Regular readers of Virtualization Review will notice something different about the site today. Namely (see what I did there?) that we have a new name: henceforth, we are Virtualization & Cloud Review. Rolls trippingly off the tongue, doesn't it?
I'm excited about the new name, despite the length. It's a long time coming, but is a necessary tweak. We needed an updated moniker to reflect our evolving coverage. We still cover virtualization, of course; all aspects of it, as always. But the role of virtualization has changed dramatically over the years, and we needed to address those changes.
When I was the founding editor of Virtualization Review, way back in March of 2008 (has it really been nine years?), virtualization was the new kid on the datacenter block. It had an air of mystery about it; most IT folks then couldn't have told you what a hypervisor was, what it did, or why it was important.
Now hop in the DeLorean and fly back to the present. Hypervisors are deeply embedded in datacenters, and are mostly a "set it and forget it" technology. Virtualization, though, has grown by leaps and bounds, into most areas of IT-related infrastructure.
Today, virtualization is still at the forefront, but in a different capacity: it enables most of the cloud to work. Through technologies like software-defined networking and storage, hyper-convergence, containers, application virtualization and so on, the cloud sits on a foundation of virtualization. Here's how I've been describing it lately: if cloud computing is the house, virtualization is the plumbing. You don't have one without the other.
Our coverage has changed over the years to reflect this new reality, and our name needed to similarly change. It's growing increasingly rare to read an article here that doesn't touch in some way on the cloud, whether private, public or hybrid. We write about Amazon Web Services (AWS), Microsoft Azure, Google Cloud, VMware's various cloud efforts (like VMware Cloud on AWS), and more. If it's cloud infrastructure-related, we're all over it.
That doesn't mean we'll be forgetting about virtualization. Take a look at the new title: Virtualization remains front and center, and will remain so. But now, our title better fits our comprehensive coverage.
Virtualization isn't going anywhere. Neither is cloud computing. Neither is the new Virtualization & Cloud Review. Let me know what you think of the change: like, no like, or no care?
Posted by Keith Ward on 03/07/2017 at 11:27 AM0 comments
LAS VEGAS – VMworld 2016 still has several days to go as I write this, but most of the big news announcements have been made, so it's safe to discuss the main things I learned about VMware's strategy in the near term as outlined in both keynotes, as well as discussions I've had with multiple company executives.
- VMware is determined to make this "cloud" thing work. Despite the lack of uptake for vCloud Air in the public space, it hopes that its "software-defined" tech, like NSX software-defined networking and VSAN software-defined storage, will find a place in the Amazon Web Services and Microsoft Azure platforms. This is not mere hope, either: both technologies are cloud-ready and enhance cloud computing. It's also smart, in that if customers like what NSX/VSAN offer, they may look at other products having it, like hyper-converged appliances.
- IBM isn't a bad place to plant the Cloud Foundation flag. VMware's Cloud Foundation – a suite of products including vSphere, NSX, SDDC Manager and VSAN – is being made available first on IBM's public cloud. Although IBM is generally considered to be the fourth-place public cloud platform, behind Amazon Web Services, Microsoft Azure and Google Cloud, it's an up-and-comer, and feels like a good fit for Cloud Foundation. IBM remains an exceptionally strong brand, and partnering with it helps burnish VMware's public cloud image.
- It's amazing (at least to me) how important NSX has become to VMware. I doubt that many believed that VMware's purchase of Nicira in 2012 would yield such ripe fruit. It was clear that software-defined networking (SDN) would be big, but one could make the case that it's become VMware's most important product going forward. It's the main driver for so much of what the company's doing in both cloud and end-user computing. Its micro-segmentation abilities are crucial for security, for example.
- vSphere Integrated Containers (VIC) has a bright future. Admins and developers both love containers, and VMware has embraced the hot new technology fully. It showcased some of VIC's new technologies, including a private container registry for better ease-of-use and security, and a management portal. The updates address significant user concerns, and managing containers on top of an environment they already know in vSphere will encourage further use of containers, both Docker and alternatives..
