This article was originally published on Technology.Info.
As part of our continuing strategy for growth, ITProPortal has joined forces with Technology.Info to help us bring you the very best coverage we possibly can.
How do you manage data when it just won’t stop growing? That’s the question that many IT teams are now asking, as they struggle to cope with data volumes that are growing at clips of 40 percent, 50 percent and even 60 percent per year, according to John Rollason, director of product, solutions and alliances marketing in EMEA at storage company NetApp.
“They’re suddenly having to pay attention to storage in a way that they haven’t needed to before,” he says. In particular, the gains they’ve seen from server virtualisation and consolidation - increased utilisation rates, easier management - are prompting them to ask whether similar benefits might not be gained from virtualising storage. After all, says Rollason, with traditional networked storage, woefully inefficient utilisation rates of just 25 percent to 30 percent are not uncommon.
“Customers are starting to think: if we can virtualise storage, too, by abstracting away from the hardware layer or, to use the latest buzzword, make it more ‘software-defined’, then we might be able to create dramatic new efficiencies, just as we did with server hardware.”
At the same time, it would allow them to create a virtual ‘pool’ of storage resource from the different disks sitting in different arrays scattered around their environments, opening up a path to private and hybrid cloud storage.
That’s where ONTAP comes in: it’s the company’s core storage operating system, the intelligence that runs all NetApp storage, from small systems for small companies and remote offices to the multi-petabyte environments used by large enterprises. And thanks to recent investment and development of the product, it’s also the software that will enable customers to abstract raw hardware to create a virtualised, private-cloud-friendly pool of storage resource. In effect, it will put the software in software-defined storage for them.
“ONTAP is our big bet, and all our major investments are now in software,” says Rollason. In June, the company announced an update to ONTAP - Clustered Data ONTAP 8.2 - to handle bigger clusters and require less downtime.
Clustering is hugely important to NetApp: making clustered systems work at scale is important to the company’s customers but, in the past, they’ve reported that its systems can become difficult and complex to manage at scale.
In response to these challenges, the company first started supporting clustered NAS operations in version 8.1.1, released in early 2012. Clustered or scale-out NAS enables multiple NAS devices to be linked together with a common operating system and a parallel file system running right across a cluster, so that the storage resources beneath can be scaled to provide large amounts of capacity.
The latest release of ONTAP provides a considerable jump in capacity on this earlier version. Clustered Data ONTAP 8.2, with its networked file and block data access, provides non-disruptive operations and supports up to 69 petabytes of capacity, 24 controller nodes, 49,000 LUNs (logical unit numbers), 12,000 NAS (network-attached storage) volumes accessible by more than 100,000 clients and single file containers of up to 20 petabytes in size.
In the past, scale-out was largely confined to high-performance computing (HPC) and large-scale enterprise environments. With these upgrades to ONTAP, NetApp is making it applicable to more general-purpose uses.
That’s good news for IT teams under pressure from massive data growth and business expectations for fast response times, said Terri McClure, an analyst with Enterprise Strategy Group, at the June announcement of Clustered Data ONTAP 8.2.
“Businesses are seeking storage infrastructure solutions that deliver agility and availability, that ensure fast provisioning of underlying IT services to keep applications up and running, as well as bring new applications on board quickly. And doing so in a manner that ensures applications can coexist in the same infrastructure without the worry of a rogue application taking resources from business-critical ones.”
“With clustered Data ONTAP, NetApp is virtualising the underlying storage infrastructure in such a way as to enable this type of flexible and efficient services-oriented data storage infrastructure,” she continued.
Rollason says that while ONTAP’s roots go right back to the founding of the company in 1992, the product has been written “from the ground-up” over the last five years to offer support for clustering. Around 90 percent of all storage workloads could be run on a clustered architecture, he reckons.
“You can see we’re turning into a software company, because we’re announcing a new software release every 12 to 18 months,” says Rollason. “We’ll have an 8.3 release to announce in the next 12 to 18 months.”
There’s clearly no doubt over the importance of ONTAP to NetApp’s future prosperity among the company’s executives. According to Rollason: “Clustered ONTAP is now the biggest differentiator NetApp’s got.”
The company must now work force to capitalise on that differentiator and, at the same time, stay one step ahead of accelerating customer demand for software-defined storage.