The eXtensible Access Method (XAM) provides an API for content aware storage (CAS) platforms.
By Christina Casten
Content aware storage (CAS) is a category of automated networked storage whereby large volumes of fixed content are stored over extended periods of time. Unlike traditional NAS that is designed to facilitate collaboration and file sharing, or SANs, which focus on performance, CAS is specifically designed for fixed content that has a significantly extended lifecycle compared to transactional data.
In late 2005, the Storage Networking Industry Association (SNIA) voted to accept a foundational document, and its Fixed Content Aware Storage Technical Work Group (FCAS TWG) adopted the eXtensible Access Method (XAM) specification as a starting point for development of a fixed-content API.
The XAM specification is aimed at providing applications with a standard interface and metadata to communicate with object-based storage devices such as CAS platforms.
XAM addresses the rapidly growing and increasing role of fixed content storage within overall storage architectures. According to the Enterprise Strategy Group (ESG), fixed content is growing in volume at an annual rate of 92%.
How it works
The XAM interface specification defines a standard access method between “consumers” (applications and management software) and “providers” (storage systems) to manage fixed content. XAM provides a standard interface and metadata for applications to communicate with CAS systems. XAM annotates objects with metadata, providing for the management of information at a semantic level. This coupling allows policies to make intelligent decisions about the management of objects without referring back to the application. XAM abstracts the access method from the storage.
The interface is intended to achieve interoperability, storage transparency, and automation for information lifecycle management (ILM) practices, long-term records retention, and information assurance (security).
Today’s CAS products provide little, if any, interoperability. Existing products use incompatible access and management methodologies (e.g., naming, retention, expiration, and placement of content) that require that application software be modified, often extensively, to operate with any particular CAS system. No provisions are made for content sharing between different applications. No standards exist for movement of content between different CAS products, and each product provides a different management interface.
Unlike file systems, XAM tags content with meta- data and provides a technology- independent namespace, allowing software to interpret the content independent of the application.
XAM and ILM
One of the major areas of impact for XAM technology will be on the adoption and implementation of ILM practices. For example, XAM provides an object-oriented location and storage technology-independent approach to data storage.
By tying content to a globally unique name, XAM can efficiently manage the content without application concern for a specific physical location of that content or technology it resides on.
XAM also raises metadata-contextual information about the content being stored-to the same level of importance as the content itself. By bundling content and metadata together, applications can more easily manage and share information about stored content, facilitating data sharing and ILM.
And as an interface between applications and the physical store, XAM metadata allows policies to make intelligent decisions about the management of objects without referring back to the application-the ultimate ILM.
Requirements from business units will demand that data be managed and meet various retention requirements in a cost-effective manner. Over time, XAM- compliant devices will be one of the available resources for maintaining the longevity of business data. By abstracting the physical assets of storage with a data persistence perspective, the underlying storage can age and run the course of its product lifecycles while the XAM architecture provides a consistent view of the managed content and all of its attributes. This capability is important as the adaptive data center will also be able to age and retire physical assets while preserving the various services through other abstractions for the different types of resources.
Another important role that XAM will participate in is in the emerging concept of data grids. A data grid can be viewed as a data resource location service that provides a unified view of many, if not all, data repositories or sources within a pre-defined domain. This can include databases, file-level and object-level network services, and archives.
As the maintainer and container of data and metadata, XAM can contribute the content, and possibly the metadata, to this data resource location service for enterprise-wide data mining and analysis, digital asset discovery, and other purposes.
While still under development in the SNIA’s FCAS TWG, the emerging XAM specification will offer a variety of benefits. For example, application providers will benefit by “writing once to many,” resulting in lower cost of development, support, and maintenance. Also, storage vendors will see a larger market and greater demand for XAM-compliant devices. Ultimately, however, it will be the end users that will benefit from storage technology independence as well as increased mobility of their data.
Recognizing the importance of these benefits, more than 95 individuals from 34 companies representing storage vendors, application providers, and users are contributing to the XAM development effort.
XAM demonstrations are expected at the Storage Networking World 2007 conference. For additional information, go to www.snia-dmf.org/xam/index.shtml.
Christina Casten is XAM co-chair for the SNIA (www.snia.org).