Target VTS Load Calculation for Source Selection in Storage Systems

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In a storage system with multiple virtual tape servers (VTSs), the existing technologies fail to efficiently select the optimum source VTS for data copy requests, leading to increased system load and potential performance degradation due to unconsidered job processing states among source VTSs.

Innovation Solution

A method where the target VTS calculates load values based on attribute information, including job processing ability, unprocessed data amounts, and priorities, to select the source VTS that minimizes the overall system load by distributing job loads effectively across multiple source VTSs, thereby optimizing copy processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the target VTS requests data copy from multiple source VTSs simultaneously, then data redundancy is improved, but system resource waste increases due to simultaneous writing to both VTSs

Engineering Contradiction:
Improvedata redundancyVSAvoidsystem resource waste
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The system performs preliminary assessment of source VTS states before initiating copy operations. The target VTS evaluates the current load and processing status of potential source VTSs, then selects the most appropriate source in advance, avoiding simultaneous copy operations that would waste system resources while still achieving data redundancy

Inventive Principle:
Principle #10Preliminary action

2Ease of operation

If the target VTS selects source VTS without considering job processing states, then copy operation simplicity is maintained, but system load increases and performance degrades

Engineering Contradiction:
Improvecopy operation simplicityVSAvoidsystem performance
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The target VTS continuously monitors and receives feedback from source VTSs regarding their current job processing states, load conditions, and availability. This feedback mechanism enables the target VTS to dynamically select the most appropriate source VTS for each copy request, optimizing system performance while maintaining operational simplicity through automated decision-making

Inventive Principle:
Principle #23Feedback

3Speed

If the target VTS requests copy from a source VTS with high current load, then copy request processing speed is maintained, but overall system load increases and job completion time extends

Engineering Contradiction:
Improvecopy request processing speedVSAvoidjob completion time
Core Design Contradiction:
SpeedVSLoss of time

Solution Approach 1:

Before initiating copy operations, the target VTS performs preliminary evaluation of source VTS load conditions and processing states. By assessing the current system state in advance and selecting source VTSs with lower loads, the system avoids adding excessive load to already busy VTSs, thereby reducing overall system load and preventing extension of job completion times while maintaining efficient copy request processing

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2466446B1Storage system, method, and program, comprising a plurality of storage devices
Publication Date: 2014.02.26 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • EP2466446B1 patent drawingFigure 1
  • EP2466446B1 patent drawingFigure 2
  • EP2466446B1 patent drawingFigure 3(a)~3(b)

AI summary

In the entire system including plural source storage devices, a new selection method is provided, which a target storage device selects which source storage device to accept a copy request from the target storage device so as to minimize the load on the entire system. [Solution]The target storage device must select a source storage device as a copy request destination of a specific data in the entire system. A first and a second source storage device are in processing plural data. Firstly The method calculates first and second load values as job loads being processed respectively. A first total system load value for the entire system is given by adding a job load value of the specific data to the first load value for the first source storage device as a first final load value and then adding the second load value to the first final load value. Furthermore a second total system value for the entire system is given by adding the job load value of the specific data to the second load value for the second source storage device as a second final load value and then adding the first load value to the second final load value. Lastly the first total system value is compared with the second total system value to select which can minimize the load on the entire system, the first source or second source storage device.