Comprehensive & Scalable Disaster Recovery Solutions

If a disaster strikes your IT infrastructure, destroying data and/or computer hardware like servers and storage, at a minimum you need to be able to recover from a copy of your data stored at a remote location. The "poor man" approach to disaster recovery has been to ship physical tapes containing backup data to a remote site, but as data growth rates have skyrocketed and recovery requirements become more stringent, tape-based DR has not been able to keep up. Distance is a key consideration in setting up a DR plan: certain disasters like hurricanes or tornadoes can have widespread impact, and your remote site needs to be far enough away from your primary site so that a single catastrophic event cannot destroy both sites. Industry best practices dictate that the remote site should be at least 200 miles away from the primary site.

There are two levels to a DR plan: data recovery and application recovery. Data is useless if there are no applications to use it. Your DR plan must provide for both. Tape-based DR covered only data recovery, while recovering applications was a manual process left to administrators to perform in real time in the event of a disaster. Manual recoveries require sophisticated administrative expertise to rebuild servers appropriately and result in lengthy recovery times that are fraught with risk.

If you've attempted to perform a disaster recovery from remote tapes, then you are likely to have run into at least one of these problems:

  • Surveys suggest that for a given data set that must be retrieved from remote tapes, that data is not recoverable 15% to 20% of the time; this may be due to a number of factors, such as lost tapes, tapes that contain incomplete backups, or tape media integrity problems, among others
  • The most recent data on remote tapes may be 3 to 4 days old or older, resulting in significant data loss on recovery; re-creating this lost data by hand can take months if it can be done at all
  • Recovery times using tapes can be several days to a week or more; consider that to recover from tapes, they need to be found (among the many tapes that you may be storing at the remote site), shipped to a recovery site (an operation that can take 24 to 48 hours or more), restored to disk, and validated (to ensure that all the recovered data is usable); servers on which the recovered data will be loaded must be appropriately configured prior to restoring data to them, a process that requires significant administrative expertise and can be very lengthy and manually intensive if systems need to be brought up to date with new configuration parameters and various patches so that they match the configurations of the servers that they are replacing

InMage Scout provides a better way, offering options for both data and full application recovery that can be fully automated. Here is how InMage Scout is better:

  • As new data is created, it is continuously transferred across a local LAN to the InMage CX, which immediately replicates it to one or more target locations; this approach gets new data to remote locations within seconds or minutes of its creation, supporting much more stringent RPO requirements that minimize data loss and the business impact of recovery events
  • InMage Scout supports asynchronous replication over cost-effective IP networks so that it can support short and long distance (more than 200 miles) DR configurations
  • InMage Scout employs a number of storage capacity optimization technologies that minimize the amount of data that has to be sent to target locations to enable very granular recovery operations; the bottom line is that it provides the most WAN-efficient comprehensive DR solution on the market today
  • All data recoveries come directly from disk, allowing InMage Scout to recover data even from remote locations within minutes; this shortens recovery time from potentially weeks (using tapes) to minutes
  • InMage Scout can be configured to support "push button" application failover/failback, an automated process that recovers application services on new servers using any previous recovery point an administrator wants to select; the combination of CDP, asynchronous replication, and application failover/failback means that, in response to a catastrophe, you can restore application services at a remote location from current data and without having to rebuild servers manually, literally within minutes