New infrastructures, such as Edge Computing or the IoT-Edge-Cloud computing continuum, make cloud issues more complex as they add new challenges related to resource diversity and heterogeneity (from small sensor to data center/HPC, from low power network to core networks), to the geographical distribution, as well as to the increased dynamicity and security needs, all under energy consumption and regulatory constraints.
In order to efficiently exploit new infrastructures, we propose a strategy based on a significant abstraction of the application structure description to further automate application and infrastructure management. Thus, it will be possible to globally optimize the resources used with respect to multi-criteria objectives (price, deadline, performance, energy, etc.) on both the user side (applications) and the provider side (infrastructures). This abstraction also includes the challenges related to the abstraction of application reconfiguration and to automatically adapt the use of resources.
The Taranis project addresses these issues through four scientific work packages, each focusing on a phase of the application lifecycle: application and infrastructure description models, deployment and reconfiguration, orchestration, and optimization.
Project leader: Christian Perez
— Project objectives
The objectives of the Taranis project are to contribute to advancing the state of the art with regard to a number of issues:
- Specification of a formal, static (structure) and dynamic (behavior) semantic framework — for the definition and verification of a generic application and infrastructure description model taking into account the specific features of target platforms: geographical distribution, heterogeneity, dynamicity, volatility/resilience, multi-tenancy, energy impact, security, etc.; realization of proofs of concept; feasibility study of model evolution in production; realization of adapted development environments (smart Cloud-native IDE).
- Specification and proof-of-concept of a methodology and tools for defining domain-specific application description models; applicability study of existing models and methodologies; validation via application case studies in representative domains.
- Specification of a configuration language for uniformly modeling all layers of a global distributed system; realization of proofs of concept; study of applicability and interoperability with respect to existing languages.
- Specification and realization of proofs of concept for safe, robust and decentralized automatic management of applications and infrastructures (algorithms and techniques for choreographing decentralized, high-performance, fault-tolerant and secure orchestrators for deployment and reconfiguration); case studies of deployment and reconfiguration on various infrastructures.
- Study and implementation of solutions (exact solver, (meta)-heuristics, control and machine learning-based solutions) to optimize infrastructure use with respect to multi-criteria objective functions and integrating application (user) and infrastructure (supplier) considerations; optimization of an application and a cohort of applications on target infrastructures; case study of integration into simulators and operational systems.
Addresses the complexity of cloud-edge application and infrastructure models: formal verification and optimization of these models, multi-layer variability, the relationship between model expressiveness and efficient solution computation, lock-ins of proprietary models, and heterogeneity of cloud application and infrastructure modeling languages.
Studies deployment and reconfiguration related issues to reduce management complexity and increase support for provisioning and configuration languages, while improving operations certification and increasing operations concurrency. The workpackage also aims to reduce the complexity of the bootstrapping problem on geo-distributed and heterogeneous resources.
Aims at extending the orchestrators for the IoT-Edge-Cloud continuum, while making them more autonomous with respect to dynamic, functional and/or non-functional needs, in particular with respect to the network partitioning problem specific to IoT-Edge-Cloud infrastructures.
Aims to revisit optimization problems associated with the use of IoT-Edge-Cloud infrastructures and the execution of an application when a large number of decision variables need to be considered jointly. It also aims to make optimization techniques aware of the ioT-Edge-Cloud continuum, the heterogeneous distributed platforms and the wide range of application configurations involved.
Project work packages
