Do I need a specific type of disk subsystem to implement disk-to-disk backups?

Do I need a specific type of disk subsystem to implement disk-to-disk backups?

    Requires Free Membership to View

    Login

With the quickly growing popularity of disk-to-disk backups, this question is raised quite frequently. The primary distinction that needs to be made is to determine whether you want to backup-to-disk or replicate-to-disk. Replication is typically implemented to take advantage of specific key features such as rapid access to an online copy of the data should the original become unavailable. This typically requires a better performance, higher end disk subsystem.

True disk-to-disk backups are usually controlled by backup software to create what is typically referred to as "near line" copies of the data. This also means that this copy must be written back to a specific location (restored) in order to be accessed. In addition, a solid backup scheme should typically include the creation of a (tape) copy of the backup data to be stored offsite. For these reasons, the availability and performance of the disk array for backups becomes secondary. One of the appealing aspects of disk-to-disk backups is that they offer better overall performance than tape backups but at a lower cost than data replication.

Therefore, pricing is probably one of the main evaluation criteria followed by scalability when considering the selection of disk storage to use for disk-to-disk backups. These last points are at the base of the growing popularity of ATA and SATA disk technologies for disk-to-disk backups.

This was first published in July 2004

Join the conversationComment

Share
Comments

    Results

    Contribute to the conversation

    All fields are required. Comments will appear at the bottom of the article.
    Back to top