IBM is a major force in computer research and development in the U.S. The company is always working on next generation computing systems and components. It also has several machines that it has designed, built, and installed on the list of the fastest super computers in the world. IBM has an unnamed client that needs a new supercomputer with a record-breaking amount of storage.
The storage system that the team at IBM has constructed is many times the size of most existing storage systems – its capacity is a whopping 120 petabytes. That is 120 million gigabytes of storage. Technology Review reports that the storage depository would be large enough to hold about a trillion files. To reach that capacity, the storage system uses 200,000 conventional HDDs that are working together.
More storage is required for super computers that are working on very complex simulations and in industries like weather and climate modeling as well as the petroleum and genetic research industries. The most staggering factoid about the 120 petabyte storage system is that it could hold 60 copies of the largest file backup system on the web, the 150 billion page WayBack Machine the Internet Archive runs.
"This 120 petabyte system is on the lunatic fringe now, but in a few years it may be that all cloud computing systems are like it," Bruce Hillsberg, director of storage research at IBM, says. Just keeping track of the names, types, and other attributes of the files stored in the system will consume around two petabytes of its capacity."
Many issues have to be overcome with such a large storage system. Things like efficiently combining the 200,000 physical drives and allowing the supercomputer to continue operating when one or more drives are down are among the issues that must be addressed. IBM uses a system where when a disc breaks, the data is written back to its replacement slowly so the supercomputer continues operating normally. Hillsberg claims that the result is that the machine shouldn't lose any data for a million years and with no performance compromises.
Combing this massive storage system with the Neurosynaptic chip that IBM is working on would result in a very impressive machine.
