VMware, a Falling Giant?
New submitter Lashat writes "According to Ars Technica, 'A new survey seems to show that VMware's iron grip on the enterprise virtualization market is loosening, with 38 percent of businesses planning to switch vendors within the next year due to licensing models and the robustness of competing hypervisors.' What do IT-savvy Slashdotters have to say about moving away from one of the more stable and feature rich VM architectures available?"
Microsoft started offering their own Virtual PC software for free, but it's shit compared to features of VMware products. Granted, it's not really an enterprise product either. But VMware's products will save you lots of headaches, they perform better and offer much more features. It's sad to see companies don't appreciate quality software anymore, because VMware has always produced just that. That has been the trend lately, just like companies are moving towards Google's products just because they are free, even while there are much better products on the market.
VMWare has a proven history. They have set the bar and when potential problems are brought to their attention, they address it. Everybody else is simply a VMWare wannabe.
Management may think they're going to make the switch, but when it comes to actually doing it, it'll prove to cost more in terms of effort than they'll save on licensing. There's a hell of a lot more to VMWare than just the virtualization of servers, and it doesn't take a propeller-head to effectively use the tools. Can the same be said of the alternatives?
.nosig
Bar none, PowerVM still has VMWare beat in most areas that matter, but vSphere 5 is a step in the right direction.
If we're just talking x86, though, I keep hearing that KVM will be the top virtualization solution going forward.
FUNK!
If not, they should be. I mean, if they're going to continue being the best in the business, it makes sense for them to build and run an environment as part of their business which simulates the extremes of the challenges that their clients experience. If they build a competitor to Amazon's EC2 or SliceHost or other systems which make heavy use of virtualization, they could really increase their own bottom line. In fact, there's no reason that they couldn't get big enough to convince those sorts of companies to outsource their infrastructure.
The CB App. What's your 20?
Oracle VM 3.0. Xen based enterprise product with 24x7 global support. Check that!
I come to Slashdot only to read sigs. One you are reading is mine.
We use OpenNebula/KVM here.
Both are free as in speech, I can do live migration, it's easy to manage, etc.
I'm running the whole thing on an NFS share from an AoE storage backend.
100% libre software solution, and it kicks ass.
Good luck vmware.
mov ah, 4ch
int 21h
Why should be pay attention to stories where the headline ends with a question mark?
Coder's Stone: The programming language quick ref for iPad
Sure, VMWare is dominant in business virtualization. It has great features and if you're going to do some server consolidation inside a single facility it makes great sense. So there are 1000's of corps out there invested in it. Now, look at the really big virtualization facilities like RS, Amazon, etc and they're never going to touch it with a 10k foot pole. It has its niche, and as long as that niche remains relevant VMWare will probably dominate it. The real question is whether in 5 years anyone really gives a poop about that market segment anymore. Beyond that what's going to happen with Xen/KVM/etc. It is out there getting hammered on everyday in huge web-scale facilities. At a certain point can VMWare compete with that any more than IIS was able to compete with Apache? IIS is AROUND, and not even irrelevant, but it is still basically a bit player. The same thing is likely to happen in the longer term. Nor do I think HyperV is going to be relevant in the longer term. It might eat VMWare's lunch some day? Yeah, by which time nobody will care.
"Malo periculosam, libertatem quam quietam servitutem." -- Jefferson
Anyone mind extrapolating on what VMware's demerits may be? I've only used their virtualization products on a desktop and they work lovely. Full Linux support (both as the host and the client) and very easy management tools. Getting a vm up and running from an .iso file of Windows was just a few minutes of point and clicking in a well made gtk gui. In my experience, it is a very good and user friendly product.
Football Odds
First of all let me say VMware makes one rock solid hypervisor. Out of all of the hypervisors I have used theirs is the least prone to issues.
VMware has a couple of really BIG problems in their platform.
1. Their management tools are windows centric and so is Virtual Center for that matter
2. Their licensing model is confusing as hell and requires a spreadsheet to figure out what you need without overpaying
3. They have so many products that it gets downright confusing to determine which one works for your purpose.
4. They use "old school" sales tactics that just don't work for more modern companies.
When I am engineering a solution and have a problem to solve I am presented with many challenges to present VMware as a solution. Finding the product that suits our needs, Figuring out what license would suit our needs, getting a quote from the vendor without a lot of harassment after the fact trying to close the sale. Rather than deal with all of that many people have found that the open source projects like KVM and XEN are good enough for their needs. Not to mention the huge numbers of free cloud products such as Openstack that gives you enterprise features "for free". At the end of the day I don't care what product get's used as long as the problem is solved with the minimal amount of budget and effort.
A small startup does not want to deal with legacy software and maintaining licensing and dealing with windows boxes. They do great with the "enterprisey" douchebags with their complex setups that cause more outages than they solve but lean and small companies don't want their stuff.
The reality here is the world is slowly changing. Big monolitic companies are failing because their business models are unsustainable without cheating and people are getting fed up with the cheating. VMware has to answer a question to themselves. Do I want to serve the needs of the dying dinosaur companies or do I want to be in business in 10 years?
