Not too long ago, companies were more concerned about disaster recovery (DR) than business continuity—as if they were two different aspects of how businesses operate; but essentially, they’re one in the same. If a disaster occurs—whether it’s a system disk going down or a flooded data center—business comes to a halt, interrupting continuity.
Larger enterprises were the first to fully grasp this concept and took prudent steps to support it, using replication software and dedicated backup boxes, for example. Of course, these enterprises had the fiscal resources and IT departments to make sure this worked as planned.
"When you look at things from a number of perspectives, including cost, functionality and ease of use, PowerHA can be very beneficial even for small companies,”
says Keith Livingood, senior developer for Minnesota-based APi Group.
Without similar resources, small and midsized business (SMBs), on the other hand, would often cross their fingers and hope nothing horrible happened. If it did, they would have to rely on daily, weekly and monthly backup tapes—assuming they had them—and rush to an off-site location to restore their entire systems—a long, tedious and tense experience.
That’s rapidly changing, however, even for businesses that can’t afford or don’t want large IT departments to continuously monitor critical systems.
It’s due in part to the continuing evolution of the IBM PowerHA* solution, which doesn’t break the bank, require constant care or deploy as a mere afterthought. Rather, it’s highly integrated at the hardware level.
As a result, DR and business resiliency have become one, and PowerHA technology offers SMBs a way to do what the big guys have been doing for years—now affordably and largely hands-off.
A Wild Situation
One company investing in business continuity and DR is the Minnesota-based APi Group, headquartered in New Brighton.
As Keith Livingood, a senior developer and the organization’s only senior System i* administrator, simply puts it, “We can’t afford any unnecessary downtime, especially unplanned downtime. So other than a period right before our backup process, when we’re down for only about a half an hour, we’re up all the time. We have to be, because we have a U.K. office that’s six hours ahead of us and working by 1:00 in the morning, our time.”
The organization, which is a holding company for more than 35 other companies in the specialty construction and life-safety businesses, has locations in the U.K., Canada and throughout the U.S. It has annual revenues of around $1.5 billion.
Most of APi Group’s companies use a homegrown ERP application running on an IBM i POWER7* blade in a BladeCenter* chassis attached to an IBM XIV* Storage System in a storage-area network (SAN) environment. A similar blade hosts an instance of a third-party ERP system that, according to Livingood, is “used by around 20 users during the day.” The main ERP system handles several transactions, including job costing, payroll, accounts payable and accounts receivable.
“We’re up all the time. We have to be, because we have a U.K. office that’s six hours ahead of us and working by 1:00 in the morning, our time.”
—Keith Livingood, senior developer, APi Group
Due to the critical nature of this system, the company decided to deploy PowerHA technology on the primary blade, with backups being fed into a PowerHA-dedicated POWER6* system with onboard disk. It decided not to run PowerHA on the second blade because of the low level of transactions it processes.
“Because we were acquiring so many companies and bringing them into our computing fold, we thought it would be prudent to have a quick failover system in place in case of a disaster,” Livingood says. “If we didn’t have such a system, we might have to take tape off-site, start up another system and restore. That would be pretty wild.”
Which is why it was beneficial that APi Group had a PowerHA solution in place when its BladeCenter system lost power in July 2012. Although the POWER7 system was completely down, the company was able—thanks to the flexibility of PowerHA—to roll over to the POWER6 system, which then became its primary system.
Livingood was on vacation and golfing when his technical lead called to inform him of the situation. “They were scrambling a bit, so I told my technical lead, ‘Call IBM support, get to the PowerHA team and tell them you lost your production system and need to fail over.’ He did, and it was simply a matter of IBM support telling him which command to type in. That’s all they had to do.”
As Livingood further explains, “It takes only one command to move over to the other system. In our case, it took only 15 minutes, although we did our due diligence to ensure everything was running properly on the POWER6 system. As it turned out, everything was OK. With our old system, that might not have been the case. It was too complex, and we had limited knowledge about how best to use it.”
This DR-type situation meant the company was, in fact, resilient and able to get its core system back online in seemingly no time. And as far as its users were concerned, the actual system outage was only a minor inconvenience compared to the recovery time another DR solution might have involved.
