I don’t want to sound alarmist or anything – but if the economy fails, it’s because nobody really got a handle on that whole cloud/enterprise integration thing.
Fourteen Technologies Needed to Implement a Fully Functional Private Cloud
Or, at least, that’s what Natalie Kilner Hughes, founder and principal consultant at PBSI Enterprise Consultants, says in a recent Wired blog post.
And in case you think I’m exaggerating, let’s look at an actual quote:
“The economic consequences of the current market stagnation are potentially catastrophic. In the past two weeks, the press has reported Apple’s first profit drop in a decade, and IBM’s biggest stock price decline in eight years. During the same period, major corporate operations interruptions were reported at American Airlines and The Chicago Board Options Exchange. Though these events were reported as standalone incidents, when you connect the dots, they are, in fact, symptomatic of an industry in trouble.”
Now, PBSI does offer IT services like data quality and coding, so keep that in mind. Still, it’s hard to ignore the data points when you “connect the dots,” as she puts it.
For instance, she writes that technology companies have been losing “billions of dollars in market capitalization,” which is in part due to a slow-down in business spending on enterprise software upgrades.
That would make sense if they didn’t need the upgrades, but actually, they do, she writes, pointing to several recent and public technology outages and failings. So why aren’t enterprises investing in software upgrades?
In a word: integration. In two: cloud migration.
In recent years, enterprise software vendors have focused on a cloud strategy that requires modernization and a willingness to take advantage of what the cloud offers. But enterprises are notoriously stodgy about new technology, and the cloud is no exception.
“Solutions advocating the replacement or rewriting of billions of lines of enterprise software deployed today raise very realistic fears of risk, disruption and cost over runs that have historically prevented significant migration efforts for modernization to the cloud and mobility,” Hughes writes. “Though virtually every major tech company purports to address this enormous enterprise computing opportunity, none has a credible enterprise software modernization solution to address this essential requirement.”
Of course, security is a major sticking point in enterprise cloud adoption. But what seems to be the real sticking point is integrating enterprise business rules with cloud software and mobile devices.
That’s key — we’re not talking about data integration, here. We’re talking about integrating the applications and the rules that govern them.
“It is now clear that cloud platforms and isolated smart devices on the market today, even with enterprise level security, will not deliver their promised value until the vendors address the enterprise customers’ requirement to integrate those offerings with existing legacy software that runs corporate operations every day,” she writes.
I’m sure SaaS and established enterprise vendors will beg to differ, probably by sending me an email right away. (Feel free to just comment below instead, please!)
But in some ways, I’m not sure it matters what businesses believe if you don’t.
Actually, Hughes even says there are platform-specific tools for migrating this legacy code to the cloud that could save up to 80 percent of the cost of rewriting or replacing code. (This may be where she’s pitching her company’s services; honestly, I’m not sure.)
One solution that’s being touted as a way to solve integration between cloud services and enterprise applications is ye old enterprise service bus technology (ESBs). Many web-based integration companies couple the ESB with API management tools to address the kinds of problems that have paralyzed enterprises.
Plus, as ZDNet’s Joe McKendrick points out, there are several good arguments for using ESBs to solve this challenge, including that you’re probably already using an ESB to handle the exact same complex integration issues internally.
But, then again, as I’ve been told a million times, “It never works like they say it will.” So maybe this is less about finding specific technologies to solve the cloud-to-on-premise challenge than it is about eliminating more of the headaches. Or, maybe it’s not even that. Maybe enterprises just need to be convinced that it’s not only possible, it’s actually worthwhile (and possibly, as Hughes points out, a Sarbanes-Oxley violation created by risking a software meltdown).
What do you think: Is this a legacy integration problem or an unwillingness to be pushed to the cloud?
