We continuously work with customers to help them implement strategies for developing higher quality software that are measurable and repeatable.
The motivators for improved quality are generally to reduce bugs in the field and lower overall support costs.
A key element of the VectorCAST approach to quality is measuring testing completeness with code coverage analysis.
I spent some time this week with a global telecom customer. We are working with their DevOps team to design customized workflows that are based on VectorCAST and will enable them to reduce time to market and increase release quality. The following is the phased approach we are implementing:
Integrate test coverage into all flavors of existing testing
Aggregate coverage from all sources to get a sense of testing completeness
Examine poorly covered modules and assess risk
Extend existing test suites to improve coverage
- Invest in unit test for high-risk modules or modules that are difficult to test during functional testing
Obviously this new workflow will take some time to implement, but the customer is confident that the end result will be higher quality applications that perform well in the field and have lower lifecycle support costs.