Software Artifact Traceability in Big Data Systems
Erik M. Fredericks, Kate M. Bowers · Auerbach Publications eBooks · 2021
Traceability is the glue that holds together software artifacts at different levels of abstraction, from system design to software requirements. Traceability links ensure that code is maintainable and can minimize faults stemming from improper implementations, misunderstood requirements, and unexecuted test cases. Effectively, traceability ensures that all aspects of the software design life cycle are interconnected for both assurance and certification purposes. For big data applications, traceability remains a critical activity to ensure that the system can adequately support each big data dimension (i.e., “V”). For instance, a big data application may be subject to unexpected combinations of data types in a JSON packet that may result in an exception error, or alternatively, the rate at which data is received may be faster than the system can support, resulting in dropped packets. This chapter focuses on software traceability in the presence of big data, comprising topics in software traceability, link discovery, and link maintenance.