by Ian Masters
When it comes to backing up data, tape has been the preferred option for many years. At the end of each day, some poor soul has been responsible for managing the transfer of data from production servers to tape cartridges, which then have to be transported off-site, just in case company IT systems suffer a fatal issue the next day. As the employee watches the spool of tape revolve to its conclusion, the thought must at some point enter his or her head: “There has to be a better way.”
What IT managers want
IT managers are responsible for delivering services to users efficiently and cost-effectively. Having a business continuity plan that works has a significant part to play in fulfilling these fundamental responsibilities, as any IT manager who has presided over periods of lengthy downtime will no doubt appreciate.
A good business continuity plan needs to meet several objectives:
-It needs to guarantee that, following a disaster, data is not lost and is recoverable (preferably quickly);
-It must provide multiple recovery options, as no two disasters are alike;
-It should be fully automated, as the last thing anyone needs when the building is on fire is to have to run around pressing buttons;
-It must comprise processes that don’t need to be constantly monitored or scheduled.
Finally, given modern working practices and technologies, it needs to have the flexibility to quickly scale and be able to recover to either physical or virtual services without issue.
What tape cannot give
Even if a company has been making regular tape back-ups for years and as a result has a library full of media, the IT manager may be plagued by doubts. Will the data have written correctly? Will the media be okay after a long time? What if there is a disaster and all the company’s data needs to be recovered? How long will it take? Will it work? How much data will be lost? How much trouble will I be in if it does not go to plan? The multitude of possible disaster and recovery scenarios asks a lot of an IT manager’s faith in tape, for good reason.
Drawbacks of tape
Many people will remember the music and film industry’s foray into tape media. Who can forget the hours spent sitting next to the stereo sellotaping snapped media together, loading puddles of tape back into the cartridge with a biro, the continuous hiss of the terrible sound reproduction and the way that even that quality flagged over time? The entertainment sector could not get out of tape quickly enough, with first a move to optical media and latterly ‘the cloud’. All those problems apply to tape designed for data storage.
Tape media and tape back-up processes do not provide the back-up options required by 21st century business. The minimum amount of time to recover from tapes that are stored remotely needs to be calculated in days. Why would a company invest time and money to make redundant copies of data and then not be able to have it at their fingertips to use at any time? IT managers need more options for effective data and server recovery than tape can provide.
What else is there?
There are several technologies more cost-effective than tape that offer near zero data loss and recovery time. Data replication and operational recovery technologies, for instance, not only provide real-time replication of data changes, they are also able to save several hundred if not thousands of versions of those files to be recovered when needed. Restoration of data and systems, to any point in time, is automatic and seamless.
If you are looking to replace tape then you are likely to consider a centralised back-up design or site consolidation using some form of data replication. This strategy will at least reduce, and possibly eradicate, tape used within remote offices by replicating data back to a centralised location for ease of management.
The central site can then be used to provide disaster recovery for remote locations as they will have the ability to restore their servers either locally or, depending on the level of the disaster, at the central location. All of which will be quicker and better than waiting for the Iron Mountain van to arrive. The lower cost of disk storage and efficiency of virtualisation technologies enables most companies to take advantage of the basic and more advanced data recovery and management practices.
With data replication and continuous data protection functionality developing rapidly, the ability to use back-ups for more than just recovery is becoming the norm.
For example, users can be empowered to find files they have deleted or recover their own systems with little involvement from IT support. Full server protection options safeguard the entire system state of applications and servers, providing data centre managers with the ability to perform maintenance on those servers without extensive downtime.
It is probably fair to say that most IT managers will not fully know what service pack their servers are operating with or have full documentation on how the vendor configured their mission-critical applications. With full server protection it is less of a concern because the solution will capture all settings and any changes made to the production server and apply them to the back-up systems.
If an IT manager can recover an exact duplicate of a server from the OS up and not have to install all the applications with the exact configuration and all the appropriate service packs, recovery will happen in a blink of an eye, in less time than it takes to say, “who knows where the tape is?”.
Add in continuous data protection and you not only have the ability to recover an entire server, you have the ability to recover versions of files that were changed. What if you get a virus that infects your e-mail? It would be nice to have rewind capabilities to scroll through your versions of files to find the exact point in time before that infection occurred so it could be prevented.
Proof trumps belief
It is not enough to believe data is backed up and available – IT managers have to know. If a company’s IT system suffers a disaster, and tape is the primary back-up and recovery solution, any faith in its ability to help get systems back up and running quickly, will soon receive a severe blow. Data replication technology has usurped the old way of doing things by rapidly proving its ability to help IT managers meet their responsibilities and deploy innovative management strategies.