MEASURING SOFTWARE QUALITY - A CASE-STUDY

Over the years, the National Security Agency has become extremely depe ndent on the software that makes up its information technology infrast ructure. NSA has come to view software as a critical resource upon whi ch much of the world's security, prosperity, and economic competitiven ess increasingly rest. If anything, dependence on software and its cor responding effect on national security makes it imperative for NSA to accept and maintain only the highest quality software. Cost overruns o r software systems that are defective or of low quality can impose a s ignificant burden on national security and NSA's mission. NSA is no mo re immune than the rest of the software industy to the problems of low -quality software. Software development is intensely manual in nature, and it inevitably falls victim to rushed schedules, constantly changi ng requirements, poor process, and failure to adhere to software engin eering practices. So what is NSA doing about software quality? The NSA 's Software Engineering Applied Technology Center has done metrics ana lysis on some 25 million lines of code. The result is a highly correla ted set of measures that we have developed into a streamlined set of c ode-level release criteria that we apply to code written at NSA organi zations. The author describes these critical measures and has drawn up a case study to illustrate the benefits of applied quality assurance and code-level measurement activities.