Although VMware made no new friends with their v5 licensing debacle, they're still the undisputed technological king-of-the-hill. Microsoft, Citrix, and KVM are slowly catching up, but they're still a ways off on many fronts (DRS, storage DRS, HA, etc). Hyper V (R2 SP1) is just now adding overcommit - a technology that's been in vSphere for years..
Most big entireprise clients are leaving VMware for licensing and cost reasons, not technological. Microsoft is not a small player, so when you can save hundreds of thousands a year of licensing costs for a product that does more or less the same thing (minus the higher end features), there's a compelling arguement to be had. Not to mention, with the v5 licensing debacle, many customers are having to shell out big $$ just to upgrade. VMware softened the blow by re-tooling the licensing after community outrage, but they're still very expensive.
The number of people talking about VirtualBox and VirtualPC in this thread is astounding. We're talking about "enterprise" virtualization here, not keeping some dev boxes on your desktop. I think you need to be talking about Hyper-V and Xen, as well as all the competing VDI solutions.
I really like the functionality, stability, and feature set of VMWare, but due to their licensing change, I'm sticking with vSphere 4 for now -- I'd have to buy too many new licenses to move to vSphere 5 because of the amount of memory we have.
We're evaluating HyperV now and may end up with a split cluster - HyperV for our Windows servers and VMware (or maybe even Xen) for the Linux side.
We've already started the migration to Redhat's KVM. Our testing environment has been completed. We beginning the rollout for Production now.
Why switch? The price.
VMware took a nosedive for us when their Linux support began to suck. When VMware Server 1.x and Ubuntu 9.04LTS was around, I could compile the vmmon and vmxnet modules and the virtualization software worked very well. When I upgraded to 10.04 and all the way to current, VMware Server 1.x refused to compile despite the patches offered by VMware. I tried using other distros as well but all resulted in failure.
Vmware Server 2.0 was such a huge pile of crap, I tried it once and found that it was all web-based and ditched it. Even if the networking worked properly (which it STILL didn't), the fact that I couldn't use the VMware Server Console except through a web browser was beyond irritating and forced me to look elsewhere.
I now use VirtualBox with the Extensions pack and can access my VM consoles via MSRDP and it doesn't use a web browser at all. I can build, start, and stop vms through an SSH session with no difficulties. The only downside is that while VMware Server would allow you to build networks of VMs tied to multiple Virtual Networks that stayed on the host but could be bridged or natted to the outside world, VirtualBox only gives you the same (Host Only, NAT and Bridged) three options, but if you have two VMs running together in any mode but Bridged, they cannot communicate with each other.
This appears to be a Linux Only issue, as in Windows, the VMware networking works exactly as it is expected to which lead us to believe that VMware just doesn't care about their Linux users and would prefer to sell their other products rather than attempt support of their existing lines of supposed Linux-compatible VM server products.
I recently built a new server and decided to go with Xen using HVM. Although this is mainly a personal project, I can't really see the point in purchasing closed source software when the open source alternatives support everything I need and more. I admittedly am pretty new to this game but what advantages can I get from VMware when Xen runs a VPS totally seamlessly?
I'm a fan of VMware but clearly any competition - and specifically, the more robust the competition gets the better the pricing models will be for the end users.
Right now we have very little if any issues with our virtual infrastructure - although View could use some work (we're still on 4.0 though...). VMware's support is excellent. Their tools are excellent. Their online documentation is excellent. Other than $ there wouldn't be a lot of motivation to start shopping for another vendor anytime soon.
But I also trust and expect that VMware will continue to be cutting edge in the VM sphere since it's the focus of their business (vs. Microsoft - who's focus is who the hell knows these days).
D
Expensive, bloated and chock full of unnecessary middleware and abstraction layers? That's usually what "enterprise software" means.
So, which features are these that Virtualbox provides and are in the most expensive package for vmware? I was under the impression that everything virtualbox does vmware does for free as well, and the cost only comes in should you want the infrastructure stuff... which is completely absent from virtualbox.
For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
After having a look at the PDF I wonder which businesses they did ask. First, most large companies that I know of run their servers with Linux, no one would even dare to suggest a MS hypervisor. Second, the hypervisors that I've seen in the wild are (apart from Citrix and some VMWare hosts) mostly OpenVZ, Virtuozzo or Xen. Just think of all those root v-servers you can rent for cheap. Xem is big in companies and backed by major players, e.g. IBM. The survey numbers just don't make sense.
Computer simulation made easy -- LibGeoDecomp
With production enterprise experience with Xen, HyperV, and VMWare hosting linux VMs.
It's still VMware, just based on some of the showstopping issues encountered with Xen and HyperV.
It's only a matter of time until VMWare competitors catch up, which is good for all of us, however based on my personal experience VMWare is still my preference.
Note, environments vary. Just based on environments I've worked with.
Awesome!
> What do IT-savvy Slashdotters have to say about moving away from one of the more stable and feature rich VM architectures available?"
Um, how about "over my dead body"?
If you want free stuff like VirtualBox or VirtualPC, more power to you. It helps push the envelope and provides for competition.
For large enterprise installations, there is VMWare. I'm sure that won't always be the case, but for now, you'll have to pry my vCenter from my cold dead hands.
