NAS or SANs? Will Your (Future) Storage Manager Decide?

Enterprise Storage Forum content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.

If your company is short on storage, should you turn to an outside SSP (storage services provider) or go it alone? If you do keep storage management in-house, does it make more sense to invest in SAN (storage area network) or NAS (network-attached storage)? More organizations are now hiring their own storage managers to help deal with these matters, according to some speakers at last week’s Internet World show.

In many organizations, though, storage management still falls somewhere between the cracks, said Jack Domme, Senior Director of Product and Sales and Engineering at StorageNetworks, during one session at the conference in New York City. Systems administrators, network managers, and DBAs (database administrators) each own some – but not all — of the requisite skills.

“Each understands a niche,” Domme maintained. A DBA, for instance, might go to a network manager or systems administrator for help with disk partitioning. Meanwhile, in many places, nobody on board is well acquainted with the ins-and-outs of SAN switching fabrics, for example.

More platforms, more complexity
In another session, Vixtel CTO Stuart Berman noted that heterogenous environments can bring even more complexity. Berman pointed to research by the Aberdeen Group, in which 75 percent of survey respondents said they were unwilling to deploy SANs across multiplatform Unix/Microsoft Windows enterprises.

Smaller companies often add storage to the IT director’s already long list of responsibilities, Berman said. Larger organizations dish out storage management chores in myriad ways.

Several panelists pinpointed a trend toward adding storage administrators to the company payroll. Hewlett-Packard is one company that already uses this strategy internally. HP’s storage administrators collaborate not just with other technicians, but also with “knowledge engineers” and “content managers” from HP’s business side, said John Selep, product marketing manager in HP’s Data Protection division.

HP’s internal storage management is further broken down into subcategories. “We even have tape management people,” Selep elaborated. “A lot of specialization can emerge.”

Selep didn’t advocate this level of specialization for everyone, however. “You can’t afford to double your staff every two years,” he said. “You need tools.”

Tools can add problems to storage stew
Storage management tools, though, can be problematic in and of themselves, according to Berman. Right now, storage companies show a high level of “vertical integration” – that is, integration between storage products in their own line-ups – and not enough “horizontal integration,” or cross vendor interoperability.

As a result, customers tend to get locked into their vendors. Moreover, within some vendors’ management suites, quality can vary markedly from one tool to the next.

In the future, though, storage software will become more “commoditized,” according to Patty Barkley, storage networking marketing manager at CNT. Vendors will also start to release “predictive and automated” storage tools.

The speakers also harkened to future adoption of industry standards in areas ranging from fabric switching to plug-and-play disk drives.

Some SSPs aren’t doing so hot
Contracting with an outside SSP is another approach to handling data. “But a lot of SSPs aren’t doing so well these days,” Berman acknowledged. “People really want to keep their own data inhouse.”

When companies do decide to take on outside partners, they’re generally more amenable to consultants who will work on-site, he added.

Will tape wind down one day?
“Does tape have a future?” asked a user attending one of the sessions. “Yes,” answered Mark O’Malley, Quantum’s senior manager for business planning. “Disk storage doesn’t give you as much protection against data loss in case of disaster.”

Quantum plans to start doubling the capacity of its tape storage products every six months, according to O’Malley. He admitted, though, that tape will probably wind down at some point, citing emerging technologies such as holographic storage and solid state chips.

Several attendees raised questions around choosing between NAS and SANs for disk storage. The decision should revolve around the application, together with cost and storage management considerations, according to the speakers..

As Domme sees it, NAS is appropriate for storing distributed Web files, for instance. “Yahoo runs on a series of NAS boxes. NAS is cheap to put in, and you can get it going within 30 minutes,” he added.

“The jury is still out,” though, on whether it’s wiser to install separate NAS boxes, or to install NAS software on top of a SAN.

Don’t use NAS with databases
Typically, NAS runs over Ethernet, whereas SANs operate over fiber channel, Berman observed. SANs are particularly well suited to transactional database applications.

“NAS doesn’t work well with Oracle or DB2,” he illustrated. On the other hand, SAN still faces scalability constraints, even when FC is routed over IP, since FC relies on only a single routing protocol.

Lately, some industry analysts have noted a trend toward the use of SANs for enterprise storage consolidation. Berman, though, dismissed this theory as “vendor hype.”

Tiny SANs
To the contrary, most SANs running today are small in scope, according to Berman. As evidence, Berman cited a study by Enterprise Management Associates, another industry analyst firm, showing that today’s “average SAN” consists of only 5.2 servers.

Many corporate department and small business are completely unaware that they’re already running SANs, according to Berman. Vendors such as Compaq are now embedding SAN technology into their RAID racks.

Berman also predicted that future industry standardization will go hand-in-hand with “Cisco-ization” of SAN fabric switching. He dubbed this movement “the Cisco effect.”

Cisco recently decided to take spin-off outfit Andiamo back inhouse, thereby throwing a jumbo-sized hat into the ring against current industry leaders Brocade and McData.

‘Muster the IT guy’
Berman doled out other tips, too. Corporate decisions to add more storage capacity are often driven by “business strategy and goals.”

When these decisions reach the IT department, IT will often retort, “But we don’t have the infrastructure to do this.” Instead, organizations should “get the IT guy mustered right up front.”

For their own part, IT staff should “start small, and then extend at a reasonable pace,” Berman advised.

“Don’t touch a falling knife,” he added. “Be careful about how much you invest in storage technologies that look like they might be on their way out.”

This article originally appeared in Crossnodes.

Jacqueline Emigh
Jacqueline Emigh
Jacqueline Emigh (pronounced "Amy") is a veteran tech journalist, with literally thousands of published articles to her credit. She is currently a columnist for Internet.com, as well as a frequent writer for a number of other leading magazines and Web sites. She was previously a senior editor at Ziff Davis Media's Sm@rt Partner Magazine. Before that, Jacqueline spent seven years as a bureau chief for Newsbytes News Network, a daily newswire specializing in computer technology. Her work has been translated into dozens of languages.

Get the Free Newsletter!

Subscribe to Cloud Insider for top news, trends, and analysis.

Latest Articles

15 Software Defined Storage Best Practices

Software Defined Storage (SDS) enables the use of commodity storage hardware. Learn 15 best practices for SDS implementation.

What is Fibre Channel over Ethernet (FCoE)?

Fibre Channel Over Ethernet (FCoE) is the encapsulation and transmission of Fibre Channel (FC) frames over enhanced Ethernet networks, combining the advantages of Ethernet...

9 Types of Computer Memory Defined (With Use Cases)

Computer memory is a term for all of the types of data storage technology that a computer may use. Learn more about the X types of computer memory.