Time Compression System Testing Workload Simulation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current system testing methods fail to accurately replicate the varying workloads and transaction rates over extended time periods, leading to mismatches between testing and real-world production system workloads, especially with the rise of big data and cloud implementations.

Innovation Solution

The implementation of a method using an independent test clock that allows for dynamic time compression, enabling the execution of workloads in a specified sequence and date, and dynamically updating the test clock to simulate real-world scenarios, thereby replicating production system workloads over extended periods in a compressed time frame.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If testing is conducted over extended time periods to accurately replicate production workload variations, then workload modeling accuracy is improved, but testing time and productivity deteriorate

Engineering Contradiction:
Improveworkload modeling accuracyVSAvoidtesting time
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent combines multiple test executions spanning different time periods into a single consolidated test by merging their workloads. The workload merger component aggregates workloads from multiple test executions, allowing the system to capture workload variations over extended periods without actually executing tests for that entire duration, thus resolving the contradiction between accuracy and testing time.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent performs preliminary actions by collecting and storing workload data from multiple test executions before merging them. The workload collector gathers workload information during individual test executions, and this pre-collected data is then used to create an accurate consolidated workload model without requiring extended testing time.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If multiple test executions are merged to capture workload variations, then workload modeling accuracy is improved, but test complexity increases

Engineering Contradiction:
Improveworkload modeling accuracyVSAvoidtest complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces intermediary components to manage the complexity of merging multiple tests. The workload collector serves as an intermediary that gathers workload data from multiple test executions, and the workload merger acts as another intermediary that combines these workloads. These intermediaries abstract the complexity, allowing accurate workload modeling without directly managing the complexity of multiple simultaneous tests.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Reliability

If testing duration is extended to cover varying transaction rates and types, then system reliability assessment is improved, but loss of time increases

Engineering Contradiction:
Improvesystem reliability assessmentVSAvoidtesting duration
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent creates a copy of the extended-time workload characteristics by merging workloads from multiple shorter test executions. Instead of actually running tests over extended periods to assess reliability, the system copies the essential workload variations (transaction rates, types, and patterns) from multiple tests and combines them, achieving reliable assessment without the time loss of actual extended testing.

Inventive Principle:
Principle #26Copying

Data Source

PatentUS10664389B2System testing using time compression
Publication Date: 2020.05.26 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US10664389B2 patent drawing
  • US10664389B2 patent drawing
  • US10664389B2 patent drawing

AI summary

Provided are techniques for system testing using time compression.A first program and a second program of a workload are executed in accordance with a test clock, wherein the test clock is independent of a computer system clock, and wherein the first program and the second program are to be run in a specified sequence and each at a specified date and time. In response to the first program completing, the test clock is dynamically updated to the specified date and time of the second program to start execution of the second program.