PrEstoCloud Banner
PrEstoCloud logo
Newsletter October 2017 - Issue 1

Message from the Coordinator

Hello everybody and welcome to the world of PrEstoCloud!

PrEstoCloud stands for an efficient exploitation of infrastructural resources from edge devices to the clouds. The inflow of Big Data streams from static or mobile edge devices is challenging cloud infrastructures by their ever-changing velocity, volume and data variety. The PrEstoCloud software platform will improve cloud infrastructures in order to cope with Big Data applications.
PrEstoCloud is being developed in close partnership with industrial partners from different business sectors: Logistics, Media and Surveillance, which will ensure validation against various business needs such as cost-efficiency, spatio-temporal and response-time related business requirements.
I'm pleased to present the first project e-newsletter and hope that you find this e-publication useful and of course we welcome your feedback!

Enjoy reading and stay curious!

With best regards,

Birgit Helbig,
Project Coordinator

Meet our Project Team

The multinational research consortium consists of very experienced researchers and technology developers, who have already provided parts of the system that will be used as a basis for developing the PrEstoCloud platform. The consortium unites a total of eleven partner organizations from industry and research. Software AG has been entrusted with the leadership of the consortium.

PrEstoCloud Dubrovnik

ACTIVEeon logo Adtitess logo cnrs logo cvs-mobile logo iccs logo
ijs logo liveu logo amram-tech logo nissatech logo Ubitech logo

PrEstoCloud Use Cases

Validating PrEstoCloud against Real Scenarios

When the PrEstoCloud research consortium members met for a face-to-face in Dubrovnik, a joint understanding between all project partners regarding the setting of the business use cases under investigation was reached. Focusing on quite different scenarios will ensure that the envisioned software platform is validated against varying business needs such as cost-efficiency, spatio-temporal and response time related business requirements. What they all have in common is the inflow of big data streams from static or mobile edge devices challenging cloud infrastructures by their ever-changing velocity, volume and data variety:
  • The "Big Data Media" scenario combines consumer and producer feeds and offers personalized and flexible service; for example, on-the-spot freelance reporting of a big sport event will be combined with traditional news as well as news feeds from social media in real-time. PrEstoCloud will enable cost optimization of the technical infrastructure and an extended real-time media offering.
  • The "Big Data Logistics" application addresses real-time needs from large logistics companies which aim to monitor their fleet and entrusted goods. PrEstoCloud will reduce insurance costs while increasing safety on the road.
  • The "Big Data Surveillance" application covers the supervision of special areas like public urban places by camera clusters and drones. PrEstoCloud will enable the evaluation of Big data video streams in real-time in order to ensure instant threat detection and automatic alert in accordance with the new European data protection regulations.

PrEstoCloud Architecture and Technologies

Architecture Overview

One of the most previous important results is the conceptual architecture that describes the components of the system and the relationship/interaction between them. Since PrEstoCloud aims to enable a continual improvement (reconfiguration) of a complex computing infrastructure, the architecture follows, on a higher abstraction level, the self-adaptivity pattern: Monitor Analyze Plan Execute Knowledge (MAPE-K model; so called "architectural blueprint for autonomic computing", introduced by IBM).

The PrEstoCloud architecture has been structured across five different layers:

  • The Meta management layer mainly consists of decision logic capabilities required for enhancing the PrEstoCloud Control layer. Modules of this layer will use as input the situation details, the variation of the Big Data streams and the context of the mobile devices at the extreme edge of the network in order to recommend, at the appropriate time, the necessary adaptations of used resources in the real time processing network.
  • The Control layer manages resources of the Cloud infrastructure layer and contains modules which will monitor and manage cloud resources capabilities that can be extended to the edge of the network. Moreover, this layer is responsible for the scheduling of big data applications execution over the resources of the real time processing network.
  • The Cloud infrastructure layer will provide dynamic placement and scheduling capabilities allowing the utilization of the extreme edge of the network, assimilated to private clouds, jointly with public clouds. The placement of microservices and application fragments will be handled based on placement constraints reflecting the requirements of the expected behaviour of the application. Constraints will express different properties like response time, security constraints or other provider wishes.
  • The Cloud-Edge communication layer contains 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. This also includes the control of the inter-sites network virtualization process in a secure way, through a Security Enforcement Mechanism. This layer will also relay data streams on and off the PrEstoCloud platform providing standard publish/subscribe event brokering capabilities.
  • The Devices layer consolidates any kind of device that can be used as a Big Data stream source or as a mobile computational node at the extreme edge of the network.

Following figure illustrates the mapping between these layers and MAPE-K model:

PrEstoCloud on Tour

Creating a trusted and transparent cloud market was the key theme of the CloudWATCH Summit which took place in Amsterdam on the 20th of September 2017. Pioneering SMEs, industry leaders and policy experts discussed, how the European cloud computing market will be shaped in Europe over the coming years. The CloudWATCH roadmap on understanding the economic, financial and market dynamics shaping the cloud computing market in Europe, and globally has been outlined in the summit.
Startups and projects pitched their innovative products, services and research results and described the importance of ensuring that cloud products are truly ready for market needs. Presenters highlighted the positive impact of the new GDPR (General Data Protection Regulation) regarding trust and effectiveness for both European and non-European resident market players who operate in Europe.
An expert panel discussed how to manage risk and how to look at the business value of cloud standards. Security certifications and interoperability standards are essential for ensuring that cloud-based services can be trusted by end-users and become a unique selling point for providers. Experts demystified the standards landscape and offered guidance on adopting essential standards allowing cloud technologists to stay safe and avoid vendor lock-in.
PrEstoCloud was represented in the Summit by Birgit Helbig, Software AG, and Dimitris Apostolou, ICCS, who presented a poster at the event highlighting the innovative aims and ambitions from the research project.

