UBITECH has actively participated in the kick-off meeting, in Darmstadt, Germany (January 24-25, 2017) of the PrEstoCloud Research and Innovation Action, officially started on January 1st, 2017. The project is funded by European Commission under Horizon 2020 Programme (Grant Agreement No. 732339) and spans on the period January 2017 – December 2019. The PrEstoCloud project that make substantial research contributions in the cloud computing and real-time data intensive applications domains, since it provides a dynamic, distributed, self-adaptive and proactively configurable architecture for processing Big Data streams. In particular, PrEstoCloud combines real-time Big Data, Mobile Processing, Cloud computing and Fog computing research in a unique way that entails proactiveness of Cloud resources use and extension of the Fog computing paradigm to the extreme edge of the network. PrEstoCloud enables a smooth transition toward proactive, real-time Big Data, cloud-driven systems which are able to: (a) sense the need for adapting data-intensive services proactively, (b) define on-the-fly the most suitable changes in the real-time processing architecture including offloading of processing tasks at the extreme edge of the network, (c) predict reconfigurations in the underlying Cloud computing infrastructure resources, and (d) optimise continuously the infrastructure performance. PrEstoCloud is driven by the microservices development paradigm and has been structured across five different conceptual layers: i) Meta-management; ii) Control; iii) Cloud infrastructure; iv) Cloud/Edge communication and v) Devices layers.
UBITECH undertakes the technical integration lead of the project’s R&D activities, while UBITECH R&D team heavily contributes and leads the technological choices towards the definition and design of the PrEstoCloud Cloud-based Self-adaptive Real-time Big Data Processing Framework as well as of a semantic and extensible model that will describe the associations between Big Data processing types, data-intensive application fragmentations, fragments distribution constraints and potential workload requirements based on the characteristics of Big Data streams, allowing Cloud application developers to annotate their code in order to highlight meaningful fragmentations of the application functionality, declare distribution constraints and denote their potential workload needs. Moreover, UBITECH significantly contributes in the implementation of the Control layer components, which manages resources of the Cloud infrastructure layer, and of the Cloud-Edge Communication layer components – in particular, (a) the Autonomic Resources Manager which involves monitoring and management of Cloud resources capabilities that can be extended to the edge of the network, incorporating advanced capabilities for reliably and adaptability, dealing additionally with resources management at the edge of the network; (b) the Autonomic Data-Intensive Application Manager which is responsible for the scheduling of big data applications execution, for distributing meaningful fragments of data-intensive applications across clusters and launching the appropriate workers for controlling the processing workflow (deployed on the Cloud infrastructure layer); (c) the Spatio-Temporal Processing Mechanism which copes with clustering groups of devices, network congestion detection and geo-fencing capabilities for providing location-awareness with respect to cloud resources management and on/offloading processing tasks to the devices layer; (d) the Security Enforcement Mechanism that is responsible for providing the appropriate access and usage control mechanisms with respect to accessing, reallocating and reconfiguring network and edge resources as well as exploiting their processing outcomes, ensuring the protection with different granularity levels and enforcing this protection in different operational layers, all the way from the networking infrastructure up to the cloud application itself; and (e) the Inter-site Network Virtualization Orchestrator for coping with the need of virtualizing hardware resources situated in multi-cloud environments and managing their orchestration and provisioning across different and heterogeneous providers, including the control of the inter-sites network virtualization process in a secure way, through a close interaction with the Security Enforcement Mechanism.