Oliver's law of assumed responsibility: If you're seen fixing it, you will be blamed for breaking it.
Two years ago, VMware was the only serious player in the enterprise hypervisor market. It could demand a price premium and had weight with other software platform vendors to demand support.
Now, with Hyper-V being somewhat more mature and with the Xen product line, VMware is falling into a competitive market. Naturally, there will be an erosion of market share in that case.
The bigger threat that faces VMware is the same threat that faced Netscape in the 1990s-- VMware is a competitor of Hyper-V which Microsoft can include "for free" in its server operating systems. And Microsoft still has the same monopoly influence over the major hardware vendors (to discourage pre-installs or reseller agreements). And it can control the licensing for its operating systems to inconvenience VMware customers (you have to buy a separate license for each ESX VM, but if you run on Hyper-V you get 10 VM licenses for free) and/or control its support of its enterprise application stack (We'll only support Exchange/Sharepoint/SQL Server/Link/IIS/whatever if it's running on Hyper-V. If it's ESX, please reproduce the problem on physical hardware to make sure it's not an ESX issue)
--Joe
I had considered VMWare for our virtualization needs but the licensing was a nightmare. Complex and very expensive. It seemed like the price was alright up front but then figure in teh costs for everything else you need to manage it. This is a big part of what pushed us to Hyper-V. Licensing is simple and affordable. I can't really complain much about Hyper-V, it suits our needs for a reasonable cost and I've had no major issues with it. I don't see the benefit to switching to VMWare, for us. We are smaller than most companies that use ESX though, so that may be worth considering. I do not have any extensive experience with ESX though do have some experience with VMWare on top of Windows.
I work for a major *.edu -- we use VMware, we pay their exorbitant pricing, and subsequently get threats of additional fees for not renewing support on time (an amusing tactic). We don't really find ourselves using the fancy feature sets. In fact, a large part of our *.edu is going KVM -- probably for similar reasons, more likely pricing. As others have said, VMware continues to change their licensing models -- it ends up being nickle-and-diming for features, where I'd just rather pay one flat price and just be able to use the entire product. KVM/Xen, et al, are still being actively developed and hammered on. I don't see any reason why those products couldn't eventually, significantly supplant VMware in areas such as mine where they get the job done effectively. If we want support for KVM/Xen, we can pay for that... much less money in the end.
Libvirt and the improvements to KVM plus Xen getting mainlined (is that the right term?) has to be hurting VMWare. Rackspace, along with a lot of major players, are spearheading OpenStack which ought to be a major open source enterprise player when it matures. Also, cloudstack recently went 100% open source which puts even more pressure on VMWare.
Also, projects like OpenVSwitch are putting major pressure on the proprietary vendors too.
I came to the datacenter drunk with a fake ID, don't you want to be just like me?
Plain and simple, VMWare is pricey. I'd love to run them where I work, but it's extraordinarily expensive compared to Xen and Hyper-V.
Hyper-V is about 5 years behind and XenServer is about 3 years behind in terms of functionality and stability, mainly due to the fact that VMWare has been doing it for so long. VMWare is rock-solid and feature rich, and I'd love to use them. Currently we use XenServer, but with Citrix recently closing down their hardware API's and not playing nicely with anyone it looks like it is going to be the first casualty. I've been very upset by XenServer's HA so far, plain and simple it has sucked. I've had hosts reboot from crashes and the virtual machines go down, but the host thinks it has the machines and all of the other hosts think it has the machines. I've done everything XenServer has asked (HA quorum on a separate LUN, patches, etc), but it still just sucks. I've yet to see a host fail and the machines to go elsewhere, and the configuration is absolutely right and has been reviewed by Citrix. Maybe 6.0 will be better, but I just heard of major issues today with it. Hyper-V is really where the competition is going to come from, especially with how engrained it is in everything coming up. Want to run Exchange 2010 SP2? Recommendation is Hyper-V virtual machines.
God I miss VMWare.
I like ESXi for the handful of random non-production systems I use. I just don't buy that VM is the right direction for every company as a primary platform. Sure, small scale VM has it's benefits, but in a large scale scenario the overhead and vendor lock in becomes short sighted. Yes, eventually with enough VM in your datacenter, you'll save money, but at what long term expense? What's that vendor proprietary solution going to do for you in 10 years when you want to move to the next big thing? I say build out your DC using commodity hardware and design your applications and network with fault tolerance and efficiency in mind. Need a more efficient footprint? Try microservers http://www.dell.com/us/business/p/poweredge-c5125/pd 1. Focus on getting the best bang for your buck with commodity hardware 2. Focus your people on streamlining operations for this model (instead of focusing on how to integrate VM with existing models, etc) 3. Design your applications/architecture around not having some magical single box with a thousand mac addresses that can move around the data center on a whim. Who would be dumb enough to believe in this model? Google and Facebook, for starters.
Forgive them... It's what most people get exposed to. After all you can quickly run that on your desktop to do cool stuff. Try setting up a Xen Server at home and run a few instances. Not that it's hard (I've done it on an old Athlon MP 2400+/4GB RAM as I got exposed to Xen at work and wanted to look deeper into it), but it's far from typical desktop use. Add in "weird" (for the desktop world) hardware like fibrechannel SANs, etc, and the population who have gotten exposure to enterprise level virtualization dwindles.
