VMware To Join OpenStack Foundation

hypnosec writes "OpenStack Foundation, backed by virtualization players like Rackspace, Red Hat and IBM, is going to get a unexpected new member – VMware. According to a post on the OpenStack Foundation Wiki, the agenda of the Board of Directors meeting on August 28 includes the Gold membership of VMware as one of the topics. VMware is not the only one applying for Gold membership as Intel and NEC are also standing in line for their memberships as well."

Comments

Political mess?

[Anonymous Coward]

With so many conflicting interests how long until the project starts spinning in place?

Will VMWare donate their patents?

[Picass0]

VMWare's membership is a great opportunity to plant poison pills that can be later exploited to shut down any development originating from this partnership. I would hope Openstack has obtained usage rights in exchange for their membership.

Re:Will VMWare donate their patents?

[Rogerborg]

I dunno, do you reckon they're retards?

Look at the state of patents. Look at Apple just being gifted a billion dollars and change for "inventing" rectangles with rounded corners. Even those who rail against patents are busily engaged in stockpiling "defensive" ones.

So no, I'm guessing that no sane business will lay aside their patents, not until the rules of the game are changed.

Re:Will VMWare donate their patents?

[Picass0]

I 'reckon' VMWare is well aware of Apple, and as my question suggests last weeks court decision is in my mind as well. Unless VMWare offers indemnification to Openstack Foundation they are a fox in the hen house.

Openstack should secure usage rights in perpetuity, else VMWare can make strategic suggestions that steers development towards their patent portfolio. Years from now VMWare might decide the products of Openstack Foundation represent an obstacle to be eliminated.

Re:Will VMWare donate their patents?

[postbigbang]

I think you're paranoid on a good day.

VMware is late to the party, and is getting their clock cleaned by the competition. This why they came to the table: to eat lunch, like the rest of them. VMware's roots are open source, although they charge dearly for their proprietary stuff. Nonetheless, the industry points towards stack-related tool kits that spin up hardware of near-legendary value. You don't think they don't want a piece of that pie? Loads of it's FOSS and loads of it's Ruby, Ajax, and stuff that almost predates them going back to thoughts of Xen and LVMs.

So, no, don't worry. This is about being left behind, not total annihilation an destruction.

Re:Will VMWare donate their patents?

[Picass0]

>> "I think you're paranoid on a good day."

Are you following me?

Re:Will VMWare donate their patents?

[Anonymous Coward]

You could just do the research: the Openstack contributor's licence agreement (individual and corporate; an employee working for their company is expected to be covered by both before their contribution will be accepted) seems to include all the necessary boilerplate.

An interesting commentary

[thePowerOfGrayskull]

An interesting and relevant commentary on OpenStack;

https://gist.github.com/3456841

Re:An interesting commentary

[vlm]

He misses the point that in

How do you commercialize

the easiest way to commercialize is to hand hold and support and project manage.

Selling a proprietary virtualization empire is, in the long run, about as likely to succeed as writing a text editor, so going all "IBM" and moving into services might be a pretty wise move for vmware.

Re:An interesting commentary

[LordLimecat]

Selling a proprietary virtualization empire is, in the long run, about as likely to succeed as writing a text editor

You mean like Microsoft Office, which at its core has a glorified text editor that is one of the cores of Microsoft's profitability? Or how about an OS (for anyone whos read "In the Beginning was the Command Line")?

If VMware can keep up innovation, and can fix some of its licensing issues, I dont see why they could not have a long future in selling a "virtualization empire".

Re:An interesting commentary

[VortexCortex]

Selling a proprietary virtualization empire is, in the long run, about as likely to succeed as writing a text editor

Point proven. Text editors being successfully embedded into damn near everything these days. Even my Universal TV remote. It's possible you pontificated precisely the words you submitted using some other technology, but Occam's razor points to you using a text editor instead...

Think of that fine fellow who was paid to implement the text editor in my universal remote. Why, after completing that task I bet the engineer went on to create other software, and got paid for making it too. One might even ague that monetising the scarcity of ones work might actually be a more valid business strategy than trying to profit from the artificial scarcity of software copies.

Indeed, without patents or copyright to create the artificial scarcity, the demand for new entertainment and better products would still exist. There would also still be people willing to work in order to meet the demand for a fair price... Thus, not only is "moving into services" a wise move for information marketeers, it's the only logical move since bits are in near infinite supply, and copies are inversely worth next to nothing.

Congratulations, you passed the first lesson of Economics! Now, if only we could get the RIAA and MPAA and BSA to do the same.

Re:An interesting commentary

[Anonymous Coward]

Just like Red Hat, VMware wants to make sure (at least) its hypervisor is fully supported by OpenStack so it can engage in talks with customers.

It might also want to move the project towards something that will better integrated with its own virtualization management tools.

Now for being "good stewards", I would not hold my breath.

Re:An interesting commentary

[Anonymous Coward]

If VMware can keep up innovation

It really hasn't, and now - unlike the arena of office suites - open source and other alternatives have come up hand over fist.

While VMware isn't in any real danger of going anywhere, they haven't had an MS Office style corner on the market since the 90s, nor are they ever likely to hold that sort of position ever again.

Re:An interesting commentary

[vlm]

Selling a proprietary virtualization empire is, in the long run, about as likely to succeed as writing a text editor

You mean like Microsoft Office, which at its core has a glorified text editor that is one of the cores of Microsoft's profitability?

Thanks that was Exactly my point. The market has space for one big text editor and maybe a couple "one man shop" specialized editors. And thats about it. Sucks to be word perfect or bank street writer or whatever. Not so bad if you're the one guy who won, but what are the odds.

