Skip to main content Skip to main navigation Skip to search Skip to footer

HCL Technologies

Non Functional Risk Assessment Framework to Increase Predictability of Non Functional Defects

Non Functional Risk Assessment Framework to Increase Predictability of Non Functional Defects

The standard approach to non-functional testing, which in most cases is limited to performance testing only, is to identify a subset of use cases, which are anticipated to be executed frequently in production, and conduct a set of non-functional tests. Such tests could include Load Tests, Stress Tests, Endurance Tests et al using any renowned commercial off the shelf (COTS) tool and then publish the test results. Where the test results indicate that certain transactions do not meet the stipulated Service Level Agreements (SLAs) (if they were defined in the first place) then either an exception is sought to go-live as per schedule or execute a re-test post tuning. Such tuning attempts are predominantly limited to modifying certain configuration values either at the application server or database server layer or both to alleviate the symptoms of performance issues. This superficial approach to non-functional testing tends to leave gaping holes in the system and exposes it to serious performance and non-functional quality issues when the “rubber hits the tarmac” in production. Given that there could be various reasons why the non-functional issues were not detected in a pre-production environment, one of the most critical elements is the ability to simulate the anticipated risks in a production like pre-production environment. To accomplish the same, the system under test should be systematically studied and all the potential risks thoroughly understood in the context of historical production non-functional issues and futuristic workload and deployment changes. This white paper explains how to undertake a Non-Functional Risk Assessment exercise to ensure the non-functional test strategy and scenarios have a fool-proof coverage of all the technical and non-functional risks to maximize predictability of non-functional quality attributes of the system.


Contact Us