Computerworld

Original ideas still in archiving

Archiving may be an old application, but innovation in the field has not finished yet. The breakthrough in gen-i's archiving application, on which it has a patent pending, is that it does not need a server, says developer Alan Moore.

Archiving may be an old application, but innovation in the field has not finished yet.

The breakthrough in gen-i’s archiving application, on which it has a patent pending, is that it does not need a server, says developer Alan Moore.

“Generally, in the past, archiving has been done through a server, which controls the file system being archived,” he says. “With ours, the user can just point the program at any network share and archive the files directly. The owner of the file instigates the archiving and restoration, and the program runs in that user’s own PC.”

When a file is archived, it leaves a shortcut on the owner’s screen. Clicking on that shortcut will recover the file when needed, says Moore.

When the program was first discussed, in the context of gen-i’s attempt to patent the process (see Own wares stiffen gen-i support for patents), it immediately sparked letters saying the process was not new. But all the archiving procedures described in these letters as having been in place as long as 15 years ago use a server or a mainframe, Moore says.

Most “hierarchical storage management” systems, he says, have a mainframe ancestry and have never got rid of the perceived need for a central machine controlling everything, he says. “We come from the opposite direction, from the desktop and control by the user.”

The current version of the program is written for Windows NT, but versions for other operating systems, both Microsoft and not, are planned, he says.

Meanwhile, US-based Princeton Softech also claims to have enhanced the archiving and recovery process, but in a different direction. Its “active archiving” tool can prune the infrequently used data in a relational database, claims Princeton Softech senior consultant Bruce Fischer, but preserve all the links between items so that the little-used rows from a table and the rows in other tables associated with those can be stored away together and recovered together.

An insurance company might, for example, say “archive everything related to this incident, because the work on it has been finished”, Fischer says. With the unneeded data out of the way, database retrieval becomes much quicker, he says.

US company National Steel archived 70% of its database, that had been sitting on high-speed disks and rarely or never accessed. With the space and cost saved and the productivity improvement resulting, the company was able to implement a data warehouse without needing to purchase any extra hardware, Fischer says.

Wellington’s ITB solutions has the local agency for the Princeton product, and has found its first buyer in BNZ bank’s insurance subsidiary.