The goal of the VISION project is to advance the competitiveness of the EU economy by introducing a powerful ICT infrastructure for reliable and effective delivery of data-intensive storage services, facilitating the convergence of ICT, media and telecommunications.
This infrastructure will support the setup and deployment of data and storage services on demand, at competitive costs, across disparate administrative domains, providing QoS and security guarantees. In doing so, VISION will lay a foundation for the Future Internet, radically extending earlier generations of service infrastructures.
The prime deliverable of VISION will be an architecture and a reference implementation of a cloud-based infrastructure building on open standards and new technologies, capable of optimized delivery of converged data-intensive services. We will demonstrate how this infrastructure, where virtualization plays a key role, supports emerging Telco/Media services that are not supported today. We aim to achieve significant and quantifiable improvements in service delivery productivity, quality, availability, reliability and cost.
Scientifically, our work will advance the state-of-art in several dimensions, including:
-New Model of Data Storage, using objects with metadata to support effective access and management, enabling unprecedented scalability and simplification of storage services.
- Data Mobility and Federation, enabling comprehensive data migration and interoperability across remote locations, overcoming current limitations.
- Computational Storage, introducing a technology for specifying and executing computations close to storage, reducing overheads.
- Content-Centric Storage, facilitating access to data by content and its relationships, relieving applications and users from the need to be aware of the underlying storage implementation.
- Capabilities for Cloud-based Storage, supporting delivery of data-intensive services securely, at the desired QoS, at competitive costs.
The role of SICS
SICS leads the work on computational storage in VISION. We will provide an architected solution for bringing computing to storage, enabling secure execution of computational tasks near their data, applicable both to user-driven computational tasks as well as autonomous data derivation and transformation. They may involve user driven computational tasks as well as autonomous dynamic data derivation and transformation. The programming model for developing complex data-intensive applications is not limited to supporting only map/reduce, but also other kinds of large-scale data-intensive applications.
IBM (Coordinator), SICS, SAP, Telefonica, Siemens, Engineering, Nat. Tech. Univ. Athens, Deutsche Well, RAI, Umeå, SNIA, Telenor, FT, Uuiv of Messina, iTricity
PI at SICS: Per Brand, perbrand [at] sics.se
Duration: Oct 2010-Oct 2013