Hard out there for an incumbent.
Most enterprises would love to embrace the future. The problem is they’re so heavily saddled with the past. This is as true for customers as it is for vendors.
Nowhere was this more evident than at VMworld this past week, where VMware, beset on all sides by containers, public clouds and other existential threats, embraced them all. Every single one of them.
Given its obvious conflict of interest—VMware’s crown-jewel virtualization tech pays the bills even if containers and cloud draw the developer crowds—there’s a real question whether any incumbent vendor, even a great one like VMware, can usher its customers into the future of computing.
Developers Steering The Ship
Developers have never been more powerful. Dubbed “the new kingmakers” by Redmonk, developers are driving fundamental shifts in how enterprise IT is delivered and consumed. In virtually every instance, developer-led IT is driven by agility and convenience, not more established concerns like security.
This, in turn, has powered Amazon Web Services to hit and surpass $1 billion in revenue faster than any software business in history. It also explains much of the enthusiasm for middleware containers such as Docker, which aim to simplify both app development and IT infrastructure management by eliminating software dependence on particular operating environments (virtual or otherwise) and hardware.
As such, the claim by VMware executive Bill Fathers that IT operations is regaining control over developers represents wishful thinking. At best.
What Choice Do They Have?
It’s not surprising that VMware would be hopeful, however: it has so much invested in the old world of virtualized data centers that it cannot help but pray that it can bridge the chasm between what it provides and what the market now wants. They’re not the same thing.
Not that VMware has much of a choice. As Windows analyst Paul Thurrott highlighted to me:
In other words, you can believe, as Thurrott does, that “VMware is on a long downward trajectory” while still agreeing with Pivotal’s Andrew Clay Shafer that there isn’t an obviously better solution.
It may be that containers are complementary to virtualization, whether by design or by necessity. As Pivotal’s Andrew Clay Shafer posits, “the preponderance of containers will run in VMs for the foreseeable future.” He’s probably right.
But it’s not clear this will last. As former Gartner analyst and current Apprenda executive Chris Gaun notes, there are some companies that are using containers to significantly lower their VMware investments. In one particular case, it’s apparently a VMware customer running thousands of applications and saving millions of dollars by making the switch to containers.
And so VMware, like its customers, is forced to shore up its legacy business by partnering with Docker and the OpenStack crowd.
“Bimodal IT” To The Rescue?
Gartner analyst Lydia Leong calls this “bimodal IT.” As she explains:
Bimodal IT is the idea that organizations need two speeds of IT—call them traditional IT and agile IT…. Traditional IT is focused on “doing IT right,” with a strong emphasis on efficiency and safety, approval-based governance and price-for-performance. Agile IT is focused on “doing IT fast,” supporting prototyping and iterative development, rapid delivery, continuous and process-based governance, and value to the business (being business-centric and close to the customer).
Not surprisingly, it’s generally not possible to use the same team to get these very different results. As Leong suggests, enterprises need “different people, processes, and tools supporting each” mode of IT. You’re simply not going to find a good agile-minded DevOps person that wants to carefully plod down the policy-laden path of traditional IT.
So you embrace both, but with different groups.
VMware’s approach, according to Leong, fails to achieve this. She summarizes VMware’s VMworld announcements as proffering a “heavy overtone of reassurance that the VMware faithful can continue to do business as usual, partaking of some cool new technologies in conjunction with the VMware infrastructure that they know and love—and control.”
But, as she goes on to argue, “VMware needs to find ways to be relevant to the agile-IT mode, rather than just helping traditional-IT VMware admins try to improve operations efficiency in a desperate grasp to retain control.”
It doesn’t help VMware that its developer-friendly assets like CloudFoundry are run out of a completely separate company, Pivotal. That makes it that much harder for it to offer a truly cohesive story on how these VMware-savvy enterprises are going to become cloud/container-savvy enterprises.
The Future Is Here: It’s Just Really Messy
This isn’t to pick on VMware. In a Clayton Christenesque Greek tragedy sort of way, VMware is here a victim of its own success. No company did more to make data centers dramatically more efficient, or to profit therefrom.
But the same is true of enterprises in general. They see these cool new things like containers and want to embrace them, but they’re constrained by the technology investments they’ve already made, including VMware’s virtualization technology.
The answer may be to spin up a bimodal approach to change, following Gartner’s recommendation. The reality, however, is that it’s really hard to change, no matter how you go about it.