Ahhh...the great dumpster continuum. Many a free computer will be found there. -- sowth (748135)
Dear VMware,
I started out with you at Workstation version2. Since that time, I have never seen anyone really do Virtualisation as well as you did. Ever. Workstation is still sold with new features, and some nice pricing. I played and ran ESXI 3-5 and at each step its been accompanied with a rising tide of pain. No matter how brilliant a product is, if you start throwing in silly licensing and serious costs - and stupid complexity (in licensing) - well - you get what VMware is right now.
And the worst part is if you talk with their low and middle staff, they KNOW they still have the tech, and the cool. They also nod each time you state the obvious thing they can't fix. I have no idea who the board are at VMware. I only know they have the best product, a bleeding edge product, and that they have started the process thats going to actually kill it. Being the best isn't actually relevant. Being the best with a fair and sensible model means people will use you - and not lose you.
Right now, there are only two types of VMware (enterprise level) customers. Those who are paying with eyes watering and teeth grinding, and those who are at least looking seriously at moving away.
And I speak as someone who has serious love of VMware stuff, and they've reached a stage where they are so arrogant they don't even talk to me now. So I guess thats why HyperV sits in my racks these days *despite* being lesser to me.
The problem with being the best, and getting too serious a dose of arrogance, is that come the fall, there is no way back.
I'd really like them to get back to ESXI being the foot in the door brilliance it once was, and to having a sensible curve upwards in cost that people could look at and say its great, "what if we grow?" Now it just seems like growth? Haaa, pay us a lot of $$.
Its still the best virt stuff I have used. Period. But the gap between it and other stuff that works pretty damn well is smaller than its ever been. So they need to wake the fuck up and get real.
We`re all equal
Both of your points are incorrect. Virtualbox does do far more for free on it's *open-source* product, and if you need the infrastructure support you can purchase it from Oracle. Their current product is called Oracle Virtual Desktop Infrastructure 3.2.
brandelf -t FreeBSD
I'm glad Hyper-V, Xen, and to a lesser extent, RHEV are providing some legitimate competition to VMware--and typically at lower software licensing costs. But, the hurdles to adopting these competitors are high: sparse ISV support, less rich ecosystem of 3rd party tools (backup & recovery, capacity planning, etc.), and existing investment in VMware licenses, training, SOPs, etc.
VMware is no longer the only game in town for enterprise virtualization, but their position is firmly entrenched for at least the next 3 years. Switching costs for environments of any substantial size are just too high compared to the licensing cost premium VMware demands.
Enterprise Solution - Solvent used for dissolving piles of cash in corporate vaults.
ESXi 5.0 has native driers for Realtek 8111/8168 and many more consumer-level devices, just so you know.
until they started to look and sound like oracle.
Good people go to bed earlier.
VMware's original edge was their code patching hack which allowed visualization on hardware that didn't really support it. Once x86 machines got some virtualization hardware support, that hack was no longer needed, and anybody could write a hypervisor.
Now most of the virtualization issues are more about systems for managing instances of virtual machines. The hypervisor itself is a small part of the overall product.
So 38% of virtualization customers are planning on switching, but 2/3 of all virtualization customers are VMWare with the other 1/3 being somebody else. There's a lot of floating data points here.
We can come up with lots of fun theories...
Maybe VMWare numbers will drop to 30% of the market and those will all get sucked up by Hyper-V.
Maybe everyone using Hyper-V thinks it blows and are going to Xen, leaving 22% of the original survey to allocate to leaving either VMWare and Xen for one of the two they're not using.
Connect the dots any way you want, but without knowing which camps the answers come from, this is a non-story. You probably see a lot of churn in the minds of decision makers, but nothing gets done anyway once they get to planning strategies, or crunching the numbers, as another commenter here said.
----- - The beatings will continue until morale improves
Generally, I agree... but VirtualBox is only a few features away from competing.
At *HOME*, I run VBox on my OpenSolaris box... four SATA drives in Raid-Z2... phpVirtualBox to manage...
the new phpVB allows allocating memory to the host, which allows me to block out 3gb for ZFS (I found that loading VMs over ZFS memory would cause crashes... set arcmax and VB limit, no crashes)... and it's gaining features fairly quickly.
Given a little bit of fiddling (which would need to be resolved for it to be "enterprise ready"), the server has been quite reliable.
That said... there is NO integration with host shutdown/startup... which would be critical... and the live migration support is not yet easily exposed... plus the host could use a lot of improvements to manage the network... but... I think with a little bit of focus by the VirtualBox team, they could really start competing here.
I stopped caring about vmware when their marketing people got editing rights over their technical documentation. Seriously, go to their website, try to figure out what product will work for you (if you can make it through the marketing drivel) then look at the documentation for it.
Technical documentation is supposed to be .. technical. If you screw over the IT admins managing your product they're going to search for alternatives.
Just because you disagree doesn't mean it's not true.
People don't care what software runs their virtual machines. The very target market for this is people who don't want to commit to platforms.