PrEstoCloud organised successfully the special session entitled "Intelligent systems and services for cloud accessible, data-intensive computing, ISS-CLOUD" in the 8th IEEE International Conference on Information, Intelligence, Systems and Applications, in Larnaca, Cyprus, 28-30 August 2017.
The special session focused on the integration of cloud computing with mobile services, which enables the development of systems that provide resources and services on an on-demand basis, process big data collected from mobile sensors, and support Internet of things (IoT) with massive cloud-based backend. Such systems may employ computational intelligence methods, utilise artificial intelligence approaches as well as decision making techniques for enhancing the effectiveness and efficiency of computing infrastructures, increasing security and data privacy while making infrastructure configuration, deployment and maintenance easier.

The following papers were accepted and presented in the special session:

  1. Seamless task offloading on multi-clouds and edge resources: An experiment, Andreas Tsagkaropoulos, Yiannis Verginadis, Dimitris Apostolou and Gregoris Mentzas
  2. Wide Area Video Surveillance Based on Edge and Fog Computing Concept, Georgios Kioumourtzis, Michalis Skitsas, Nikolaos Zotos and Anargyros Sideris
PrEstoCloud partners ICCS and ADITESS participated with papers which described early work on managing the seamless offloading of computation effort between cloud and egde resources (paper #2) as well as the requirements stemming from the use of cloud and edge computing resources for wide area video surveillance (paper #6), respectively.

Dr. Harald Schöning, Vice President Research at Software AG, gave a keynote speech on the topic of "A dynamic and distributed software architecture that manages cloud and fog resources proactively" at the 11th Symposium and Summer School On Service-Oriented Computing (SummerSoC) in Crete, Greece. The lecture on June 28th was about PrEstoCloud.
At SummerSoC, renowned researchers and leading industry practitioners gave tutorials and invited talks on current topics from the areas above. PhD students presented posters of their ongoing work and gave plenary presentations of their theses.

More information

Intensive discussions during the PrEstoCloud meeting in June
Taking advantage of the inspiring air in beautiful Dubrovnik, our face-to-face meeting in month six was dedicated to a joint understanding between all project partners regarding the setting of the use cases.
Business KPIs like "paid cloud resources", "improvement of QoS performance" and more were mapped to technology assets such as "fog deployment", "autonomic resource manager" and others. Equally demanding were the big data architecture discussions and the deduction of technical requirements for the envisioned platform.

Dr. Yiannis Verginadis from the Institute of Communications and Computer Systems (ICCS) participated in the 7th International Conference on Cloud Computing and Services Science (CLOSER 2017) that was held in Porto from the 24th to 26th of April. The specific conference focused on several of the established and emerging areas of Cloud Computing concerning the infrastructure, operations, and available services through the global network. Dr. Verginadis presented a joint (ICCS, Activeeon and Nissatech) conference publication entitled: "PrEstoCloud: Proactive Cloud Resources Management at the Edge for Efficient Real-Time Big Data Processing". This paper essentially discussed the main challenges that PrEstoCloud aims to address and provides details on the PrEstoCloud vision that encapsulates a dynamic, distributed, self-adaptive and proactively configurable architecture for processing Big Data streams.
During this presentation, Dr. Verginadis argued on the advantages of combining real-time Big Data, mobile processing and cloud 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. A very focused and related audience of computer scientists with background on multi-clouds and fog computing domains, allowed for a very fruitful discussion and exchange of positive opinions over the mentioned issues.

Upcoming Events

  • November 1-3 in London, United Kingdom
    ACM Internet Measurement Conference (IMC): CNRS will present its work within PrEstoCloud, as well as preliminary results, with a poster entitled "Towards non-intrusive measurements of available bandwidth for multi-cloud applications"
  • November 7-8 in Nicosia, Cyprus
    Technical plenary meeting - Focus topic: Architecture


  • PrEstoCloud: Proactive Cloud Resources Management at the Edge for Efficient Real-Time Big Data Processing, presented at the 7th International Conference on Cloud Computing and Servcie Science in Porto, Portugal
  • Seamless task offloading on multi-clouds and edge resources: An experiment, presented at the 8th IEEE International Conference on Information, Intelligence, Systems and Applications, in Larnaca, Cyprus, 28-30 August 2017
  • Wide Area Video Surveillance Based on Edge and Fog Computing Concept presented at the 8th IEEE International Conference on Information, Intelligence, Systems and Applications, in Larnaca, Cyprus, 28-30 August 2017

Dissemination Material

Related Project Links

  • CloudLighting (CloudLightning: self-organising and self-managing heterogeneous cloud -, now in its final year, is delivering a Service Oriented Architecture for the heterogeneous cloud which moves the management burden from the customer to the provider. PrEstoCloud aims to further support the provider with tools for efficient management of cloud and edge resources.
  • COLA (Cloud Orchestration at the level of application -, a project that started together with PrEstocloud, will define and provide a reference implementation of a generic and pluggable framework that supports the optimal and secure deployment and run-time orchestration of cloud applications. PrEstocloud has a similar goal with a focus on hybrid clouds that include also resources at the extreme edge of the network.

Join us on Facebook, Twitter and Linkedin

We invite you to use our online networking tools to make connections, exchange scientific ideas, and keep informed about the latest research findings. On Facebook, Twitter and Linkedin you will enter rooms for exchange of experiences, among professionals and stakeholders - a great opportunity to connect with a greater scientific an industrial community.

Facebook Twitter Linkedin

January 2017 - December 2019
Reference 732339 • Volume 4.2 mio. Euro • Partners 11