- With the clearing of the final regulatory hurdle, Dell and EMC will be united in holy matrimony on Sept. 7. That means that just as VMware employees are starting to recover from VMworld 2016, another wave crashes into them in the form of the long, possibly torturous process of integrating with the new Dell Technologies behemoth. No rest for the weary.
- I truly do not believe that Michael Dell is going to be gumming up the VMware works. I asked him and VMware CEO Pat Gelsinger about possible interference point-blank at a press briefing on Monday. The response I got, combined with discussions with many, many sources over the past year, leave me feeling confident that VMware will be left alone in much the same way it has been under EMC's stewardship.
- I like Las Vegas as a venue, mainly because I can stay in the same hotel as the show. I like San Francisco a lot, too. Both are fine destination cities. But would it kill you, VMware, to hold a VMworld on the east coast in the future? Pretty please?
Posted by Keith Ward on 08/30/2016 at 7:11 PM0 comments
Click here for notes from day one.
Day two of virtualization's biggest annual event has arrived. Here are some notes from the field.
- Unofficial attendance, according to several VMware PR staffers, is 23,000. That equals last year's attendance. The 2015 number was a record, so although no new record was set this year, attendance didn't drop, either. It still does show that the interest in virtualization hasn't diminished – a very good thing for VMware.
- It's often the small things in life that matter. I left my mouse at home, and it annoyed me all day Monday. I'm much more productive on my laptop with my mouse. Fortunately, the FedEx store in the hotel had them for sale. That's the last place I would've expected to find a mouse. It saves me from having to go a mile or two to get one. Again, a little thing, but life moves quickly at the show, and anything that speeds me up is gold.
- Media breakfasts that start at 7 a.m. are hard. Veeam sponsored the Tuesday morning breakfast, and it was enjoyable. Doug Hazelman, VP of Product Strategy, gave a comprehensive overview of new (and upcoming releases) from Veeam, and many questions were asked an answered. Still, that doesn't mean that 7 a.m. isn't early.
A few nuggets from the breakfast presentation: Veeam now has about 205,000 customers, and it's been around for 10 years. One of the more unusual aspects of that longevity is that Ratmir Timashev is still with the company he co-founded. Although he's stepped away from the day-to-day operations, he hasn't left for another feisty startup, as is so common in the industry.
Finally, new Veeam President and COO Peter McKay, formerly a VMware exec, mentioned that Veeam is going to start heavily targeting the enterprise market segment with its products. That should be interesting.
- To me, it felt like the Tuesday keynote had more energy and was more relevant to the average customer. The cloud – especially the public cloud – feels more abstract and less immediate; and, therefore, more useful.
This isn't a knock on Pat Gelsinger, but rather a reflection of the content. Look at what was discussed: containers, both on vSphere and non-vSphere platforms; software-defined technologies, with a focus on NSX; Workspace One; OpenStack; hyper-converged infrastructure. It seemed like there was more audience reaction as well. To me, it moved much faster, and felt shorter, even though it was about 10 minutes longer.
- I'll end with one miscellaneous annoyance. I bought a 10-ounce bag of trail mix this morning to snack on. It cost $14. Yeah, I'm in Vegas.
Posted by Keith Ward on 08/30/2016 at 11:21 AM0 comments
LAS VEGAS -- VMworld returns to Las Vegas for the first time since 2011 (as I wrote about in my "Brief History of VMworld" article). Has it really been five years since VMworld was in the desert?
The show would still be in San Francisco, except its home there, the Moscone Center, is undergoing renovations. But Vegas does have its charms. The venue, at the Mandalay Bay hotel, is terrific (lots of pools here, to offset the inferno outside.)
One advantage of Vegas is that the conference is in the hotel, so there are no shuttles necessary for those of us lucky enough to be staying at Mandalay Bay. Not a big deal, but still nice.