VMWare doesn't do anything that is substantially proprietary.
If something comes along which does the same things as VMWare, and reads VMWare disks, but is better in some way than VMWare (for instance by being free for all the freetard masses), it's bye bye.
Once, the selling point of VMWare was "cool, you can virtualize a machine". That is a commodity now. The selling point now is the enterprise stuff. That's it.
What I've noticed with corporate IT is that they look into many options and kind of keep a tab on them, so a larger team would be looking into vmware alternatives as they await .net 5 & server 2012 or something. It's good business practice to weigh your options, but say a decision is made to look into a migration path, at that point licensing, man hours, and risk factor are weighed in and often times the project is shown unfeasible and never takes off the ground. Thereby a survey is not the best way to gauge who will be switching from what to what, an analysis of features tied to business logic is, ex. what can x feature do for us that costs x money, in the case of migrating hypervizors that's a tough one to answer, the x feature is lower cost of licensing, but what's the trade off? Saving 10k a year isn't worth the man hours to migrate a large IT network onto another hypervizor. Downtime and hardware need to be taken into consideration too, and don't forget compatibility issues... It gets kind of hairy real quick, so while I believe that the results in the article are exactly as the survey suggests, that's where the accuracy ends and the real world begins.
We use Ganeti + KVM at the OSUOSL with a completely open source stack. We can do live migration, fast deployment, easy horizontal scaling, and easy management. We haven't had any major issues with Ganeti itself only a few minor issues with KVM. Performance on KVM is still lagging compared to Xen but for what we use it for we've had no problems.
It's safe to assume OP can't afford enterprise either way, the tag starts at $xxxxx , so it's price point limits it to the consumer market. Then again there's not a whole lot wrong with esxi.... but on a stand alone exsi server, zen server might surpase it, on a vcenter environment (the enterprise class), the tables quickly being to turn with features such as vmotion working CORRECTLY, and such.
On that note knowing what I know about computers, I would recommend it to clients, but never for any venture I ever undertake lol, the licensing is too nasty.
In this case it means 1 feature you really want and 9 you don't per 10 features kind of mix.
Any idea what middleware or abstraction layers actually are OP???
VMware has a free tool called vCenter Converter that will do the migration, at which point you can use anything that understands VMware vms.
Awesome.... ....... Giggdy!
Gravity!... It's not just a good idea... It's the Law!
IT: VMWare is the best! We need to use it for all our VM needs.
Boss: It's to expensive. I read that Windows does the same thing for free.
IT: No, it only does 10% of the stuff VMWare does. We need VMWare!
Boss: Glad we agree! Start implementing Windows Virtual Server tomorrow....
That submitter Lashat is shilling for EMC.
I've been a VMware customer since 1999, and I must count myself among those disappointed by recent releases and pricing changes. Parallels, Microsoft, Citrix, and Oracle all have competitive offerings, at least two of which are substantially free software. If we hadn't invested so much time and energy into VMware at work, I'd seriously consider switching to HyperV or Xen.
I'm proud of my Northern Tibetian Heritage
Vmware+Veeam+Offsite = Perfection.
When the foot seeks the place of the head, the line is crossed. Know your place. Keep your place. Be a shoe.
And libvirt is, IMHO, the library that will be used for managing VMs going forward. It's virtual platform agnostic (right now it's Xen and KVM, but I think it will expand in the future), and has bindings for many languages.
If I mod you up, it doesn't necessarily mean I agree with what you've said, sorry.
This is a video poking fun at VMware, thought I'd share.
What do I know, I'm just an idiot, right?
Generally, I agree... but VirtualBox is only a few features away from competing.
At *HOME*, I run VBox on my OpenSolaris box... four SATA drives in Raid-Z2... phpVirtualBox to manage...
I admire your pluck. And I cried when it left, but OpenSolaris is dead. Long live OpenIndiana.
The Admin and the Engineer
Shortsighted CFOs and the like probably think, 'ya we are totally going to switch to product $X' without considering the fact that the true cost of moving a large infrastructure might be more than they expect.
More servers will be required: It should be obvious that you'll need a home for the new VMs without first decom'ing the existing VMware environment. Maybe this is done with a server refresh at the same time and becomes a non-issue.
More storage will be required: Converting a VM from VMDK to VHD is not a zero-sum process. If you have 50TB of VM datastore requirements you'll need at minimum 20% over that to act as swing space. Thats assuming you can devise a strategy to swing storage like that. It also assumes you dont have massive array-based replication requirements that would force a resync requiring you to either bring the target equipment back locally (IF its not also serving production data) or hammering a WAN circuit for a few days/weeks/months
Retraining will be required: For VMware admins, troubleshooting Xen and its great expanse of 32character UUIDs (ex: 83750d2d-dd36-787c-da1c-951d7afc2cf0) is frustrating as hell and is a great detachment from normal administration through direct console or vMA. Hyper-V is probably ok as long as you already have System Center to tack on SCVMM and are comfortable with running MSCS. As much as the world complains about the instability of MS products its amusing to watch people proclaim using NT6.1_x64 as the best choice for a hypervisor. Hyper-V with Server 8 also seems like it might recommend running VHDs against (SMB2.0/2.1?) CIFS shares. Anybody want to sign up for that?
