Abstract | With the emergence of a wide range of Internet of Things (IoT) devices (e.g. sensors or cameras) and new computing paradigms such as edge and fog computing, science gateways and their enabling technologies also face novel challenges. For example, collecting and analysing large amounts of data in the Cloud-to-Edge computing continuum exposes the limitations of current centralised orchestration solutions. A Cloud-to-Edge orchestration system is responsible for the automation of application deployment and their runtime management by providing simultaneous access to the heterogeneous resource landscape of the continuum. Most currently available orchestration tools, responsible for deploying and managing applications in the Cloud-to-Edge computing continuum, are based on a certain level of centralisation. Such centralisation, while relatively easy to implement, carries several disadvantages. The central component can be easily overloaded as the system scales and becomes a single point of failure and a good target for security attacks. Additionally, such a centralised approach does not fit well with the highly distributed and dynamically changing nature of the computing environment. A centralised management approach cannot react fast enough to some changes in local environments (e.g. volatility of resources) and cannot support fast adaptation of resources and application requirements (e.g. due to the movement of certain computing elements). To address these challenges the Swarmchestrate EU-funded research project, started in 2024, aims to create a completely decentralised, autonomous, secure and self-organised application orchestration system by combining and extending emerging technologies, such as Swarm computing, distributed AI, distributed ledger systems and decentralised identity management. The approach applied by Swarmchestrate is fundamentally new to application orchestration and suitable for managing hyper-distributed applications that span across large distances and the different layers of the dynamic compute continuum. As the work in the project started only recently, the main focus of this presentation will be on the generic architecture and the major technological components of the decentralised Swarmchestrate orchestration concept, based on Figure 1. From a structural viewpoint, the overall system is divided into four distinctive parts, comprising of (1) Application view – dealing with application specification, (2) Orchestration space – handling all core functions of orchestration, (3) Trusted knowledge management – secure handling and management of all system-wide knowledge and interface required to enable trust and transparency, and (4) Resource layer – representing resources of the continuum that span across multiple cloud environments and non-cloud layers. Additionally, the presentation will outline early results regarding the feasibility of the decentralised orchestration and its interconnection with the creation and management of swarms based on distributed intelligence. The system is anticipated to achieve optimised deployment of (data-processing) applications based on their performance requirements, thanks to AI-driven decision-making from cooperating swarms. Work on the project is ongoing, and the first version of the Swarmchestrate orchestrator, focusing on the initial deployment of microservices-based applications, is expected by the end of the year. The implementation will utilise the experiences of the developer team gained from the MiCADO cloud orchestrator solution, transforming that into a fully decentralised framework. |
---|