Research initiative aims to provide efficient delivery of data intensive storage services by understanding what’s inside the data
IBM this week announced that it is leading a joint EU-funded research initiative called VISION Cloud – Virtualized Storage Services for the Future Internet. As part of the initiative, IBM is working with 15 European partners to develop a smart cloud storage architecture to improve the global delivery of rich data and storage services across boundaries of countries and vendors. The research project will tackle the major challenges facing today’s storage clouds, including: cost-effectiveness, data mobility across cloud providers, security guarantees, and the massive computing power demands that are affecting quality of service.
The project aims to develop a new approach, where data is represented by smart objects that include rich information describing the content of the data and how the object should be handled, replicated, or preserved. The new architecture will use the knowledge of the content being stored to improve data mobility and enable more efficient and secure methods of computation.
According to the EMC Digital Universe Study, in 2009, the amount of digital information grew 62% over 2008 to 800 billion gigabytes. The study predicted that the amount of digital information created annually will grow by a factor of 44 from 2009 to 2020, as all major forms of media — voice, TV, radio, print — complete the journey from analog to digital.
“The world is generating data faster than we can store it,” explained Kristof Kloekner, IBM VP Strategy & Enterprise Initiatives, Systems & Software & CTO Cloud, “and we’ve become critically dependant on services that can extract valuable information from the data and help drive better decision making. By optimizing delivery of data-intensive storage services, VISION Cloud will usher in a new era of more flexible, scalable, and secure cloud storage that can be delivered in a pay-per-use model.”
The VISION Cloud aims to develop a breakthrough storage cloud architecture which combines several new concepts, notably (a) a rich object data model, (b) execution of computations close to the stored content, (c) content-centric access, and (d) full data interoperability. Based on this architecture and the supporting technologies, VISION Cloud will enable the delivery of new classes of rich data-intensive services, offering unprecedented functions and providing high quality of service and security guarantees.
For example, with VISION Cloud, a future provider of storage services could offer a “digital safe” service, where people can safely store their digital information, including photos, videos, health records, financial records, and more. Then, for example, when an expecting mother undergoes a 3D ultrasound, the image is automatically uploaded and stored in her digital safe. This ultrasound is stored together with metadata, such as the date, format, stage of pregnancy, or content description; the metadata is stored as an integral part of the smart ‘data object’. She can then give access to family, friends, or editing services, enabling them to download and view the video – no matter what kind of format their device or computer works with. Meanwhile, a radiologist with access to the video finds a slight anomaly and can tag the scan accordingly. The tagging triggers a search in a medical repository, which finds that the probable cause of the anomaly is an allergy medication.
This scenario exercises innovations of the VISION Cloud, including:
• The data model allows meaningful metadata to be associated with the scan.
• Secure access control allows the mother to share the video with her friends and family
• Data mobility and interoperability allow her friends and family to view the video, though they use different storage providers
• Computational storage provides safe compute power next to the video to transform it to the appropriate formats and resolutions for the personal devices of her friends and family. It also provides safe compute power next to the scan to suggest the diagnosis.
• Content centric access to the material in her safe allows it to be organized according to its content and their relationships, thereby allowing medications to be quickly identified in her
medical records.
As another example, a small business may upload its financial records to a cloud to only find out at the end of the year the vendor has raised the prices of it services. Today, unless the small business is willing to pay the increased cost, its only option is to download all of the content and choose another cloud provider. With the technology of VISION Cloud, which is built on open standards, the financial data can easily be transferred from one cloud provider to another, saving the small business the trouble of downloading and securing what could amount to terabytes of data. Alternatively, the small business could have its data split across the two providers. Given the open standards of VISION Cloud, the less expensive provider could offer access to data still residing on the old provider. The transfer of data from the old provider to the new one could begin as a background process or even in an on-demand fashion.
“Users now require capabilities that are far richer than simple raw bits and basic storage,” noted Dr. Hillel Kolodner, IBM Haifa researcher and lead architect for VISION Cloud. “To satisfy this need, the focus must shift to the data—where the collections of bits are semantically meaningful and have associated metadata. Data should be a first class citizen, where its importance is comparable to the computing power itself. With VISION Cloud, our aim is develop the infrastructure to support this prominence of data and data-intensive services.”
Scientists at IBM Research in Haifa, Israel will lead the project and the consortium of partners from academia and industry. The consortium partners include come from academia and industry and include IT technology, business software, and service providers: SAP AG, Siemens Corporate Technology, Engineering and ITRicity; telecommunications providers: Telefonica Investigacion y Desarrollo, Orange Labs and Telenor; media service providers: RAI and Deutche Welle; the SNIA Europe standards organization, and leading universities: National Technical University of Athens, Umea University, Swedish Institute of Computer Science, and University of Messina.