Time will be required: I hope that everyone who stated they will be moving 'next year' have already planned and PoC'd running at least some of their gear on these other products. Any time invested in moving platforms is money with salaries and potentially consultants.
Environments actually using benefits of vSphere EP+ will also have to tolerate a massive amount of feature loss. Nobody running Nexus 1000Vs or a large DVS environment will probably be able to successfully migrate. Not to mention losing auto-deploy, fault tolerance, storage vMotion/DRS, etc and on and on.
I'd take that 38% with about as much credit as 'here comes the next iPod killer' or 'linux is ready for the desktop'
Someone else clued me in. I've never heard of that product until today. Thanks, though.
Work with me here - imagining that didn't exist, what does Virtualbox do more than vmware? I'm ignoring the licensing. I far prefer Virtualbox because of that, so if you can enlighten me I'll happily drop vmware.
Quick note:I use virtualbox for all my personal stuff, because I don't personally do anything beyond "desktop" visualization. I think I can be excused for not knowing that I had a choice, here.
For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
What we're seeing here is simply that RHEV, Hyper-V, etc are becoming 'good enough' for many many customers that have until now had no other choice than VMware. But really, on the ENTERPRISE level, VMware is The Leader. It's leaps and bounds ahead of anyone else. Live data migration for starters is a lovely killer feature. In terms of usability, especially in very large environments (1000s of VMs) there is NOT ONE feasible competitor with the management features VMware provides. RHEV is just about there, but for many enterprises it's a bit too new. Everything else is crap, frankly. And even many "enterprise" solutions like AIX LPARs and Solaris LDOMs, completely lack a proper management interface, making living with hundreds/thousands of VMs a very different affair compared to vSphere or RHEV-M.
I use VirtualBox at home, and it's always been just short of annoying. The emulated sound card is never recognized, so I always have to track down the link to the Taiwanese site that has the drivers, it emulates some crappy Ethernet controller (Broadcomm?), and accessing resources on the host (shared directories or USB ports) has got to be one of the least intuitive tasks around. Once I've got everything set up it seems to work pretty well (Windows 7 seemed faster under VirtualBox on my 4-core Linux box than natively on my dual-core work machine), but I've actually put off playing with a couple of OS projects because I figure it'll be a two-hour chore to get the VM set up.
It's pretty impressive for free software, but I wouldn't call it better than VMWare.
Just junk food for thought...
I've been using Proxmox for a few years in production systems. It's an OpenVZ/KVM solution wrapped in a nice web interface. Supports live migration/clustering/automatic backups and a bunch of other nice enterprise features. All free and open source of course.
I thought the Enterprise used a duatronic hypervisor.
Just kidding.
VirtualBox can be used for more than just desktop VMs but you are correct that it really isn't competitive with VMWare.
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
Sell It, get out, create something new, just like Steve Jobs said....
Let us not forget things like KVM+ovirt
VMWare supports live replication. Why just migrate when you can have automatic, near zero downtime fail over with live replication?
Good luck, VMWare? Hehe, you seriously overestimate the live migration feature. Or rather, it is no longer that special. Even HyperV has live migration. The difference is, VMWare does it VERY well. And I've yet to see anyone come close to the live replication that VMWare does.
Don't be offended by this post. I just don't think you're talking real business. Maybe it's "kick ass" for a home setup or a 3 person office.
Recent Experience
We tried Hyper-V for 6 months, and it was the most god awful unstable piece of crap I've ever worked with. A brand new IBM x3650 m3 running 12 cores crashed on a weekly basis and corrupted its main RAID running Windows Server 2008 R2. I think we can all agree that Microsoft virtualization, be it VirtualPC or HyperV is just absolute shit.
We've since switched to vSphere/ESXi, and haven't had a single crash. Everything is running fast and stable. I can live migrate any of the machines. Should a disaster happen, I can bring up another ESXi machine in on any other server or replacement hard drives (should our hot swap drives also fail) in about 5 minutes. Time is money on my network, and I don't have time to screw around. HyperV is not a 5 minute install, and I doubt your solution is either.
I can use VMPlayer instead of VirtualPC which is also free as in beer, not speech. ESXi is also free as in beer, though it is well worth the license for more features. And yes, it does live migrations. Unlike MS solutions, VMWare supports installs on Linux. I rather run Windows 7 in VMPlayer on Ubuntu, than run Ubuntu on Windows 7.
Being in a major production environment where every minute of downtime is a lost customer, I can't play around with anyone that doesn't have major support. That pretty much means Microsoft, Citrix, or VMWare. And of those three, I've had the best luck with VMWare. I've even heard Citrix is good, especially in VDI.
I don't think VMWare is in any need of luck. I know of no serious business looking for 4th party solutions for their major production servers. As someone else said, maybe for a VERY small shop, maybe for a dev box, but not for Enterprise.
