Tool Support for Software Artefact Traceability in DevOps Practice

Abstract

Software development in DevOps practice is a widely used approach to cope with the demand for frequent artefact changes. These changes require a well-defined method to manage artefact consistency to ease the continuous integration process. This chapter proposes a traceability management approach for the artefact types in the main phases of the software process including requirements, design, source code, testing, and configuration. This chapter addresses traceability management, including trace link creation, change detection, impact analysis, change propagation, validation, and visualisation. This chapter presents a tool named SAT-Analyser that is applicable for any software development method and designed for continuous integration, multi-user collaboration, and DevOps tool stack compatibility. The SAT-Analyser is assessed using case studies and shown an impact analysis accuracy of 0.93 of F-measure. Further, the feedback by DevOps practitioners has shown the suitability and innovativeness of the proposed approach.

Machine Learning Operations: A Survey on MLOps Tool Support

“…Streamlit is a python library that facilitates web app creation in less time. It has an easy to use interface without backend configurations 21 . Streamlit supports code iteratively and views results while ongoing development.…”

“…Research has been done to explore new horizons on developing sophisticated systems for MLOps. Although several tool support is available to manage the artefact traceability DevOps practice [21], there are no major tools that address the traceability in the MLOps life cycle. Several studies have presented automation tools to maintain the artefact consistency during the DevOps-based software development [1,5].…”

Applications of Natural Language Processing in Software Traceability: A Systematic Mapping Study

“…NLP was first introduced in SAT-Analyzer for addressing artifact inconsistencies due to natural language representation (Arunthavanathan et al, 2016) -it improves the usability of SAT-Analyzer through automated generation of XML input from requirement artifacts, which was then evaluated by a case study on a Pointof-sale (POS) system (Rubasinghe et al, 2018b). SAT-Analyzer was also covered in DevOps practices (Rubasinghe et al, 2018a(Rubasinghe et al, , 2020; a traceability management tool for continuous integration and multi-user collaboration. TiQi, on the other hand, focuses on trace queries that are generally complex and naturally worded, transforming them into executable SQL statements (Pruski et al, 2014).…”

“…In a dynamic continuously integrated, continuously developing environment (Rubasinghe et al, 2018a(Rubasinghe et al, , 2020(Rubasinghe et al, , 2018b, artifacts transform constantly and this hampers continuous traceability efforts. In cases where traceability is necessary for regulations (Florez, 2019;Arora et al, 2015), the natural language used in these documents is not represented similarly to other artifacts, such as functional and non-functional requirements.…”

“…Large-scale systems pose a challenge in traceability management due to the complexity of trace links, particularly in visualisation (Rubasinghe et al, 2020;Chen et al, 2018). This also relates to time and compute resource complexities, and becomes even more challenging in environments where constant change is present (Rubasinghe et al, 2018a,b).…”