The company's Collective Intelligent Bricks project builds variously sized three-dimensional stacks out of the eight-inch modules, each filled with 12 hard drives and six network connections to keep data coursing through the collection. IBM envisions a day when hundreds of these storage "bricks" are stacked together, eventually with computing bricks in the same assemblage.
By the first quarter of 2003, IBM hopes to have built a three-by-three-by-three-brick prototype with a total of 32 terabytes of storage, said Jai Menon, an IBM fellow and storage research manager at Big Blue's Almaden Research Center in California.
A three-by-three-by-three stack would have 27 bricks, and a six-by-six-by-six collection would have 216. With a collection of eight bricks on a side, there would be 512 bricks--that is, 6,144 hard drives--in a system only a little wider than an adult's arm span.
"This is almost like a revolution in packaging. We're taking advantage of the third dimension," said Almaden researcher Robert Garner. Garner and Menon showed off prototypes of the system Friday.
Even though bricks slide in and out of the system easily, IBM imagines people will just buy a few extra bricks and leave defunct bricks in place. Software keeps track of where information is stored and ensures that it will still be available despite any failures.
Those features are the real reason IBM is pushing the design. Storage systems are expensive to manage, and Big Blue believes the resilient design will permit a single administrator to manage about 100 times as much storage capacity.
The brick storage system, devised in a project formerly code-named IceCube, is an extension of the "" trend sweeping the computing industry. The vision is that customers will stack simple modular systems piece by piece as demand grows instead of buying monolithic systems.
At present, though, these modular systems can't match the performance of the single large-system approach. The kings of the hill in high-end storage are million-dollar machines such as EMC's Symmetrix and Hitachi Data Systems' Lightning.
Storage is a very competitive market today, with EMC leaning on anwith Dell Computer to spread its wares more widely even as rivals Hewlett-Packard and Sun Microsystems invest in new designs. With customers better able to dictate purchase terms, storage companies are searching for improvements they'll be able to charge for.
New tricks for bricks
IBM is working on several advances to try to make the bricks surpass current monolithic designs.
First off, because the systems are connected on all six sides to their neighbors and must easily slide in and out, they can't be connected using ordinary plugs and sockets. IBM is using a technique Garner developed called "capacitive couplers," small pads that can send signals from one brick to another with which it's in contact.
A key part of the brick concept is data protection. As with current RAID (redundant array of inexpensive disks) technology, information such as a database file is stored on several disks so it won't be lost if a single disk fails. The brick idea takes that concept one step farther, with the bricks automatically shuffling data from one brick to another to compensate for problems such as failed drives, bricks or networking elements.
And of course the bricks suffer an extreme version of a long-standing issue with high-end computers: overheating. In current prototypes, the bricks are impaled on vertical pipes that hold a water-cooling system. Interior bricks, with no air circulation at all, would otherwise quickly suffer debilitating temperatures.
"I believe the industry is moving toward liquid cooling," Garner said.
Water cooling has its problems, though, including expense, maintenance difficulties, and a required connection to a specialized external system to cool the circulating water back down. IBM itself has switched its top-end mainframe servers from water cooling to air cooling.
More challenges arrive when building a three-dimensional lattice of networked nodes--each storing its own data along with routing information to find other data. IBM has been researching issues such as how many units can fail before the spread of data through the system is impaired.
The prototype brick system connects each brick to the other using Ethernet networks with a data transmission speed of one gigabit per second, but that version of the technology won't work for actual products because transmission delays, or latencies, are too long, Garner said.
"Latencies are one of the issues with gigabit Ethernet," Garner said. Eventually, IBM expects to use either or 10gbps Ethernet.