Progress and challenges in intelligent vehicle area networks
Communications of the ACM
Statistical Timing Analysis: From Basic Principles to State of the Art
IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
| Hi-index | 0.00 |
Hybrid multi-domain computing systems are emerging. While the context-aware self-adaptive system models are under intensive research in individual computing domains, their integration into a collective adaptive system still remains a major challenge. This position paper visions a meet-in-the-middle approach, where horizontal integration is applied to sub-system models extracted from vertical integration. The integration relies on orthogonal behavior and execution models respectively capturing the functional and non-functional features of sub-systems. The construction towards guaranteed services can be achieved with composition of static (worst-case) execution models, while best-effort services can be constructed with statistical models. Given that each computing domain has, to some extent, formulated its own design flow of context-aware systems, the envisaged meet-in-the-middle integration approach maximizes the reuse of existing models and platforms, thus is promising for the highly-complex system design process.