Despite the advances in mobile technology over the years, users have been forced to rely on bigger and bigger storage media to combat their data growth problems.
Perhaps, instead of focusing on larger storage, vendors really just need to create smarter storage.
At an IBM dinner a couple of weeks ago, Finis Conner, who is currently chairman and CEO of StorCard Inc., suggested that personal and portable computing devices could benefit largely from a personal HSM (hierarchical storage management) solution.
Conner, who is familiar to storage-industry followers for his experience in designing, manufacturing and selling hard disk drives, co-founded Shugart Associates in 1973. The company, sold to Xerox Corp. in 1977, made and sold 8-inch floppy disks and HDD (hard disk drives).
In 1979, Conner co-founded Seagate Technology LLC, where he conceived the original 5 ¼-inch Winchester disk drive. In 1986, he founded Conner Peripherals Inc., which was the first company to develop and manufacture 3 ½-inch and 2 ½-inch Winchester disk drives used in personal computers.
Under his direction, Conner Peripherals grew to $1.4 billion in its fourth year of sales and $2.7 billion in its 10th year, when Seagate acquired the company.
But Conners new company, StorCard Inc., is turning its attention toward portable solutions. Its new, portable storage product is the size of a credit card, and a single StorCard can hold 100MB to 5GB of data.
Conner said he believes that by seamlessly migrating older, unused data from portable devices onto servers or PCs, users will be able to more efficiently utilize the small onboard storage on their portable devices.
In place of the data, a personal HSM could leave intelligent file pointers that quickly locate migrated copies of data and could be residing on an external system such as a PC at home or a file server at work.
Considering the growth these days of high-speed wireless Internet access, intelligent personal HSM solutions would be able to automatically contact a home PC or server to locate and quickly download a previously migrated file.
Like any other HSM, a personal HSM would benefit users since it would clean up expensive, highly limited portable storage and transfer data that is unlikely to be accessed to an external computer.
But such a solution does face some challenges. First, while HSM has been a mainstay in mainframe environments for decades, only now—with the monstrous marketing push of ILM (information lifecycle management) from storage vendors—has HSM really become fashionable in smaller open system environments.
Much like any other storage management solution on the market today, complexity is the biggest obstacle.
One aspect of HSM that has always been difficult is the creation and optimization of migration policies.
If you make a migration policy too aggressive, you will wind up wasting valuable time sending requests out to the HSM server and retrieving data from nearline storage.
While a megabyte or two would be OK to retrieve on a broadband connection with a wireless device, doing retrieval operations for dozens of files could be more than a little painful, especially if you are with clients or late for a meeting.
On the flip side, creating policies that are too laid-back makes the HSM system inefficient, which may force clients to buy additional portable storage to satisfy their needs.
Another practical thing to consider is that a user would have to make sure his or her HSM server is always on and always connected to the Internet—two factors that could lead to security headaches.
While the personal HSM may not be a solution we will see in the near future, Im sure that technologies will be developed to help mobile users optimize the storage of their devices.