Performance Testing :- To Do or Not to Do
Categories:
Performance Testing
Categories: Performance Testing
| image courtesy :-imdude.com Scenario :- A Company to develop or acquire a new Web based Software which is intended at speeding up it's business process, justifying it's Return On Investment , enabling it to have a leaner team, aimed at cutting costs or migrating from a paper -based to an electronic system. It invests time, money and human resources into unit testing, undergoes comprehensive System and Integration Testing Cycles, performs User Acceptance Testing and the system is deployed into Production . The system has been developed or installed according to Vendor or Architect specifications , there are robust Business contingency and Disaster Recovery plans in place and the Business can start to see money. Everything is hunky-dory until the usage of the system peaks and the system comes down on its knees and all hell breaks loose trying to figure out if the problem is with the application, the underlying infrastructure or the back-end database. What went wrong? In spite of everything that was seemingly right in the above scenario, there was one thing missing which was proposed as part of the request for Project funding, but could't be implemented due to Time or Budget constraints. And that not-so-insignificant thing was Performance Testing aka Stress or Load Testing. All of us who have either been consumers of a new Software System or closely involved in it's development, Project Management or Testing would understand the enormous importance of Performance Testing and building it into our Business proposals and Project Plans, particularly if the Software is to be rolled out to a wider audience. So where would you highlight the importance of Performance Testing to a Project Sponsor?
And how would you justify their ROI in the Peformance Testing Component of the Project? by asking these thought provoking questions:-
The discerning Project Sponsor is then compelled to think about the risk of not doing Performance testing and the significant costs incurred to analyse the subsequent production issues . So I have two open questions to the community :-
|




