In most enterprises, disk is looked to as a cure for all your backup woes. The goal of adding disk to the process is to reduce backup windows, improve recovery windows and make the whole process more reliable. Reality is though that while it does help the situation it seldom cures the backup problem and fails to fully address any of these problems.There are conflicting needs within the backup process that has lead to a proliferation of disk backup devices in many enterprises. There is one of rapid initial copy time vs. one of long term storage need. This in many environments has lead to not one but multiple disk to disk backup solutions being deployed. A high speed cache for quick ingest, a medium term storage repository for file and system recoveries and a long term archive pool for retention. Increasingly data centers have two or three of each of these options as they try to decide which one is best or need to address scaling limitations which leads to buying multiple units. Each has to be independently managed and monitored. This is an area where backup virtualization can help as we discuss in our recent article "Backup Virtualization Brings Flexibility To Disk Backup" by consolidating the various disk backup targets behind a single backup virtualization appliance.
Consolidating the various disk and tape resources is merely step one and brings a powerful band-aide to the problem. The next step is to just get better control of the protection process. This can be done in one of two ways. Fundamentally changing how the backup process works or insert specific policies and procedures to better manage and maintain service levels within the current process.
Fundamental change is going to have to come from the current move away from traditional file by file copy method and more to the disk based image level backups. While these have become especially popular in virtualized environments, stand alone systems are now being included either by temporarily treating that stand alone system as a virtualized server or by direct block level backup support of the stand alone system. In a way, these systems can be thought of as the ultimate use of disk because they depend on a device that is random access in nature. This does not mean tape is dead though. Several of these solutions are supporting tape, or you can again use backup virtualization to perform that integration.
Better control is going to come from the increased use of tools that can provide a realtime view into the goings on of the backup process. These solutions allow for service levels to be set and maintained as well as data to be managed. Gaining control of the process and the data that process is designed to protect is really the long term solution.
Disk backups can help and may be the most viable quick fix to a backup problem. Looking further down the road though by consolidating backup targets and instituting backup SLAs and processes may have a greater return on investment.
Track us on Twitter: http://twitter.com/storageswiss
Subscribe to our RSS feed.
George Crump is lead analyst of Storage Switzerland, an IT analyst firm focused on the storage and virtualization segments. Find Storage Switzerland's disclosure statement here.