Setup
What is your setup and why in Spagetti Monster's name are you running VM over NFS ethernet? Either go RAID or fiber to a SAN. That does not sound like a good Enterprise setup. Are we talking 10GB over a dedicated line (doable) or 1GB over switch (WTF)? If it's the latter, I bet you have a hell of a write to disk latency problem if you run more than 1 production VM. Personally, I'd go fiber SAN if RAID wasn't an option, but then again, I run 12 VMs on 2 64bit servers alongside a small array of dedicated servers. I'd laugh an AoE based VM proposal right out of my office.
I'm just taking a wild guess you aren't running a rack, or have network intensive users/applications. In my situation, I have to deal with about 80 workstations, a remote office, and 4 internet data connections on top of the servers themselves, including a 13TB file server and 20TB of backup storage. I can eat bandwidth with the best of them. When you can no longer count your routers and switches on your fingers, then I'd more willing to listen.
So, I don't know who you are posting to, because it's not really SMB or Enterprise.
I8-D
VMware is still the best product for hardware virtualization, and they do have some great home/consumer products. The biggest issue right now is their change in licensing to start charging you for how much RAM you plan to use. For a lot of customers, this may not make a difference, but at the enterprise level its definitely a nuisance and causing irritation. Will that irritation be enough to change vendors? I don't think so although managers may want it.
I will shred my adversaries. Pull their eyes out just enough to turn them towards their mewing, mutilated faces. Illyria
We're talking about VMWare. A few dev boxes on your desktop absolutely is relevant, because that's what many of their corporate licensees use it for.
I am trolling
DESQView/X!
1311393600 - Back to Black
I like VMware Workstations a lot. VirtualBox isn't bad in Linux/Debian, but it has issues like:
1. Missing drag and drop between host and guest.
2. Unable to exit VirtualBox after pausing the guest session.
3. Windows key sometimes don't work or break keyboard navigation.
Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
I was the ultimate fan of VMWare, however, with the way the competition has been improving their business models to grab at the corporate sector, if VMWare wants to stay in the race, they will have to start changing their thinking a bit. Just like Microsoft, who wants to listen to no one but themselves when it comes to making things that play nice with others.... so too should VMWare take a page out of their bad decisions play book, and learn from it.
VMWare is one of the most powerful apps out there for virtualisation, however, with the out going market and time tables for setting up environments that are out of the box....you pay way to much for those skill sets. Either make learning them easy enough that anyone can start getting certified, that way admins are a dime a dozen and lowering the cost at that end, or lower the price of the product add on that usually rack you up in the tens of thousands easily just to access your own info.
I liked hearing about their move to push an android friendly environment to have 2 simultaneous phones running off of 1 device....this would allow someone who needs many separate lines to still have them all separate, but on the same device...however, that is not the corporate market, so what about companies that want to save some bucks, what can be done for them, so they do not leave your client base.
That's what the systems folks here settled on. My personal attitude is if you have to emulate a whole machine in order to keep one service from stepping on another, you are probably using the wrong OS, or not using the OS you have effectively. Not that you have a choice of OSes most of the time -- you have to run the product thePHB went out and bought, so you have to run it on the OS it supports. But personally, I haven't fired up a VM for my own use in over a decade, unless you count WINE.
Someone had to do it.
You :
/whatever) installs and configures your app on your versioned test VM from scratch,
Some of those steps are already done for you if you have pre built software. It gives you:
Stop thinking development ends when the source builds to an executable. Today we develop services not applications. That means you should be able to build all the way to a running service, preferably to a versioned file which can provide that service when it is booted.
Deleted
I didn't pay too much attention to HyperV until I started playing around with Team Foundation Server 2010. You might ask, what does an ALM tool have to do with virtualizing servers? Well, one things we found really helpful when doing automated testing, was this...you use the lab manager to spin up all your testing environments in hyperV then when a defect is encountered, the developer can spin up that environment on their laptop and actually debug defects on the server to make sure they have the most accurate test case. This eliminates the issue of "Works on laptop, don't work in test environment".
Besides that, Microsoft's virtual stuff is pretty decent, especially when it ties in their other products.
I have had cause to deal with this both on the server and the desktop.
tl;dr VMware works, and everything else is shit.
This pains me. VMware uses the pricing structure "how much have you got? yes, a bit more please thanks." They can do this because their VMs are not shit. They are ridiculously better than anything else.
VirtualBox in particular is just incompetent. It can more or less run Windows or Linux. As long as you don't want to do anything fancy, e.g. USB.
As a sysadmin, I would tremendously welcome something that cost less than VMware but didn't suck. No such thing presently exists.
For your home Linux box, use VMware Player if you value your sanity.
http://rocknerd.co.uk
...
Or actually, the real one is xVM, the VDI is for virtualizing desktop users, not server infrastructure.
VMware also does VDI.
Both Sun/Oracle and VMware VDI require the xVM or vSphere to actually accomplish their goals.
Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
As a system administrator that manages a VMware Environment for ALL employees workstations (Over 1600) as well as ALL Windows servers (Over 300), I am very happy with the vSphere product that VMware has created. They have embraced the community and provided great tools so that I can use common languages to script VDI deployments (Using Powershell with the PowerCLI addon from VMware) as well as simplify disaster recovery efforts. VMware's View product has allowed around 30% of our employees to move to work at home status without a noticeable difference in performance over being in the office. I firmly believe that VMware remains the top vendor for virtualization.