Lots of folks here, as always, but I don't have an official attendee count yet. I'll update this with the information when it's available.
- The Mandalay Bay is big. Like, really, really big. It takes a long time to get to the conference center, and involves passing lots of slot machines. It's probably a good thing I'm not a gambler.
- Breakfast this morning: some sort of breakfast burrito-type thing. Really, VMworld? Sorry, but for those of us who don't like Mexican, that's just gross – especially on the first day.
- The keynote streamed well on my iPhone 6S. That's a real bonus, since wi-fi is notoriously unreliable at big shows like this.
- CEO Pat Gelsinger delivered a solid opening day keynote, as he always does. The Cloud Foundation stuff wasn't eye-opening; anyone paying attention over the last few months saw this coming. But it is important; bundling software-defined technologies like NSX and VSAN and inserting them into public cloud platforms, extending their management capabilities, is a solid direction. NSX's importance to VMware continues to grow.
- Gelsinger's interview with Michael Dell was a snoozer. I can't imagine that anyone gained any insight into anything during that little chat. Dell said nothing of value, and Gelsinger didn't ask the big question: "When will the acquisition be final?" Now that would have added some interest.
- Nice backpack, as always. Black with blue accents. Plenty of room, and a handy water bottle.
That's all for now. Enjoy the show.
Posted by Keith Ward on 08/29/2016 at 12:36 PM0 comments
Related Stories:
So, is he staying or is he going? That's been the question on the minds of many in the virtualization industry the last few days, following the explosive report by CRN Tuesday night that VMware CEO Pat Gelsinger was leaving the company after the Dell/EMC marriage is completed.
He Said/He Said
The story brought swift and
strong denials from the PR arms of both VMware and EMC, VMware's parent company, and from
Gelsinger himself, speaking at a tech conference the morning after the story broke. There's a lot to untangle here, from multiple angles, and this surely won't be the end of the story.
The key question to answer: Is the CRN story accurate? Of course, there's no way to know that for sure. Reporter Kevin McLaughlin relies completely on anonymous sources for the details. It's impossible to know from the story how many sources were used, but he does quote a "high-level EMC source" at length.
Axes and Agendas
The problem with using unnamed sources is that we don't know if they
really know anything. They could have an axe to grind, or a secret agenda that that reporter doesn't know about. I'm uncomfortable with stories that use nothing but anonymous sources, especially ones that trash a story's subject, as some of these do. Here's some of what the EMC source said about Gelsinger: "Pat is not an easy guy to work with. Pat likes to do what Pat likes," and "The goal of Dell Technologies is to work closer as a team … I don't know if Pat is completely on board with that model." Those could be the words of someone who completely understands the situation; or the words of someone with a personal vendetta against Gelsinger.
However, stories like this are often difficult or impossible to report without heavy reliance on anonymous sources. I'd imagine that was the case here; given the gravity of the topic, these sources probably wouldn't have said anything if they weren't granted anonymity. It puts a reporter in a tough situation, but on the whole, I tend to side with the reporter in this case; it's clear this isn't a one-source story.
Another factor here is that with stories like this, it's very likely a lot of vetting was done before publication. I don't know anything about the reporter or CRN's editing methodology, but traditionally, the bigger the story -- and the more it relies on unnamed sources -- the more careful the reporter and publication are. CRN has a good reputation in the industry, and that has to count for much.
Swift, Strong Denials
On the other hand, the denials from Gelsinger and the PR statements were swift and unequivocal that the story was a fabrication. Here's what Gelsinger said: "I categorically deny it, EMC categorically denies it, and Dell categorically denies it. There's absolutely no merit or substance to the rumor whatsoever." Those denials would be unusual if the story had any truth to it. Typically, if a story is true, the denials will be more "non-denial" denials, giving some wiggle room.
