ca service virtualization vs mocks stubs

CA Service Virtualization vs. Mocks/Stubs

Compare key areas of differentiation between mocking / stubbing tools and CA Service Virtualization

DefinitionsSetting the ground

3 © 2017 CA. ALL RIGHTS RESERVED.

Stubs Mocks

“Provide canned answers to calls made during the test, usually not responding at all to anything outside what’s programmed in for the test.”

Development or Testing “objects pre-programed with expectations which form a specification of calls they are expected to receive.”

http://martinfowler.com/articles/mocksArentStubs.html#TheDifferenceBetweenMocksAndStubs

http://martinfowler.com/articles/mocksArentStubs.html


Service Virtualization

A method of capturing and simulating the behavior of unavailable or incomplete systems.

Takes a modern approach to the practice of “mocking” or “stubbing”.

Uses sufficiently “real” virtual services without depending on complexity and constraints of real hardware or software.

Benefits include: – Earlier detection of defects– Shorter Release cycles– Reduced cost of development– Reduced physical infrastructure

dependencies

Capabilities & Benefits - Compare & Contrast

Service Virtualization vs. Mocks/Stubs


Reduce Dependencies and Constraints in App Development

Reduce cycle times, earlier defect discovery, more efficient use of resources.

Service Virtualization Mocks/Stubs


Virtualize test data for comprehensive test paths

Faster setup/teardown, greater stability for test automation.


Test data must be designed for specific use


Enable high performanceenvironments

Better realism and quantity of performance testing.


Not designed for performance testing


Complete testing ofapplication business logic

Test each app component individually and automate the validation of business processes that span across multiple internal and cloud based services.


Inside Application Code


Provide reuse and collaboration across development teams

Virtualize a Service once and reuse across development teams and applications.


Every app requires writingmocks that could be shared


Incorporate into Continuous Integration Process

Hook into all stages of CI environments. (Dev., QA, Performance, and Integration)


Limited outside of Dev. Environment


Reduce time and cost ofMaintenance for Dev andTest environments

Increase dev time to focus on applicationcode vs. maintaining mocks and test data that are code dependent.


Code Dependent


Expansive Code Coverage

Enable a consistent and standardized SDLC/ CI Process with solution that supports all code bases.


Java Only


UI to modify request/ response

Resources with no developer skills can create and update models.


Coding required


Bulk import/ recording oftransactions

Leverage large volumes of actual behavior and data.


No recording or importingof transactions


Magic Strings/Dates

Test request that don’t have a “specific” response, as patters are identified and made available. .


Static Information


Think Time

Add “think time” to transactions to simulate performance in production.


Single Thread


Various Capture Methods

capture services through record a live system (for accuracy) or import data.


No Capture Mode


Learning Mode

“Learn” live system data over time or pass-through to w/o reconfiguring SUT.


No Learning Mode


CA Service Virtualization

Leverage throughout Continuous Integration (CI) process – Development, Pre-Production, Performance, Integration, QA.

Test Data Management challenges may disappear entirely.– Edge conditions, negative test scenarios and error handling– Virtual service never “burn” since they play back behavior responses.

Support across the enterprise including:– web traffic (HTTP), web services (SOAP/XML), integration layers & ESB

(JMS, etc.)

Simulate transactions and connections with: – underlying mainframes (CICS, CORBA, etc.,), databases (JDBC, etc.) and

third-party services.


CA Service Virtualization

STS security Plugin module cannot be mocked, but are easily supported as a container with Service Virtualization

Drive performance testing. Traditional Mock/Stub tools - not designed for performance testing environments.

Represents more realistic behavior than stubs and mocks. Delivers stand-alone, decoupled services suitable for sharing

between teams and organizations. Does not require proprietary source-code level changes or

modifications of the application under development.

For more information visit

ca.com/SV

http://www.ca.com/SV

ca service virtualization vs mocks stubs

Software