While most of the current cloud services operate on a best effort basis and provide no timing guarantees, there is a growing interest in cloud applications for industrial and critical applications. However, for cloud technologies to be fully embraced by industry, a higher degree of determinism is desired. This calls for new techniques and methodologies to design predictable and reliable cloud applications.
To achieve this, real-time performance will be required in all layers of the stack:
- The compute virtualisation layer must provide deterministic behaviour
- Similarly, the network layer has to provide bounded and controllable communication latencies. For example, in time-sensitive networks this implies a need for a higher-level controller integrated with the cloud orchestrator
- Data plane services, such as messaging and storage services need to provide guarantees on latency bounds
- The control plane services, e.g. scaling and resource management, have to be aware of the real-time requirements and to use appropriate real-time techniques.
The RT-cloud workshop aims at being a forum for researchers working with topics related to real-time cloud and real-time virtualisation techniques. In particular, researchers are encouraged to submit papers regarding:
- Deterministic hypervisors and cloud execution environments, e.g., real-time containers or real-time serverless compute
- Real-time cloud applications
- Use-cases and requirements for RT-cloud applications, e.g., automation and cloud robotics systems
- Experiments and experiences with RT-cloud
- Management and orchestration aspects, i.e., design and assign based on real-time requirements
- End-to-end resource management and resource scheduling
- Co-scheduling of virtual network and compute resources
- Assurance in RT-cloud, deterministic scaling, fail-over, migration, etc
- Modelling of network and cloud, execution stacks and applications
- Tools and technique for resource sharing and isolation
- Programming models for RT-Cloud
- Predictable orchestration and cloud operation, e.g., how to allocate and configure cloud resource in bounded time
All the submissions will be reviewed by the members of the technical program committee and accepted papers will be included in the workshop proceedings, that will be available online on the workshop website. For every accepted paper, at least one author is expected to register for the workshop and present the paper.
Submitted manuscripts should be written in English . The submission deadline will be announced soon.
- Johan Eker (Lund University/Ericsson)
- Luca Abeni (Scuola Superiore Sant’Anna – email email@example.com)
Technical Program Committee
- Karl-Erik Årzen (Lund University)
- Daniel Bristot de Oliveira (Red Hat)
- Tommaso Cucinotta (Scuola Superiore Sant’Anna)
- Gerhard Fohler (Technische Universität Kaiserslautern)