For instance, Gelsinger could have said something like "That was a ridiculous story, and I have no idea where CRN got their information from. Don't believe everything you read on the Internet." Those kinds of statements make alarm bells go off in my head, but the denials in this case are clear and straightforward. Thus, we're left at an impasse. And I can see the logic in either Gelsinger staying, or moving on after the merger is complete.
Pros and Cons of Leaving
Here's the argument for him leaving: Things have changed enough since the Dell acquisition was announced that a parade of high-level VMware executives have already departed. That does indicate some level of unhappiness within the company. Gelsinger may be experiencing the same thing.
And if you were Pat Gelsinger, and looking for an exit strategy, leaving right after the merger would be a natural time to do it. The upheaval that's been happening (including 800 layoffs in the first half of this year) might convince you it's time to seek greener pastures. You've successfully led the transition from Paul Maritz to the next, post-merger CEO, and helped VMware grow to be a $6 billion-plus company.
Here's the argument for staying: This would be the worst time to leave, given the challenges facing the company. There's going to be enough to do simply integrating with Dell; adding a new CEO to that, with a new vision, leadership style, management style and so on, is the last thing VMware needs. Michael Dell can be a persuasive guy, and Gelsinger said that Dell expects him to continue in his role.
Whatever the truth of the CRN story, it's clear that VMware is facing a challenge it never has before: the hypervisor market simply won't sustain it as a company anymore, as it's reached saturation. It's branching out into numerous new directions, some of which (e.g., end-user computing) are looking more promising than others at the moment (e.g., hybrid cloud).
Dell's Decision
Ultimately, the decision may not even be Gelsinger's. Once Dell owns VMware, it may decide to replace him, whether he wants to stay or not. Until then, however, Gelsinger's denials would appear to be the final word on his future.
For now.
Posted by Keith Ward on 05/12/2016 at 10:30 AM0 comments
What was your first VMworld like? What was the strangest thing that happened to you at a VMworld event? What do you like most/dislike most about VMworld?
I'm putting together an article for an upcoming edition of Virtualization Review magazine, and I'd like your feedback on these things. It's a remembrance-type of piece on VMworld, which started pretty small and has grown to be the single biggest event in the virtualization industry.
What I'm looking for is your stories. There's no limit on what you can discuss. Here are some ideas:
- How has VMworld changed for you over the years? Has it gotten better or worse?
- What surprised you most about your first VMworld?
- What was your favorite keynote speech, and why?
- What was the craziest thing you saw on the show floor?
- What was the best swag you ever received?
Those are just some ideas; you can pick any VMworld-related memory you want. If you'd like to contribute a reminiscence or story, here's the process.
- Write up your piece. There is no length requirement -- long or short, it doesn't matter. But when you're done writing, get away from the article for an hour or a day, and then come back and edit it. You'll be surprised at how many mistakes/errors you find in it. In other words, please make sure your copy is as clean as possible. It saves me work, and also ensures that your story is clear to the readers. You may also want to include your job title, but this isn't required.
- Save it as a Word file, either a .doc or docx. If you don't have Word, you can dump it into the body of an email. If you do that, make sure you still do an edit pass over it before hitting Send.
- Include your name and city/state.
- Email it to me as an attachment. CRUCIAL: Put in the Subject line "VMworld Submission." If it doesn't say that, your story will be deleted unread.
- Attach a high resolution (at least 300 d.p.i.), clear headshot of yourself. Note that this is optional. You don't have to do this, but your chances of having your article appear in the print magazine are increased by including a headshot.
- Send your contribution to me by May 16.
I'm at kward at 1105media dot com.
If you have any questions about this, use the same Subject line, i.e. "VMworld Submission," to guarantee that it won't go into my spam filter. One final note: VMware employees are more than welcome to submit.
I look forward to hearing your stories!
Posted by Keith Ward on 04/15/2016 at 9:07 AM0 comments
When you think "virtualization," the name VMware is almost sure to follow close behind. They go together like The Colonel and Kentucky Fried Chicken.
But when you think "Microsoft," the word "virtualization" doesn't spring as quickly to mind, if at all. You think "Windows." You might think "PC." In this day and age, you might even think "Azure." "Virtualization," though, is unlikely to make that list.
Although that's quite natural, it also undersells how involved Microsoft is with virtualization. Redmond has had its eye on the benefits of virtualization since it unveiled Hyper-V, lo these many moons ago. In fact, it helped kick off a skirmish back then when it gave away Hyper-V for free as part of Windows Server. At the time, VMware's ESX hypervisor wasn't free, but soon thereafter VMware started giving it away as well.
Hyper-V, of course, has been vastly improved over the years, and it's very popular in enterprises, often sitting side-by-side with vSphere. Despite that, Microsoft keeps Hyper-V a low-profile technology; it would much rather talk about Windows Server or Office 365 than break out its hypervisor separately.
That minimization applies to all of Microsoft's virtualization tech; it's a slice of the pie, but just a slice. It helps customers economize, and now it's a key factor in its Azure public cloud stack. It almost never takes center stage, and that's the way Microsoft likes it.
But there's no doubt that Microsoft sees virtualization as integral to its strategy going forward, whether it's in the datacenter or the cloud. And, of course, it's pretty integral to a magazine with the name Virtualization Review. That's why we've inaugurated a brand-new column to focus exclusively on Microsoft's virtualization efforts.
It's called Virtual Microsoft, and the first column is now out. The author is veteran Microsoft writer and admin Paul Schnackenburg. Paul's written about Hyper-V and Microsoft for years, and regular visitors to this Web site will recognize his byline. He runs an IT consultancy focused on Microsoft, and uses the technology on a daily basis.
One of Paul's strengths is his in-the-trenches knowledge of Microsoft virtualization. He understands how to get most out of this stuff, because it's what his clients use. His articles have a real-world feel to them, because he's out there implementing and managing virtualization in the same sort of environments many of you work in. He knows what's important to you as an admin, because they're the same things that are important to him as an admin.
Virtual Microsoft will cover everything from Hyper-V to Azure to Windows Containers, in Paul's down-to-earth, easy-to-read style. There will be how-tos, technology overviews and sneak previews of coming products. I think you'll find it full of delicious, useful information.
Maybe even finger-lickin' good.
Posted by Keith Ward on 03/22/2016 at 6:27 AM0 comments
Is the "Stop the presses!" Dell buyout of VMware parent EMC in trouble? That's what some media outlets seem to think.
As pretty much everyone in the IT world knows by now, last October Dell pledged to buy storage titan EMC for $67 billion. EMC's crown jewel is VMware, which has become the crown jewel at least partially because it continues to operate independent of its owner. EMC has been hands off, and that's just the way VMware likes it. (Whether that same laissez-faire attitude will continue under Dell was the subject of a December cover story in Virtualization Review magazine.)
The site Seeking Alpha reported yesterday that financing for the mega-merger could be coming off the rails. It cited, among other factors, a New York Post story claiming that Dell's having trouble selling off Perot Systems to raise cash for the deal, and the banks helping to raise more of the $45 billion needed are also facing difficulties.
Technology analyst and writer Kurt Marko saw the news, and Tweeted out his skepticism as well:
Seeking Alpha also noted that although Dell still believes the deal will go through, that there is a $4 billion "breakup" fee it will owe, if things fall apart. That isn't small change for any company, and will certainly motivate Michael Dell to get this thing done.
I won't speculate on whether or not the merger's in trouble, mainly because I don't have enough information to make that determination. I do know that the deal is nightmarishly complicated, in part because Dell's a private company (now), while VMware remains a public company.
In the cover story referenced earlier, some analysts speculated that Dell could raise money for the acquisition of EMC by selling off VMware. That would be true if Dell doesn't see the same product synergies for VMware that it obviously has with EMC, and how desperate it becomes to make the deal a reality. Whether that's the case or not, again, I have no inside knowledge.
But if the tech market doesn't bounce back quickly, or gets even worse, it's possible Dell will need to look at alternatives to raise enough scratch to buy EMC. Things could get very interesting this spring and summer.
Posted by Keith Ward on 02/12/2016 at 10:09 AM0 comments
As time rolls on following the announcement that Dell is buying EMC, things seem to be getting murkier for VMware, rather than more clear. The latest fallout from this confusion is that VMware stock is plunging, losing 20 percent of its value.
"Stock in VMware, which trades under its own symbol, dropped nearly 20% Wednesday after EMC said it would reorganize the two companies' cloud-computing offerings into a subsidiary named Virtustream, which is expected to go live early next year," reported Robert McMillan of Dow Jones Business News.
New Cloud Business
No one seems to know exactly what will happen to VMware, as the particulars of the Dell acquisition, a
$67 billion purchase that's the biggest in tech history, continue to be released. But one major new direction involves Virtustream, a cloud services company that EMC acquired over the summer to beef up its cloud offerings. According to a
VMware press release, VMware and EMC, its parent company (for now, at least), will merge their cloud businesses, to "… incorporate and align cloud capabilities of EMC Information Infrastructure, Virtustream and VMware to provide the complete spectrum of on- and off-premises cloud offerings."
The new business, which will maintain the Virtustream branding, will be owned by both EMC and VMware, and will be run by Virtustream CEO Rodney Rogers. Its financial results will be folded into VMware's, beginning with Q1 2016.
More Public Cloud
VMware is positioning Virtustream as an expansion of vCloud Air, its hybrid cloud offering. Currently vCloud Air is used mostly for on-premises clouds, and hasn't made big inroads in the public cloud space which is dominated by Amazon Web Services (AWS) and Microsoft Azure.
David Goulden, EMC Information Infrastructure CEO, said in the press release that the combining of the cloud offerings "… is central to our strategy to help customers move all of their applications to the cloud. The new Virtustream business will enable customers to implement a hybrid cloud-based IT environment that incorporates the best of both public and private cloud quickly, and from one source."
If it all sounds confusing, that's because it is. It's still unclear how all this will shake out, with Dell having gone private some years ago, and VMware apparently staying public. The assumption is that EMC will go private as well, but again, those details are sketchy at best. The financial site Motley Fool had this to say, when it analyzed the VMware stock drop:
"The bearish analyst army cited weak forward guidance numbers and an unclear future. Majority owner EMC is in the process of going private under the wing of computer systems giant Dell and a consortium of private equity firms. The buyers took great pains to distance their VMware interests from their outright ownership of EMC, and this report did nothing to clarify the big picture."
Federation Complications
EMC has suffered much over the last few years with its "Federation" model of buying essentially separate IT businesses, and letting them operate as independent fiefdoms. Will Dell now become the overlord, adding servers to the storage, virtualization, cloud, security and other businesses? If so, how will it be any more successful than EMC, which has suffered under the current model?
Perhaps that was why EMC CEO Joe Tucci attempted to assuage those fears during a conference call with investors, as reported by Dow Jones Business News's McMillan. "I could tell you that what I think didn't come across was that putting these together will actually save money and actually put more profit on the VMware bottom line," he quoted Tucci as saying.
A Long-Term Play
The question warranted by that statement, though, is simple: Why does he think that, given EMC's recent history? EMC's problems haven't seemed to harm VMware, because of its independence, but the long-terms effects of the Dell acquisition on VMware won't be felt for some time. What's apparent in the immediate aftermath, though, is that the uncertainty isn't helping it. At all.
Posted by Keith Ward on 10/22/2015 at 10:25 AM0 comments
So, Dell now owns EMC, and by extension, VMware. There is so much to say about this, even though so little is actually known.
