@misc{15984, keywords = {Anomaly detection, Self-Healing, artificial immune systems, runtime diagnosis, fault containment, dependability}, author = {Moeen Naqvi and Merve Astekin and Sehrish Malik and Leon Moonen}, title = {Adaptive Immunity for Software: Towards Autonomous Self-healing Systems}, abstract = {Testing and code reviews are known techniques to improve the qualityand robustness of software. Unfortunately, the complexity of modernsoftware systems makes it impossible to anticipate all possibleproblems that can occur at runtime, which limits what issues canbe found using testing and reviews. Thus, it is of interest toconsider autonomous self-healing software systems, which canautomatically detect, diagnose, and contain unanticipated problemsat runtime. Most research in this area has adopted a model-drivenapproach, where actual behavior is checked against a model specifyingthe intended behavior, and a controller takes action when the systembehaves outside of the specification. However, it is not easy todevelop these specifications, nor to keep them up-to-date as thesystem evolves. We pose that, with the recent advances in machinelearning, such models may be learned by observing the system.Moreover, we argue that artificial immune systems (AISs) areparticularly well-suited for building self-healing systems, becauseof their anomaly detection and diagnosis capabilities. We presentthe state-of-the-art in self-healing systems and in AISs, surveyingsome of the research directions that have been considered up tonow. To help advance the state-of-the-art, we develop a researchagenda for building self-healing software systems using AISs,identifying required foundations, and promising research directions.}, year = {2021}, journal = {28th IEEE International Conference on Software Analysis, Evolution and Reengineering (SANER)}, pages = {521-525}, publisher = {IEEE}, isbn = {978-1-7281-9630-5}, url = {https://ieeexplore.ieee.org/abstract/document/9425979}, doi = {10.1109/SANER50967.2021.00058}, }