“I don’t think they would have been able to come back online so quickly with their previous solution,” remarks Mike DuBois, chief operating officer and co-owner of the Minneapolis-based TSG Server & Storage, an IBM business partner that works with APi Group.
A Cautionary Tale
DuBois notes another example of a Minnesota-based company where the data center’s air conditioning failed over the weekend, with no apparent notification to the IT staff. When employees returned the following Monday, they discovered that two drives had failed. Without a DR solution or plan in place, it took the company about 36 hours to recover, he says.
This cautionary tale is instructive, especially for companies with limited IT staffing. APi Group’s IT department, for example, has only eight developers who handle everything related to its IBM i systems, and three of those are consultants. This means the company can’t constantly monitor a business-continuity or DR solution.
In fact, some companies have no IT staff at all, despite having servers, storage and networking in their data centers. If something were to go wrong in such cases, these companies would have few options to bring their operations up in a timely and cost-effective manner.
But Livingood doesn’t have to worry with APi Group’s PowerHA solution. As he explains, “I make sure it’s running in the morning, and then I’ll run a few reports every week or so to check for possible mismatches, such as a user profile being out of sync. But that’s really all I have to do.”
“It takes only one command to move over to the other system. In our case, it took only 15 minutes, although we did our due diligence to ensure everything was running properly on the POWER6 system.”
—Keith Livingood, senior developer, APi Group
This is a boon for SMBs, according to DuBois. “From a disaster recovery point of view, it’s pretty much set it and forget it. We’ll occasionally get alerts from our smaller customers that don’t have IT staffs that something may be out of sync, but those are rare. As far as I’m concerned, PowerHA is more straightforward to deploy and use than anything else that’s out there today,” he says.
Livingood advises people to do their homework before implementing a PowerHA solution. In his case, he created a sandbox on one of his systems and began moving the company’s programs and data into an independent auxiliary storage pool (IASP), which is a collection of disk units that can be brought up or taken offline independent of the rest of a system’s storage, to make sure they were available.
As he explains, “You want to test things thoroughly in the case of IASPs, so you can make sure you can still see your data and get to it with nothing being different. IBM has some very good Redbooks* publications regarding IASPs on IBM i Power Systems*, so I recommend that people look them over before diving in too deeply.”
A Big Plus
Livingood sees PowerHA technology as simple enough—yet powerful enough—for SMBs to use for both DR and business continuity, which, given today’s world of clock chasing, has become more or less the same thing. SMBs now have a robust, easy-to-use tool in the PowerHA solution that levels the playing field with larger organizations.
“I spend very little time with PowerHA, except to run those reports every few weeks,” Livingood says. “If something’s out of sync, I just fix it, but other than that, it’s never gone down out of the blue. It’s handy, too, if I’m applying PTFs to the POWER6 system, because all I have to do is suspend PowerHA on the production system, apply the patch, resume PowerHA and then watch as the systems go back into sync. It’s all pretty simple.”
Simple is the keyword here. As DuBois indicates, he’s receiving increasing requests for more information about PowerHA—as well as increased installations—which indicates that more SMBs will be, or already are, following in APi Group’s footsteps.
And for good reason: Many of these organizations have around-the-clock demands, whether it involves offering Web-based customer self-service, having worldwide or time-zone-dependent operations, or being key components in larger companies’ global supply chains. Global manufacturers, for example, are increasingly requiring stricter standards for business continuity. If SMBs aren’t compliant with those policies, they could lose business.
“All kinds of thoughts go through your mind when you’re considering the ramifications of downtime,” Livingood notes. “So when we were looking at replacing our POWER6 server with a POWER7 server, talk of PowerHA naturally came up. When we took a longer look at it, it turned out that PowerHA was considerably cheaper than the alternatives and, importantly, is native to IBM i, which means you don’t have the administrative overhead related to other products. That was a big plus for us. So when you look at things from a number of perspectives, including cost, functionality and ease of use, PowerHA can be very beneficial even for small companies.”
IBM Systems Magazine senior writer
Read More: (http://www.ibmsystemsmag.com/power/Systems-Management/High-Availability/smb_api/?page=1)