Keeping an enterprise IT system running on a daily basis certainly has its multiple challenges. Ensuring that the enterprise can rebound quickly and resume business in the event of a natural disaster, power outage, terrorist or hacker attack, or other unplanned incident is equally important in this day of real-time Internet business.
But there’s a new problem for IT managers that will be impacting enterprises in the near future. A deluge of new data — much larger than most people realize — is being created, and most of it will be business-oriented. Researchers IDC, Gartner Group, Forrester Research, The 451 Group and others have all predicted a huge glut of new digital data that will need to be stored and protected by BC systems.
We’re talking about many exabytes [an exabyte is equal to 1,000 petabytes] of storage. Multiple thousands of exabytes are expected to be created by individuals and companies from every corner of the planet, thanks largely to the proliferation of inexpensive handheld digital devices and videocams.
This storm of data, in turn, is predicated on the swift growth and subsequent workload processing and storage of high-resolution video, surveillance video, financial and health records, scientific data, high-end video games, and high-resolution photos and graphics.
Can BC products and services handle this? Certainly the technology itself is improving almost daily, thanks largely to increasing use of virtualization, deduplication and thin-provisioning methods to keep only the important data and winnow out the chaff. Increased areal density of both hard disk drives and NAND flash solid-state memory are also playing important roles.
Better data center management tools and services also are coming into the market, from a slew of companies that include CA, BMC, Symantec, Dell MessageOne, Hewlett-Packard, Clearview, and Continuity Software.
Despite this, the main issues yet to be solved in the BC world are human-related, such as imagining what the problems might be, planning how to solve them, and implementing final deployments.
Author/researcher Michael Croy of the Forsythe Solutions Group consultancy noted the following:
–Nearly one-third of current businesses that suffer a major IT disaster go out of business within two to four months.
–Two-thirds of mid- to large-sized companies have experienced from one to 24 hours of unplanned down time, with an average cost per incident of $3 million.
–Two-thirds of U.S. companies say it would take days to weeks to recover from a significant business interruption.
Good BC Can Save the Life of an Enterprise
Thus, a good business continuity strategy and process not only can keep vital data juices flowing, it can potentially save an entire enterprise.
Virtualization of computing and storage resources, which came of age in 2007 and 2008, has widened the technical boundaries for business continuity. BC now can be the savior to guard against something as simple as someone tripping over a power cord and disabling a home business desktop computer in the middle of an online transaction, for example.
In this recessionary age, the small example of business continuity is becoming as relevant as the enterprise version as large numbers of people move into independent businesses that include sales on eBay and Craigslist, among others.
In fact, CEO Jacques Baldinger of Paris-based virtualized storage software maker Seanodes uses that specific example when he is demonstrating his company’s products to a potential customer.
“I’ll be showing a system streaming a video, for example. Then I will ask my guest to pull out the [server rack] plug from the wall,” Baldinger said. “Usually, they will shout, ‘No! No! Don’t do that!’ Then I’ll pull the plug for them. Of course the video keeps playing, uninterrupted, no matter what. That’s the beauty of virtualization and BC.”
In Seanodes’ case, the software already has virtualized the storage capacity in an entire system into one large pool with servers located around the world, so it can be carved up and used for both storage and computing. The video streaming process isn’t stopped because since it is using the whole system, it doesn’t care if one or several servers suddenly go off line, blow up or get flooded in a monsoon.
Data Growth the Major Issue
Fadi Albatal, director of marketing for FalconStor — a highly regarded storage and data protection OEM for a number of major systems providers — explained to eWEEK about his take on the pent-up need for more storage that will be unleashed in the next two years.
“Data deduplication will be paramount,” Albatal said. “Other data storage methods, such as online storage backup, will become popular to amortize customer investment in storage infrastructure.
“The cloud will provide a continuum of local infrastructure that will allow customers to accommodate increasingly rapid data growth and to massively scale their storage capacity both in space and time. Also, the integration of intelligent processes and data management solutions will become a necessity rather than a nice-to-have feature,” Abatal said.
The adoption of more current solutions, such as continuous data protection, will replace traditional backup applications that are responsible for the increased pressure on storage capacity, Albatal added.
Good Planning Will Determine BC Success
At any level, like other IT system processes, BC must be conceived and planned long before it goes into live production. It is generally those early planning steps that determine success or failure of such a system.
In his book, “Are We Willing to Take That Risk: 10 Questions Every Executive Should Ask about Business Continuity” (iUniverse, 2008), BC expert Croy lists a series of tough but important questions that must be faced before any kind of BC strategy is deployed.
What is the worst that can happen? How prepared are we? Where is the business vulnerable? How do we know our plans will work? What about our people? How do we determine our risk tolerance? How can we leverage our technology? Are we willing to take that risk?
“At the end of the day, this is not about infrastructure technology; it’s about the business,” Croy told eWEEK. “Business executives need to focus on this in realistic terms first. The 10 questions I pose in the book often don’t get asked. You’ve got to start there.”
It used to be that, in the old days, that an enterprise had one big machine, in one room, that when you recovered something, it was just that big machine, Croy said. “It was one application, one process, one function. Now, we’ve got multiple processes that have to be restored in the correct order, or nothing will work well.”
Specialized BC software is the only way to handle this complicated workflow.
How to Exist During Recovery
Business continuity doesn’t simply mean that a process restarts and continues to operate following a disruption. CTO Kelly Lipp of data protection provider STORServer told eWEEK he believes the biggest business continuity problem facing enterprises is that while many have thought about how to recover, not enough time has been spent on how to exist during the recovery period, which sometimes can be hours.
“Not all recoveries happen quickly,” Lipp told eWEEK. “Developing contingencies and an ‘exist-without plan’ will help eliminate the impact of a disaster and is where the most time needs to be spent.”
Imagining various possible outages is the best way to start this process, Lipp said. Using the loss of e-mail as one scenario, Lipp said that if e-mail is the key communications method for an organization — and it is for most — one can expect big problems fairly soon. It’s the nature of the business.
“If you have a good BC plan, you’ve thought through how you will communicate if e-mail is unavailable,” Lipp said. “Your plan might have time triggers to help you determine which alternative you will employ. Based on information from the recovery team, you can begin to implement your alternatives.”
Never Overlook Best Practices
Iron Mountain has been protecting business information since 1951 and acquired Stratify, a provider of advanced electronic discovery services for the legal market, in 2007. Here are the BC best practices that Iron Mountain/Stratify recommends to its customers:
1. Make backup copies of all critical data. Companies need to ensure all of their critical data is backed up — not just primary servers and the data center.
2. Protect your backup data off-site. Send your backup data off-site to a trusted third party that will ensure you have your data when and where you need it.
3. Have a disaster recovery plan and test it. With regular testing of their plans, companies have “pre-made” decisions that are more difficult to make during an actual crisis.
4. Test your backup solution. Companies should conduct simple restore tests with their backup solutions either weekly or after critical events, such as the closing of the quarterly books.
5. Communicate early and often with your disaster recovery vendors. Make sure you understand the protocols for “declaring a disaster” with your off-site data protection provider.