There are several industry organizations and vendors hard at work developing cloud storage standards and the race is on to determine how data will be moved and secured in the cloud. The most recent race entrant in this race is a little company called IBM.
IBM (NYSE: IBM) today launched a joint research initiative of 15 European partners to develop a so-called “smart cloud storage architecture.” The effort centers on delivering storage services within and across cloud boundaries through a better understanding what’s inside the data.
The EU-funded initiative, called VISION Cloud – Virtualized Storage Services for the Future Internet, has been formed to develop a new approach to cloud storage, where data is represented by smart objects that include information describing the content of the data and how the object should be handled, replicated, or preserved. The end game is to improve delivery of data and storage services across boundaries of countries and vendors (See Alan Earls’ recent piece for Enterprise Storage Forum, “Could Borders Bring the Cloud Down to Earth?” for more on that subject).
The VISION Cloud storage cloud architecture concept combines (a) a rich object data model, (b) execution of computations close to the stored content, (c) content-centric access, and (d) full data interoperability.
The VISION Cloud initiative will be spearheaded by scientists at IBM Research in Haifa, Israel, and supported by partners, including SAP AG, Siemens Corporate Technology, Engineering and ITRicity, Telefónica Investigación y Desarrollo, Orange Labs and Telenor, RAI and Deutche Welle, the SNIA Europe standards organization. The National Technical University of Athens, Umea University, Swedish Institute of Computer Science and University of Messin, will also contribute to the effort.
A noble effort, to be sure, but Big Blue has some big vendors trying to make their architectures and delivery methods the de facto storage standards in the cloud.
Oracle staked its own claim to the cloud in recent weeks when the computer giant proposed the Oracle Cloud Elemental Resource Model API, which covers the common elements of a cloud implementation by specifying the relevant machines, storage volumes and networks. Specifically, the spec submitted to the DMTF describes how a machine can be provisioned from an image; how a volume can be attached to a machine; and how a machine can connect to a network (see “Oracle Proposes Cloud Management Standard”).
Oracle (NASDAQ: ORCL) said the goal of its proposal is to encourage open standards, noting the Oracle Cloud API follows the Representational State Transfer (REST) architecture style and uses HTTP methods to interact with the resources to achieve provisioning, associating, modifying, and retiring of entities.
Meanwhile, the Storage Networking Industry Association (SNIA) has been quietly putting in work via the SNIA Cloud Storage Initiative (CSI). The CSI recently completed work on its first cloud standard, the CDMI (Cloud Data Management Interface). The CDMI provides standards for the data path to clouds, includes the ability to manage service levels that data receives when it is stored in the cloud, and includes a common interoperable data exchange format for securely moving data and its associated data requirements between clouds.
The CDMI is based on a RESTful HTTP protocol, and requires providers to implement stringent access controls and encryption of data for security purposes. The standard enables developers to mark cloud storage containers and data objects with Data System Metadata to enforce service-level requirements for the data (see “SNIA Advances Cloud, Green Storage Standards”).
It will be interesting to see how the co-opetition plays out. Is a common management method for the cloud on the horizon? Or, are we doomed to proprietary clouds with proprietary data formats?