SINGAPORE -- The main problem with large storage systems is not a lack of available capacity but the need to get at that data quickly and simply, according to Richard Blaschke, executive vice president of marketing at Xiotech Corp.

As demand for storage capacity grows by over 75 percent each year, the technique known as storage virtualization will become mandatory for end users, Blaschke said during a presentation at Diskcon Asia here Thursday.

Storage virtualization means dividing the available storage space into "virtual volumes" without regard to the physical layout or topology of the actual storage elements such as disk drives, RAID (redundant array of independent disks) subsystems and so on.

From users' perspective, the storage pool is a reservoir from which they may request any amount of disk space, up to some specified maximum. The goal of the intervening software and hardware layers is to manage the disjointed disk space so it looks and behaves like a single attached disk.

This approach has several benefits for end users, according to Blaschke. These include:

-- increased ease of use, reduction of capacity planning issues and reduced downtime

-- better performance by harnessing the power of all the disk drives as data can be "striped" across multiple drives

-- efficient use of storage capacity

-- ability to run multiple operating systems seamlessly

"When you're scaling to 200 terabytes, 24 by seven, across the world, and on many disk systems, you have to use virtualization," he said. "It also enables people to say, 'I don't know and I don't care what physical device the data is on and what operating system is running.'"

According to Blaschke, storage virtualization is an enabling technology which sits above architectures such as NAS (Network Attached Storage) and SAN (Storage Area Networks).

"All these approaches are converging and everyone is trying to get into everyone else's business," he said.

One virtualization technique proposed for SANs by Israeli company StoreAge Networking Technologies Ltd. is known as SAN Volume Manager (SVM). This uses a combination of a metadata center and volume drivers for creating and managing virtual volumes while enabling direct data transfer between server and storage subsystems.

By having multiple storage subsystems working in parallel with multiple servers, total performance is increased up to the maximum of the fiber channel fabric bandwidth, according to StoreAge.

SVM separates the handling of metadata from the data path. It consists of a metadata center that "sees" the physical storage and allocates virtual volumes, and a volume driver at each of the servers on the SAN.

The volume driver retrieves the volume configuration from the metadata center and presents virtual volumes to the operating system as if they were disk drives. When the operating system sends an I/O to the virtual volume, the volume driver intercepts the I/O, translates the volume address to the physical address, and sends the I/Os directly to the storage subsystems.

According to Blaschke, initiatives like this are becoming increasingly more important, as storage investments will represent between 65 percent and 80 percent of total hardware and software systems costs by 2003, and will be a market worth US$86 billion by that time.

ITworld.com

• Email
• Print
• Share
  • Slashdot
  • Digg
  • Stumble
  • Reddit
  • Newsvine