So when the question arose of who would know best about the real potential of HA for small and medium-sized businesses (SMBs), Craig immediately came to mind.
Overcoming the Skeptics
In those companies, back in the days of the AS/400 and the System/36, I'd find the systems tucked away beneath the back stairwell. A fresh tape cartridge was set on a shelf, waiting for the night's backup schedule. Another one was tucked in the purse of the office manager, who would take it home and store it in her closet that night.
System availability back in those days (and maybe even today for some sites) wasn't about "high" or "low" or even really "availability" at all. It was simply about backup! It wasn't pretty, and it wasn't fancy, but it worked! (Mostly!)
So when Craig started talking about real high availability for SMBs, I immediately thought about those thousands of office managers' purses, stuffed with tape cartridges on their daily rotation between home and office.
Not that I thought that those small organizations didn't need a better strategy for their disaster recovery plans. It's just I've spent too many hours in the offices of financial managers who viewed IT as an "expense" and not as a vital element of their operations. And don't forget, back then, having even a single AS/400 represented a "luxury expense" for those companies.
Changing Times and Changing Business
First of all, business continuity has become a more visible requirement for these organizations. One reason is that the System i is delivering a lot more service, and it's connecting to a lot more systems. And while, in the old days, a company might be able to lose a couple days' worth of data transactions if the system went down, there's no such luxury today.
Back then, everything was still being written on paper, and if the clerks had to spend extra hours rekeying the data, that was their job. But today, nearly everything is entered online, and those companies that have their systems connected to supply chains or that are using e-commerce packages don't have the option of rekeying. If the system goes down, the business will be lost.
And as Craig Copeland pointed out in the Webcast, if you've lost a single transaction in an automated process, chances are you've lost that customer too...forever.
Quantifying the Need for HA While Justifying the Expense
I was particularly intrigued during the Webcast about how Craig would quantify the need for HA in these small companies and how he would help them justify the expense of an HA system. His responses were right on target.
Instead of focusing on widgets and hardware and products, he said, "You need to know what it is you want your high availability solution to accomplish."
That was sound advice, because a lot of companies get into HA and never really understand why. They have no clear idea of what their data assets really are and what it's going to cost them if the system goes down. Without the basic knowledge of understanding their exposure, they can't estimate the costs of lost business. They just know that, should the system go down, they'll be feeling a world of pain.
So Craig's approach is to work with the clients, to understand their business needs for business continuity, and only then to begin the steps of evaluating various solutions.
Getting Involved in the Solution
The second sound piece of advice was to tell us that it's absolutely important that the individuals involved in the day-to-day operation of the system be intimately involved in the HA discussion. Why? Because, in the event of a disruption, they are obviously the ones who need to know how to make the swap between systems.
Some novices with HA believe that they're going to get immediate, automatic failover with their systems. But as Craig points out, "Then what? You've got everybody running off your mirrored system, when all that really happened was that someone kicked out the plug to a router on the network." Craig doesn't recommend "automatic" role swapping between systems, but "automated" role swapping, in which the customer is controlling the process. In most cases, Craig says, it takes longer for companies to decide to use their mirrored system than it takes to actually perform the process of role swapping.
Options for Affordability
Some of the most cogent responses from Craig had to do with HA affordability. After all, it was only a few years ago that a true HA solution would be out of the range of most—if not all—of my former small and medium-sized clients. But today, there are lots of options, including options to use hosting and remote management services.
These services are tailored to the SMBs, and they can be really affordable, without the expense of buying more hardware. Such services are new to the computing world of System i, and a lot of credit goes to IBM for engineering the viability of creating the opportunity for these kinds of services.
I wish some of my old small and medium-sized clients would listen to this Webcast. And if your small company is starting to think about the vulnerabilities of your System i to incidents of unplanned downtime, if your window of time for backups is continuing to erode, or if you're just plain tired of stuffing tape cartridges into your purse every night, you might want to take a listen to Craig's advice.
The title of the Webcast is "High Availability for the SMB: A Consultant's View," and it's available free at http://webcasts.mcpressonline.com or by clicking right here! Craig's talk was sponsored by Vision Solutions, but this Webcast was not an infomercial: It was a plain heart-to-heart conversation about the needs of HA for small and medium-sized organizations.
Who knows! Craig may be really speaking your language when it comes to preserving your company's business continuity! And for Christmas this year, maybe you and your team could spring for new, smaller purse for your office manager.