Tape Storage Data Duplication via Segment Metadata

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The inefficiency in recording data to tape storage due to fragmentation issues in disk drives, which slows down the data recording process, necessitates improved systems and methods for enhancing data recording efficiency.

Innovation Solution

A computer-implemented method that identifies metadata to distinguish between unchanged and changed segments of a data unit, reads the previous version from tape storage, and combines these segments with changed segments from data sharing storage to efficiently duplicate the data unit on tape, leveraging the faster reading speed of tape storage over fragmented disk drives.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If data is read from fragmented disk drives to duplicate to tape storage, then the data can be updated, but the reading speed is slow due to fragmentation

Engineering Contradiction:
Improvereading speedVSAvoiddata fragmentation
Core Design Contradiction:
SpeedVSStability of the object's composition

Solution Approach 1:

The patent segments the data unit into multiple portions, identifying which segments have changed and which remain unchanged. This allows selective reading from different storage sources - unchanged segments from tape (fast sequential read) and changed segments from disk (random access capability), thereby resolving the speed-fragmentation contradiction.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent creates a copy of the previous version of the data unit from tape storage and combines it with changed segments to form an updated version. This copying approach allows the system to leverage the fast sequential read capability of tape for unchanged data while only retrieving changed portions from disk, eliminating the need to read entire fragmented datasets from disk.

Inventive Principle:
Principle #26Copying

2Speed

If the previous version of data unit is read from tape storage, then reading speed is faster, but additional reading operations are required from both tape and data sharing storage

Engineering Contradiction:
Improvereading speedVSAvoidreading operations
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

Instead of reading the entire data unit from a single source, the patent applies partial action by reading only the necessary portions from each storage type - unchanged segments from tape and changed segments from data sharing storage. This selective partial reading approach maintains high speed while managing operational complexity through intelligent segmentation and metadata utilization.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS10296221B1Systems and methods for improving the efficiency of recording data to tape
Publication Date: 2019.05.21 COHESITY INC
  • US10296221B1 patent drawing
  • US10296221B1 patent drawing
  • US10296221B1 patent drawing

AI summary

A computer-implemented method for improving the efficiency of recording data to tape may include (i) identifying a command to duplicate a data unit to tape storage after a previous version of the data unit has already been duplicated to tape storage, (ii) identifying metadata that distinguishes between segments of the data unit that have not changed since the previous version of the data unit and segments that have changed, (iii) reading the previous version of the data unit from tape storage and reading the segments of the data unit that have changed from a data sharing storage rather than tape storage, and (iv) combining, using the metadata, the segments read from tape storage that have not changed and the segments read from the data sharing storage that have changed to duplicate the data unit to tape storage. Various other methods, systems, and computer-readable media are also disclosed.