By David Kramer – President, Domain Technology Group
This article is targeted to the technically savvy, but not necessarily to the very technical.
The message is targeted to owners and managers of small to mid size enterprise (SME).
As we all know disasters do happen. One happened to George Mitten, owner of the Windsor Press, when his business caught on fire. During the fire he asked a Hamburg fireman for an air pack. He wanted to go back into his burning business to retrieve his primary computer, the one with critical business information on it (as reported in the Reading Eagle: on March 24, 2009). With disaster recovery planning (DRP) that request would have been unnecessary.
The level of business disruption caused by a disaster varies, but all these disruptions are distracting, and costly.
The Current State of the Art Business Continuity Planning (BCP)
The cost of disaster recovery planning (DRP) has dropped to the point where all small and mid size businesses (SMEs) should investigate and implement some form of DRP. The actual level of cost to recover from a disaster is dependent upon the degree of planning and preparation put into the DRP.
Simple onsite tape backup, even where “a person” takes a tape offsite nightly is no longer sufficient. Much better solutions are necessary and available. If a disaster occurs and you are relying on tape backup for restoring your operations, the engineers performing the restoration must first locate equipment and media compatible with your tapes, before even attempting to restore your applications. This can be a very time consuming process, and you do not have time to consume. The older your existing business systems are the more of a challenge it will be to locate replacements. Then there is always the question, can the tapes be read? The tapes themselves are not a failsafe. And if a complete backup run extends beyond a single physical tape, the chance of tape failure is increased.
At this point, SMEs should be looking beyond implementing simple tape backup systems.
The current state of the art DRP is to implement an onsite backup device called an appliance, to which all data is copied at predefined increments. In advanced implementations this appliance can take over for one or many failed server(s). In other words, the appliance becomes a server utilizing the systems and data that has been copied to it. The time to configure the appliance to become a server is about 30 minutes. The number of failed servers that an appliance can proxy for is dependent upon the level of device implemented. Currently there are appliances that can proxy for up to ten servers.
Server Failure
In the case of failed server(s) only, your company is back up and running in less than one hour. The appliance will act as the server while new equipment is ordered and then configured and put back on line. Compare this scenario to restoring from tape. The downtime and recovery itself is significantly more, usually one or more days.
Loss of Building
There is a second level to this backup solution for when there is a disaster, such as a fire, causing loss of facility; when either the building is gone, or can not be re-entered. In this recovery plan the backup data is copied to an appliance, after which it is transferred via the Internet in an encrypted, secured format to an industrial strength data center or centers (in some cases, even that is redundant). If data recovery is then needed the data center engineers will copy the data to media and over-night it to your new location where an IT partner can then perform system restoration. The trained partner performs what is termed a “bare metal restore”. This means a simple restore of the operating system, the applications, and data to a new file server. Bare metal restores are not as sensitive to the equipment being restored to. You can usually replace current hardware from distribution and restore directly to it. The new computer does not need to be configured exactly like the failed server as in a tape recovery scenario. A bare metal restore is a faster safer recovery process.
Disasters Do Happen
Within the last six months there were two disasters within 50 miles of Reading that affected a company’s critical business information. One company, a client of ours, took a lightning strike. That company had a DRP in place and their information was restored quickly. The other company lost their building and business systems to fire. This company did not have a DRP and had to go through extraordinary measures to retrieve their computer systems.
Company One, Unprepared.
On March 24th there was a fire in Hamburg, PA. A long established business, The Windsor Press, located at Third and State streets sustained a disaster. As stated in the beginning of this article the owner of the business, George Mitten, wanted to go into a burning building to retrieve a business critical computer. Eventually this computer was retrieved. The Press got lucky. What would have happened if the system was not retrieved? How long would it have taken for them to recover their business data.
Company Two, Prepared.
The second case is one of our clients, a business near Philadelphia. Last summer, a bolt of lightning struck a utility pole outside their building. An electric surge travelled into their building, through their uninterruptible power supply, and fried one of their primary servers. This company had implemented the DRP solution described above. The client was up and running quickly with very little downtime. The backup appliance was configured to replace the failed server with little fanfare. And that is exactly how a backup plan and restoration should work. It is the lack of event that makes this example so good.
Investment
The investment in equipment and services required to implement an industrial strength IT disaster planning solution has dropped to the point where all small and midsize enterprises (SME) that have not implemented a DRP should seriously consider it. The investment depends upon the size and complexity of your business and systems. The initial purchase may range from $1,500 to $15,000. There is a recurring monthly fee based upon the amount of data transferred and stored at the offsite data center. Compare the investment of DRP to the cost of down time, or a sustained complete loss of systems. And note: many companies that lose their total business systems never completely recover and eventually fail.
Conclusion
Because the entry point for sophisticated, pretty much fail safe, disaster recovery planning has dropped dramatically EVERY small to medium size enterprise (SME) should investigate it, and give it serious consideration.
Even during a challenging economic climate, planning and preparing for disaster is key.
There is no worse time to lose your systems then when you are in a tight survival mode.
Working without your normal systems could be the straw that breaks the camel’s back. Just think of when your email does not work, how quickly that throws you into a handicapped mode of operation. Losing all of your systems is business drama beyond compare. As we saw in the two examples above, disasters do happen, and not that infrequently, and they happen locally. Some businesses are prepared and some are not. Which are you?
DRP solutions that include an on premise backup appliance, which can also serve as a replacement file server, coupled with offsite storage at an industrial strength backup data center is key. The backup appliance takes over for a failed server or servers when your facility is in tact. The remote storage comes into play when you lose your building due to fire, flood, or any type of disaster. The DRP discussed in this article supports recovery from both types of disaster. Your firm should be prepared for both.
For those of you that have not planned or prepared for disaster, I hope this article motivates you to implement a disaster recovery plan. Please note, implementing a disaster recovery solution is not an IT management decision, it is an executive decision.
David Kramer is the President of Domain Technology Group, Inc., an information systems consulting company. He can be reached at david@domain-group.com, or at (610) 374-7644.
Well said David. The importance of a disaster recovery plan cannot be stressed enough, especially in today’s economic climate. I am very grateful that you are a member of the Technology Committee at the chamber, you bring a wealth of invaluable information to the table. Thanks