
DAN ("DINO Analyses for Networker") is a state-of-the-art, automated 'health' checking tool for Networker. It replaces slow, cumbersome and often unreliable manual inspections that are normally undertaken by consultants using checklists. Now it's possible to automatically check various configurations and be prompted with intelligent suggestions for things such as changing the multiplexing value, adding additional tape pools or changing the backups to go to disk. Hundreds of hours of manual diagnosis are effectively reduced to minutes.
New rules for a changing world
A backup in the mainframe world is usually a simple matter of one disk going to one tape . . but its very different in the open systems world. In this complex environment products typically send data over the LAN to the backup server. The server owns the tapes and writes the data out. However, since the disks typically do not occupy full tapes . . and because the data coming from the client computers don't always supply the tapes with enough data to keep the tapes moving . . the backup products often allow several different versions of multiplexing in order to keep enough data in the pipeline to keep the tapes moving. Multiple bursts of data may be coming from each client computer and several client computers can be supplying the backup server. Data on the tape may be multiple files from multiple computers, and they are ALL INTERSPERSED on the tapes. They have Full, Incremental, Differential, and Consolidated backups so that a full backup taken on Saturday cannot be released until all dependent backups ("subsequent incrementals") have expired. A simple thing like changing the expiration date of a backup can prevent an entire chain of backups from being released.
Complex environment, complex problems
Users can also backup to disk, stop and start applications before and after backup, compress files, perform archiving, allow for tape pool overlap, include additional data (like including 1 file from a directory other than the 1 being backed up), set up storage nodes so that additional servers can own tape drives but the database is maintained by a central node, allow NDMP device to go directly to tape, take snapshots before backup, backup the disk backup to tape at high watermark levels(called staging)etc, so it can be a very complicated environment that can get out of hand quickly. DAN gives clear understanding for optimum utilization of current assets
DAN gives users the ability to: