Server virtualization trends and predictions for 2015

What’s in store for server virtualization next year? This month, we’re asking our Advisory Board members for their 2015 predictions.

Once again, we’re closing in on the end of a year, and it was undoubtedly a busy one for server admins. VMware shook things up in 2014 by announcing its family of hyperconverged products under the EVO name. Cisco and VMware continued to fight for the future of software-defined networking. And, Docker seemed to emerge from nowhere overnight to reinvent containers and collected a long list of prominent partners along the way.

Just as we do every year, this December, we’re asking our SearchServerVirtualization Advisory Board to look back on what happened over the last year and share their predictions for what will make news in 2015.  But, if this year was any indication, we’re likely to have at least a few surprises in 2015.

Brian Kirsch, Milwaukee Area Technical College

With virtualization firmly planted in the data center, and network/storage gaining more ground, the focus will shift more towards the security of those assets. With so many security breaches in 2014, a renewed focus on security will happen as more companies are looking at virtualization of hardware as old hat.

You can see security getting a closer focus today. Even a product as cutting edge as NSX is introducing the concept of micro-segmentation security while companies are still trying to figure out if NSX is the right fit.  DataGravity, with its off-node compliance and security scanning engine is another example of taking the traditional infrastructure and adding an additional security layer.  Look for many more add-ons to existing products and new offerings like these in 2015. 

The converged infrastructure movement will continue to advance and only gain steam in 2015.  While several storage vendors support this movement, they’re doing so with a weary eye as these appliances continue to take a bite out of the SAN market.  Look for more vendors going with the converged approach and thereby reducing infrastructure needs from vendors such as Cisco, EMC and NetApp. 

Speaking of vendors, look for VCE to become a fully-integrated converged business unit within EMC, dropping the VCE name all together.  Cisco is already out of the picture and looks to be moving on with Pure Storage.  It’s the next steps that make Cisco an unknown.  Today the company is one of the lone large vendors not jumping on the converged market, but look for them to make that leap in 2015.  The question will be what hypervisor they want it to ship with?  VMware is not going to be on that short list. However, it is possible we will see Cisco team up with a company like Citrix or Microsoft to go after the 800-pound gorilla known as VMware.

The battles between VMware and Cisco will continue through 2015.  Look for a few more people to wonder if Cisco ACI is software-defined networking or not, while VMware NSX continues to grow its installed base. The cloud will continue to make its mark in business, but in reality, it’s the flexibility, elasticity and scalability of the software-defined data center that will continue to be the business mover in 2015.  Oops, I almost forgot, this will not be the year of VDI. It’s the year of DaaS. … Keep waiting folks, it will be here any time now…  

Maish Saidel-Keesing, Cisco Video Technologies Israel

2014 was an interesting year. I do not want to reminisce, but rather look forward to what I think we will see in 2015.

The all flash array will become the norm. Spinning disk will always have its place, but for the virtualization world, the faster the better.

Converged infrastructure
This makes a lot of sense for virtualization. That is why VMware has gone into the market even though the company continues to emphasize that it is not going into selling hardware. Unfortunately, I do not think that many admins are fully aware of the implications of going with a converged product. This will create "silos of excellence" across the organization, and will make things quite interesting for the future. More vendors will go into this market and we will see some acquisitions as well.

SDN will still be a buzzword but the number of companies actually adopting it will not be as high as many vendors would hope. Overlay networks are not as simple as anyone would make them out to be. Even more important are the organizational changes needed to accommodate such a mindset.

The hype will continue and I think we will see a whole new market develop around this technology. All the vendors will try to sell something or some added value on top of containers.

OpenStack will continue to grow, and I might even go as far as to say it will become the de-facto standard for private cloud infrastructure. Everyone is on the OpenStack bandwagon, and are looking to make as much money as they can along the way.

We all love to look into a crystal and hope that we will be able to predict the future, but none of us are prophet. Even so, these are the trends I think will be some of the major things in the virtualization world in 2015.

Happy holidays!

Christian Mohn, EVRY Consulting

So another year comes to a close, and yet again I get to look back at the predictions made for this year. I think I got VSAN and PernixData FVP pretty much spot on, but then again those were pretty much given. More surprising perhaps is that SanDisk bought Fusion-io, just to bury them somewhere at their headquarters in Milpitas, Calif.