What we do know is that the deal is done; announcements on both Dell's and EMC's sites have confirmed it. I can't even begin to understand the financials of how it will work, of how a private company buys a public company. And for me, that doesn't matter, at least in terms of the most interesting part of the deal for those of us whose main interest here is VMware.
EMC's Laissez-Faire Approach
What does matter most is this: what are Dell's designs on VMware? VMware has thrived, and EMC's misfortunes haven't been a drag on the virtualization giant. Most folks believe it's because VMware has retained its fierce independence, and EMC CEO Joe Tucci has been hands-off for the most part.
While VMware was part of the EMC "empire" -- the most valuable part, by far -- EMC recognized the wisdom of a laissez-faire approach. That approach paid off quite handsomely, as the bulk of EMC's value is derived directly from the fact that it owns VMware.
The big question, for me at least, is whether or not Dell will recognize this and be similarly flexible, letting VMware do its own thing. Michael Dell, Dell's CEO and founder, said something interesting in his short message about the merger: "Dell and EMC are both companies that have a demonstrated ability to win in fast-changing markets, and that will only accelerate from here."
Riding the Wave
That's only partially correct. EMC has
not shown the ability to consistently win in those "fast-changing markets." VMware, on the other hand, has been much better at that. As I
pointed out last week when these acquisition rumors were making the rounds, VMware has ridden the front edge of the technology wave. It's embraced things like software-defined networking, containers and hybrid clouds that aren't yet core technologies or big money-makers for it, all in the name of staying ahead of the curve.
With Dell soon to own VMware, will that change? Will Dell insist on micro-managing VMware, thinking it knows best? Or will it assert its power to force VMware to fall in line with its vision in areas where they clash? The answer seems obvious: why mess with a good thing? But it's happened before; egos get in the way, particularly in the technology industry.
It's almost certain that some layoffs and restructuring will accompany this move, as Dell decides what works in the "federation" of disparate companies that make up EMC and what doesn't. What it does with VMware will, in many ways, be the most important decisions it will make, and will shape the future of much of the tech sector.
Brave New World
Today is Columbus Day in America, when the discovery of the New World is celebrated. It's perhaps appropriate, therefore, that a new world is arising in this acquisition. Dell, EMC, VMware and the rest face an uncertain future, as did America in the early days. Their first few steps on the new continent will go a long way toward shaping their destiny.
Posted by Keith Ward on 10/12/2015 at 6:13 AM0 comments
Typically, the only way Apple is connected in any way to virtualization is when people use Parallels or VMware Fusion to run Windows in a virtual machine on a Mac. But Apple's in virtualization news today, and in a way that may make VMware nervous.
Filing for Divorce
Kevin McLaughlin of CRN reported that Apple, the world's second most valuable company after Exxon Mobile Corp. (as of today), has cancelled its Enterprise License Agreement (ELA) with VMware and decided to start using open source KVM as its hypervisor.
The story, in which McLaughlin quotes "multiple" anonymous sources, is sketchy on how much of Apple's infrastructure runs on top of VMware, and whether or not Apple is ripping-and-replacing all its VMware stuff in favor of alternatives, or just some of it.
One of the sources he quotes said that Apple pays about $20 million through the ELA, and that it "… wasn't satisfied with the licensing terms VMware offered for the ELA extension," McLaughlin reports. I would take that to mean that Apple decided it wasn't getting enough ROI through the agreement. In addition, the report said that Apple's considering OpenStack and KVM to replace its private cloud infrastructure.
Following the Leader?
Why should this make VMware nervous? After all, $20 million isn't going to materially affect its bottom line, not when it's a $6 billion-plus company. But VMware constantly touts its offerings as the simplest to install and manage. If a company with an infrastructure as gigantic as Apple's has decided that it can do without the bells and whistles of vSphere, vCloud Air and all the rest, other companies may follow suit. Apple's hugely influential, as everyone knows.
