Article

SNIA: Expect XAM products by late 2008

Antony Adshead, UK bureau chief, storage
Retaining data for extremely long periods is obligatory for some organisations, either due to legislatory requirements which range from decades for some government records to a century for patient records under some regimes – or simply because it is valuable to the business. But while companies are storing more and more data digitally, that brings its own issues – the chief one being whether you will have the applications and hardware to read it in 50 or 100 years time.

Users will be able to migrate data between platforms and there will be no vendor lock-in.
Wayne Adams
chair emeritus and treasurerSNIA
Recently the Storage Networking Industry Association (SNIA) reached a further milestone in its quest to resolve that problem, with the announcement that its XAM – eXtensible Access Method – specification was set for release in mid-2008 and that its launch would also include a software developer kit.

SearchStorage.co.uk's Antony Adshead spoke to Wayne Adams, who is chair emeritus and treasurer with SNIA as well as a technologist at EMC, about the XAM specification.

Antony Adshead: What problem does XAM address?

Wayne Adams: XAM addresses one of the bigger problems of data storage, that of the creation of information and the ability to retain and migrate it over periods of many years. Today metadata about files regarding its attributes – when it was created, when it should be destroyed, who has access, and so on – is specific to the application and platform the data is held on, so if things stay the same 50 years into the future, you'd still have to have the same apps and hardware platform to be able to access it. The chief benefit of XAM will be to free data from that dependency.

Users will be able to migrate data between platforms and there will be no vendor lock-in. The same applies to specific application skills – you will not need to retain skills just to be able to access archived data.

Antony Adshead: What market will XAM unlock?

Wayne Adams: We expect it to unlock the content-addressed storage market, which had gone as far as it could for the moment. It will also begin to feature also in enterprise content management, email archiving and vertical-specific products such as medical records systems as well as custom projects such as archiving programmes for government agencies.

Antony Adshead: There were three distinct approaches to the spec proposed: client-side API or driver; pure protocol; and file system version. Which has been adopted?

Wayne Adams: XAM will be incorporated into products as a device-specific driver, a decision that was taken in late 2006 after looking at what would work best in the market.

Antony Adshead: What demonstrations of XAM's capabilities have you been able to carry out to date?

Wayne Adams: We did a large integrated demo at Storage Networking World in Florida last month in which we were able to show the management of data with a common interface from three hardware platforms, from EMC, HP and Sun. On the application side, there were four different ones: an HP database, an EMC disk archiving application, Vignette enterprise content management and a Sun digital image management application. What we didn't demo was the ability to migrate data between platforms, which XAM will also enable.

Antony Adshead: When will we see XAM incorporated into products?

Wayne Adams: We should start to see platform products by the autumn of this year and applications by the end of this year or the beginning of 2009. Version 1.0 is out with our members for ratification at the moment and, following that, it will be submitted for ANSI and ISO accreditation.


Email Alerts

Register now to receive ComputerWeekly.com IT-related news, guides and more, delivered to your inbox.
By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy
 

COMMENTS powered by Disqus  //  Commenting policy