Did we get more cloudy in 2014? I really don’t think so – at least not in the private cloud area. The fact of the matter might just be that most enterprises really aren’t equipped for it - nor do they really see the value of going down that route. One thing is for certain though; Microsoft Azure and Office 365 got some real traction in 2014, so perhaps the hybrid-cloud is the real panacea? 

So, my first prediction for 2015 is that we’ll be seeing more of this. Now there is a real shocker, right? I admit it, easy pickings, but hybrid-cloud solutions are really maturing. Using a public cloud infrastructure to host your off-premises disaster recovery and backup solutions? It’s a good idea that just keeps getting better. And the best part is that it’s really starting to get easy to implement. Isn’t that what you call a win-win?

Another point to make is that I don’t really believe that Hyper-V is Microsoft’s biggest weapon in the war with VMware, it’s Azure. If you really think about it, what has VMware and its dominance with ESXi provided Microsoft with? A megaton of Microsoft Server license sales. The fact that everyone pops up a new server, or 10, each time they need a new application or service provided benefits to Microsoft as well. The only way that Microsoft can make more money off that is by bringing everyone in-house to charge them directly. Guess what Azure does? Forget the vSphere versus Hyper-V debate, that is not where that battle will be won or lost.

Of course, 2015 will continue to be full of container buzz. Will it still be Docker? Probably. Will it replace hypervisors in 2015? No. You still need virtualization, orchestration and configuration management. Simply put, you will not get your containers delivered without transportation. Transportation needs infrastructure. Think containers, ships and trucks. It’s not like you put all you stuff in a container and magically it gets shipped to your new address. Something outside of that container needs to happen for it to end up in the right place, with all your stuff still in it. And of course, we live in a Windows world. Even if Microsoft is adding Docker support in Windows Server, you still need Windows for those applications. That won’t change any time soon, and again, Microsoft benefits even if you run those servers on a non-Hyper-V host.

VMware NSX got a lot of buzz in 2014, but will 2015 be the year that the hoards flock to it, or will it still be hoarded by VMware? At $6000 per socket, you need to sell a lot of licenses to get the initial $1.26 billion investment back. Of course, it’s way too easy to translate an acquisition cost into a number of licenses required to break even, but it’s still fun to play with those numbers. I believe that 2015 will be the year that NSX gets some implementation traction, but perhaps not for the most obvious reasons. Everyone talks about NSX as a networking overlay solution, but I think its strongest selling point at the moment is security. Firewall rules that flow with the VM, regardless of where it is? That’s pretty nifty, and quite a lot better than how VMware vCloud Networking and Security (vCNS) worked. Oh, and the fact that vCNS is slowly just fading away will also solidify NSX as a security solution.

My next prediction is that we will see a VMware EVO:RAIL 2.0, and it will be quite different than v1.0. The hardware offering might not change much, I think that VMware is taking a real hard look at how it’s packaged and licensed and perhaps looking to sweeten the deal. If they do, they will have a gem of a product to complement their SDDC vision. Simply put; It’s too good an opportunity to let slide as is.

Lastly, there is this little thing called EUC, and the standing joke is that it is “the year of VDI.” In 2014, VMware has shown that EUC is no longer something they do as a side gig. Acquiring Airwatch and CloudVolumes shows intent, and so does the high profile hiring strategy. I’m not quite ready to call 2015 “the year of VDI,” but we are getting really close. VDI in itself might be a stopgap, but do not forget that as long as there is a need to provide desktop-environments to end users, this stopgap is required. Not all applications are ready to be delivered in any other fashion, and as long as that is true, this particular stopgap is not really temporary. I guess it depends on your definition of temporary. After all, everything is relative, even the size of stopgaps.

So there it is; 2015 neatly wrapped up. Of course, I’m bound to get most of this completely wrong, but be that as it may, I’ll end with one bonus prediction: In 2015 something new will emerge that promises to change everything, and at the end of 2015 we will all conclude that it didn’t. Funny how that works.

Dig Deeper on Introduction to virtualization and how-tos

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.