reliability

If synthetic API is used to create recipes, then replication efficiencies and differencing benefits are achieved, but processing overhead is increased

Engineering Contradiction:

Improvedata consistencyVSAvoidprocessing overhead

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent uses lightweight, disposable write tracking records that are created temporarily during FCOW operations and then converted into persistent recipe information. These temporary tracking objects require minimal processing overhead compared to full synthetic API calls, yet provide sufficient information for data consistency and replication.

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

3Productivity

If fixed sized segmentation is used for FCOW, then I/O amplification is reduced, but recipe creation capability is lost

Engineering Contradiction:

ImproveI/O efficiencyVSAvoidrecipe information

Core Design Contradiction:

ProductivityVSLoss of information

Solution Approach 1:

The patent performs preliminary write tracking during the FCOW operation to capture all necessary information before the workload completes. By proactively monitoring and recording write operations in advance, the system preserves recipe information without interfering with the I/O efficiency of fixed-sized segmentation.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12105594B2Creating file recipes for copy overwrite workloads in deduplication file systems

Publication Date: 2024.10.01 DELL PROD LP

AI summary

Embodiments for processing fastcopy overwrite backup workloads (FCOW) as virtual synthetic backups, by detecting whether or not a backup workload comprises an FCOW workload in which a base file is fastcopied to a target file, wherein portions of the base file are overwritten through overwrites by new data in the target file by a deduplication backup process. For FCOW workloads, the process creates a file recipe by trapping seek offsets and write byte lengths for the overwrites, and writing extent information comprising offset differences to a recipe table of the file recipe, and the file recipe facilitates file processing optimizations including differencing, replication, and tiering.