Open source software like KVM isn't known for being user friendly. OpenStack in particular still has a reputation for being very complex to install and manage, which is a big reason it hasn't made huge inroads yet in public cloud, which continues to be dominated by Amazon Web Services (AWS) and Microsoft Azure. If Apple believes that either 1) there's enough third-party tools out there to smooth the management's rough edges, or 2) that it will cost less than $20 million annually to hire experts or train its own to virtualize with KVM, it could spell trouble for VMware.
The high cost of ELAs for VMware has been a complaint for years. In that way, it's like most other companies that dominate a space. But when alternatives crop up, and get to the point where they're viable enough to get into production use, other business take notice.
When 'Good Enough' Becomes Good Enough
In that way, at least, it's analogous to Hyper-V. Microsoft's hypervisor was no match for ESX (and later ESXi) for several years. But Hyper-V is now good enough to compete. It's still not near as feature-rich as vSphere, but for many shops, it doesn't need to be; it just needs to be good enough.
If we're at the point where a titan like Apple has decided that KVM is good enough, it should send tremors through VMware.
Posted by Keith Ward on 10/05/2015 at 2:07 PM0 comments
I was at VMworld 2015, and watched the keynotes and read the press releases and came away with a pretty good idea of what VMware wants the public to know about going forward. And if I were to sum up what I learned at the show, it would be this: VMware wants you in the cloud, and now.
Taking Risks
This is a dangerous strategy for VMware. I thought the company did a good job laying out its vision of the Unified Hybrid Cloud platform, and what it could mean for IT. It sounds promising, too -- the ability to move seamlessly and transparently between public and private clouds is fantastic (and yeah, the cross-cloud vMotion demo was awesome). If VMware makes it easier to move on-premises apps to the cloud without a huge amount of disruption, so much the better (although I remain skeptical at how easy this will be).
In addition, VMware EVO SDDC (which replaces, in name at least, EVO:RACK) looks like a strong product. "Software-defined" is the future, and VMware is out in front on it. Doing things in software provides a flexibility and scalability that just can't be matched in the physical world.
Time Sync
My big concern for VMware is one of timing, and where most enterprises are in their transitions to the cloud and software-defined computing. How many of you are still figuring out your private cloud strategy? How many are using public cloud for nothing more than dev testing or similar non-production tasks? My hunch is that most shops are still taking baby steps into the shallow end of the cloud pool, worried about just how cold the water will be when they plunge in.
It's often been said that corporations are years behind -- often 5-to-7 years behind -- the technology curve. Companies often don't move to Windows Server nearly as quickly as Microsoft would like, for example. The same dynamic exists when it comes to virtualization; I'd bet a fair amount of money that the huge majority of datacenters are still running vSphere 5.5 (or 5.0) instead of 6.0.
Eve of Disruption
But that kind of upgrade isn't nearly as ulcer-inducing as re-architecting apps for the cloud or re-thinking your entire networking concepts. In other words, resistance to this kind of change could be greater than the normal "new version of standard infrastructure product is released," which often requires little more than learning a new UI and what apps it might break.
Containers are in a similar boat. There's lots of talk about them, and VMware played them up big time at the show. But again, how many shops are using them in regular, large-scale production? My sense is that it's still a very small percentage. They're being studied, sure, and likely rolled out on a test basis. Major implementations, however, are likely to be pretty far in the future for most companies.
Future Tense
That's how I felt about VMworld 2015: lots of cool stuff was demonstrated in the keynotes, but little that was of immediate practical use. The direction CEO Pat Gelsinger is taking the company seems sensible, in my mind; cloud and software-defined computing and containers are coming, and are the future, and it's good that VMware is getting out on the leading edge of that. But if I'm an admin at the show, and looking for something that will dramatically shape my workday right now, I didn't hear or see much that tickled my itching ears.
Posted by Keith Ward on 09/10/2015 at 9:43 AM0 comments