RALS 2020 : 1st Workshop on Robust and Adaptive Large Scale Software Systems

Motivation and Objectives

Large-Scale Software systems are the backbones of today’s society. The sheer scale and the resulting dynamics renders manual control of those systems infeasible. Instead, they need to automatically adapt to changes, both from within the system itself as well as its environment. In particular, those systems face a high variability regarding users/things, workloads, or resources. To fulfill given functional and quality requirements even during adaptation, these systems require a certain robustness in respect to change. This workshop aims at exploring challenges and techniques to tackle this interplay of adaption and robustness across the complete system stack.

As a key challenge, it is not possible to completely switch off these systems when performing changes. Hence, all adaptations need to be performed online while the system continuously provides its service in the agreed quality or even more severe, in the face of safety-critical operation scenarios. While robustness constitutes an aspect typically addressed in safety-critical systems with a focus on reliability, availability, or security threats, the outlined constant changes in large-scale software systems create a smorgasbord of new robustness challenges. Further complicating is that these systems are not built by a single vendor, instead, they form complex ecosystems where functionality is provided by different components from different ecosystem partners in complex and unplanned, ad-hoc structures. The service and quality provided by these systems is consequently an emergent property of the ecosystem that can neither be achieved nor easily enforced centrally.

Reaching the goal of robust and adaptive large-scale software systems is a holistic endeavor that cannot be addressed by research solely in a single area of computer science. Instead, it requires coordination and cross-area solutions of multiple areas of Computer Science. Consequently, this workshop focuses on bringing the research communities of all required subareas together to exchange experiences, ideas, and future solutions to achieve robust and adaptive large-scale software systems.