Note: I do not work for VMware, but rather a large healthcare organization that has recently achieved a 100% VDI environment for all employees.
I don't think you know what enterprise means.
There are a huge number of replies to this story already; and, having read most of them, it's obvious the GP isn't alone in not knowing what "enterprise" means.
#DeleteChrome
VMWare - $5000 for any cert. Classes are mandatory for certification, must be renewed as well
MS Virtualization - No class required, $125 for a cert test at Volt/Sylvan/whatever, classes are available dirt cheap at your local community college most likely
Not including all of the other solutions and licensing issues, that right there shows why VMWare will lose share. Certs matter from the techs perspective and a hiring perspective. Companies are going to hire people to implement and maintain these solutions and possibly pay to keep them current, so which are they going to go for, hmm?
We're Vmware shop for three years now and plan on staying that way for at least three more. WIth our current hardware (IBM, DS8100, Bladecenters...) and location (Europe, Eastern) this is only good option considering we need 24/7 reliability. In the past three years we had no issues whatsoever and any other solution comes very close to Vmware in terms of cost. So... we look at others, but have no reason to move.
Name one OS that is so secure that using separate machines isn't a good idea. You're clearly not refering to any modern popular/widely used OS since all of them out of the box are setup in such a way that exploiting one service gets you something good to use against another service (perhaps not what you need every time).
I'm not saying you should be using VMs to mitigate the problem, because you're actually making it worse by doing so.
I am however saying that if you think you know of some OS that is so super secure that physical service segregation is no more secure than running them all on one box, then you're just ignorant of reality.
Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
Survey sponsored by Oracle or Microsoft?
Virtualization is a great solution for running shit operating systems.
With things like LXC, VServer and OpenVZ, there is absolutely no scenario when virtualization is a valid solution in a production environment, unless you have to run Windows-only software (and then this is your problem).
Contrary to the popular belief, there indeed is no God.
Your information seems out of date. One issue with the sound, was that neither 64-bit Windows Vista or either version of Windows 7 contains AC97 drivers by default. This has since been mitigated with the addition of the emulated Intel HD Audio card (and selected by default on Windows Vista/7 guests at creation time, so you have to go out of your way to make sound not work). Ethernet is handled by emulating one of two AMD cards or three Intel cards, which guests support what also varies, and VirtualBox is smart enough to pick the appropriate one at VM creation time (assuming you aren't lying about the guest OS type, of course ;)). Shared directories and USB devices, yes they were a pain in the VirtualBox 1.x days, but they have both become quite a breeze (even shared folders from a command-line only Linux guest is as simple as "mount -t vboxsf ShareName /mnt").
Disclaimer: I use Linux as a host operating system and don't have much experience with VirtualBox on other hosts, be it Windows, Mac OS, or Solaris. If it's still troublesome on one of the others, all I can say is "Oh well" and hope the situation improves for you.
that pretty much covers it
My ism, it's full of beliefs.
VMWare but be the leader in hardware virtualization right now, but hardware virtualization is a hack to bridge the gap between current software and the cloud concept. Why virtualize the hardware when your software natively supports a cloud model? Google has no need for virtualizing hardware because their software understands a cloud architecture. With Azure, microsoft is moving in this direction and when that happens VMware will be the leader in a technology with limited usefulness.
I work at a non-profit, and we went with Hyper-V about three years ago because of the licensing. Microsoft almost gives away their products to non-profits, and you still get support. For us, it was a no-brainer. I've worked with VMWare also, and I don't really feel we're missing anything. I have an assortment of Windows, Ubuntu, and CentOS servers, and everything works the way it's supposed to for us. Just my two cents.
A fresh start-up in the Silicon Valley, sporting a CEO who also happens to be the creator of Xen, and whose new virtualization products are designed to profoundly enhance security from the metal up... and nobody mentions it? Wow. I'm a little shocked!
Oh, and if I were Vmware, I'd be a little scared too...
I work for a managed service provider, our team deploy enterprise grade VMware, HyperV and Xen solutions to various customers based on need/religion. The unanimous opinion is that nothing compares to VMware. HyperV is shit. Xen is shit. ESX simply works better than everything else, has more features, and being the incumbent has the best support options (everyone knows and uses it so googling solutions is quite trivial).
Enterprise Solution - Solvent used for dissolving piles of cash in corporate vaults.
LOL. That is absolutely true at least for the company I work for.
One thing I've noticed though, is that where the 'enterprise solution' comes from entirely depends upon the decision by corporate to buy it to dissolve excess cash on hand. For some reason they don't seem to bat an eye at spending millions of dollars for any solution that carries the Microsoft brand, even though it inevitably is less interoperable than alternatives, but they balk at paying licences for Redhat or Suse products, regardless of the actual cost. Maybe Redhat, Suse and other vendors need to add some kind of strawberry flavoring to their enterprise solution.
This is an ex-parrot!
OpenVZ is not the same as the rest. It takes the host O/S and creates skinny containers from it that appear as individual systems; but that's not the same as "virtualization" in this context. You can't run an alternate O/S using OpenVZ.