Scientific workshops

Two scientific workshops have been promoted and organized, as planned, by the SeaClouds project, and both were accepted as part of the European Conference on Service-Oriented and Cloud Computing (ESOCC) program in the last two editions. The first one, held on September 2nd, 2014 in Manchester, United Kingdom (see D1.7.1), and the second one, on September 15, 2015 in Taormina, Italy (see D1.7.3).

In both cases, the objective was to provide a forum to discuss challenges, solutions and perspectives of ongoing research and standards development activities aimed at enabling an efficient and adaptive management of service-based applications across multiple clouds. The second edition of the workshop was organized in conjunction with the CloudWave project.

From the organizational point of view, these events may be considered as a success, in particular for the EU project track, and the invited talks, not only for the number of attendees, but also for their positive feedback. As it was planned, one of the objectives of these workshops organization was the dissemination of the SeaClouds project.

Regarding the content of the workshops, there were several aspects discussed by participants, focused around the QoS-based cloud deployment, vendor lock-in problems, discovery approaches for cloud service selection matching risk-driven cost requirements, and monitoring management. On the other hand, the importance of innovation and exploitation aspects being highlighted in new EU projects was at the core of the discussions, together with the need of more EU investment in basic research.

Key conclussions from the workshops include:

  • The diversity of cloud solutions has increased the challenges for interoperability and portability. New issues are raised on QoS management when multi-cloud deployment is involved, and modules are composed in different clouds.

  • Multi-cloud deployment on heterogeneous cloud providers, has raised out new challenges for monitoring; in particular, for data distributed on different clouds.

  • Discovery has been usually related with QoS and technical requirements matching capabilities. However, other important issues have to be dealt with when risk-driven cost-benefit analysis is considered for multi-cloud selection of services.

  • Coordination and adaptation techniques have proved to be useful for improving QoS experienced by the application users.