IBM Adds Muscle to Storage Virtualization

Big Blue enhances its System Storage SAN Volume Controller to help companies move large amounts of data over greater distances at higher speeds and across a number of different platforms.

IBM introduced next-generation enhancements to its storage "virtualization engine" software May 25—improvements that help companies move large amounts of data over greater distances at higher speeds and across a number of different platforms.

Storage virtualization is a logical view and control of physical storage systems. Enterprise-class virtualization offers high levels of redundancy and advanced data management features, including snapshots and remote mirroring.

System Storage SAN Volume Controller 4.1, the new software Big Blue rolled out, includes a number of advancements, including one designed to help an organization "virtualize" its entire infrastructure more efficiently, the company said.

"In the wake of regional disasters, such as Hurricane Katrina, businesses are increasingly seeking ways to support continuing IT operations by placing data centers that share resources at greater distances than in the past—including in other cities, other states, even other countries," said IBM Storage Volume Controller Marketing Manager Chris Saul.

"The new global mirroring function in the SAN Volume Controller is designed to help customers virtualize data at locations of greater than 100 miles faster and more cost-effectively than previously."

Data—especially massive amounts of data created by large corporations—encounters limitations as it uses bandwidth to move through wires or air, and distance plays a major factor in how efficient such transfers and replications are.

"For example, think of the way television signals are delayed slightly when they are routed through a satellite from one part of the globe to another," Saul said. "SAN Volume Controller—a pizza-box appliance attached directly to the SAN—keeps the flow of data smooth."

SVC Version 4.1 features Long-Distance Global Mirror, which provides long-distance asynchronous remote replication for business continuity and disaster recovery at nearly unlimited distances, Saul said.

In addition to the new business continuity enhancements, SVC 4.1 now supports 4G-bps environments and has extended its capabilities to virtualize data on nearly 80 different disk systems, Saul said.

What are the key advantages to a business in virtualizing stored data?

"The advantages to anyone—SMBs [small and midsize businesses] and large companies—is that you get a pool of storage assets versus discrete islands," said Tony Asaro, senior analyst at Enterprise Strategy Group Lab in Milford, Mass.

"You can move data transparently between different storage tiers online and transparently. This allows you to move dormant data off of Tier 1 storage to lower tiers to save on cost and to create a chain of efficiency. It gives you better performance and reduces management complexity."

/zimages/6/28571.gifHP offers virtual tape offerings for SMBs. Click here to read more.

Companies also can protect data by using common tools for heterogeneous storage, Asaro said. This enables customers to create remote replication relationships between different types of storage systems, which can greatly reduce cost.

Dianne McAdam, director of enterprise information assurance for The Clipper Group in Wellesley, Mass., said that virtualization gives a company more options in how to transport or store its data.

"Many of the [storage virtualization] products, such as Invista and [IBMs] SVC, will copy volumes of data from one array to another," McAdam said. "So, if I put an older EMC array and a newer HDS array behind an SVC, for example, I can use the SVC to copy the data from the older array to the newer array [and later uncouple the older array]. So they can use these products to migrate from one array to another."

Or if one array is higher-performing than another and the application needs better response time, the IT manager can move the data from the slower array to the faster array to improve performance, McAdam said.

IBM in May announced it has surpassed 2,000 SVC customers who are virtualizing more than 15 petabytes of data. IDC recently reported that the installed base of appliance-based, virtualized networked storage capacity was 28.1 petabytes at the end of 2005. Based on this information, IBM said it believes that SVC manages more than half of the appliance-based virtualized storage capacity in the market today.

"Two thousand SVC customers is impressive," McAdam said. "EMC has not been the first to market with a virtualization appliance, so they have some catching up to do. Every major vendor will have some kind of product like this—if they dont have one available already."

Asaro said he believes "storage virtualization [ultimately] will be requisite. Its just a smarter way to manage your storage. Right now, we correlate virtualization with products. But in the end, storage virtualization is just disaggregating the storage controller from the disk enclosures. The intelligence and value is in the storage controller."

/zimages/6/28571.gifCheck out eWEEK.coms for the latest news, reviews and analysis on enterprise and small business storage hardware and software.

Chris Preimesberger

Chris J. Preimesberger

Chris J. Preimesberger is Editor-in-Chief of eWEEK and responsible for all the publication's coverage. In his 13 years and more than 4,000 articles at eWEEK, he has distinguished himself in reporting...