On the other hand there are legions of contractor / consultant / educator / admin / author / designer types doing the "let me help you with MS Word", or pretty much any word processor, really, as long as you're willing to pay for the "help"

Re:An interesting commentary

[LordLimecat]

Until a couple of years ago, they were the only ones with anything comparable to vMotion. They are among just one or two others these days with Storage motion (moving where the physical data for a VM is stored, while that VM is on). They are (AFAIK) currently the only ones with anything like DRS (dynamically migrating VMs between hosts based on load), and certainly the only ones with anything like DPM (dynamically consolidating VMs based on load to save power-- during non peak, migrate everything to one server). From what Ive seen and heard, their networking (dvSwitch and normal vSwitches) are rock solid and best-in-class-- even Hyper-V and Xen guys tend to agree on this.

You can criticize lots of things about VMWare / ESXi, but they are able to get away with their ridiculous vRAM entitlements for a reason-- because they really are at the top of the heap for 99% of features. Its why when you look at competitor marketing, they always focus on value, not features-- because in most cases they are striving for parity, not dominance.

Re:An interesting commentary

[LordLimecat]

Not so bad if you're the one guy who won, but what are the odds.

I mean, when you're VMWare and youve been at the top of the virtualization pack for over a decade now, Id be feeling pretty confident.

Deceptive wording

[Eravnrekaree]

Since when should VMware be associated with anything involving the word "open"? Its a closed source, proprietary product that takes away users freedom to be able to modify and understand what it is doing on their system. At least as a part of all of their licences, give users the access to source code under a non-redistribute licence.

Virtualbox is a much better choice, which I recommend supporting financially. You can trust developers more that is upfront and honest with its code, rather than hides it as if it has something to hide.

Re:Deceptive wording

[Anonymous Coward]

Are we talking about ORACLE's Virtualbox here?

Re:Deceptive wording

[unixisc]

Ok, then support KVM/QEMU?

Re:Deceptive wording

[Anonymous Coward]

You mean Virtualbox owned by Larry Ellison? I agree. Support Virtualbox because Ellison has never locked anybody in or abused his companies Market position.

Re:Deceptive wording

[Anonymous Coward]

QEMU (and AQEMU, a related, but separately developed GUI) are well worth supporting. Good quality products.

Re:Deceptive wording

[QuantumRiff]

I pay for a small VMware cluster. On the download page, I can actually download all of the open source components used. Not just the client tools, but things like their data deduplication for their integrated backup. VMWare is actually a decent proponent of opens source, because its lack of licensing issues is what is making people go so crazy over virtualization. (spinning up 20 vm's to support a new project takes a few min with linux, and days with purchasing for windows ;))

Of course, they are like IBM, in that the best tool to work with one of theirs, is another of theirs.. (why, why oh why do they not have a decent SNMP monitoring system? I don't want a seperate monitoring tool for my vmware servers as I do for everything else)

Re:Deceptive wording

[aztracker1]

You realize that VirtualBox is now controlled by Oracle, right? I'm sorry, but I'll take VMWare as a company over anything Oracle touches/infects.

Re:Deceptive wording

[arkane1234]

http://www.ohloh.net/p/virtualbox
Virtualbox OSE, the open soruce edition with GPL licensing.

Re:Deceptive wording

[Anonymous Coward]

VirtualBox is unstable last I heard, at least on Linux. And everyone recommends KVM there.

text editor != text processor

[Anonymous Coward]

text editor != text processor

Text editor: barebones text without formatting

Keep your friends close...

[Anonymous Coward]

Keep your friends close and your enemies closer.

Enough competing hypervisors already!!

[Anonymous Coward]

As a systems person, one of the hardest things right now is keeping up with who we think is going to win the hypervisor wars. Any sort of consolidation in my mind is a good thing. Right now, there's VM software available from:
- VMWare
- Citrix (XenServer)
- Oracle (VirtualBox)
- Red Hat (RHV)
- Microsoft (Hyper-V)
- Open source (Xen, LVM, etc.)

Plus, every server and network vendor (IBM, VCE, HP, etc.) is trying their hardest to push boxes built around one of these software stacks _plus_ a proprietary network virtualization layer. Now, a hypervisor is a hypervisor, but keeping up with several vendors is tough.

Each one of these has a completely different ecosystem, tool set and support model orbiting around it, and it makes it very difficult to standardize on anything. I know each have their good points and are valid choices given the environments they go into, but having all this competition is crazy if you're just looking to figure out who's going to be around in the next few years. Having software vendors bundling products like this with others, plus their own VDI stacks, adds a third layer of confusion.

Re:Enough competing hypervisors already!!

[Alex Belits]

What SHOULD happen, is all hypervisors defeated by OS-implemented host compartmentalization (LXC and similar).

Virtualization is great for development, testing and running craptastic operating systems that you really, really hate to run but can't get rid of (what means Windows for sane people).

Re:Enough competing hypervisors already!!

[QuantumRiff]

One of the things openstack has been doing, is to abstract away the hypervisor. I believe you can manage VM's whether they are running KVM, ZEN, Amazon EC2, and vmware. Which means you can drop one, and move to the other, and still keep all your same tools to manage the machines, storage, etc, and just change or deploy new hosts. (which is big, but not as big as also re-writing all your tools for provisioning and such)

Re:Enough competing hypervisors already!!

[arkane1234]

I do not think virtualization means what you think it means.

Virtualization = FULL emulation (e.g. VMWare, VirtualBox, XEN, Parallels)
Compartmentalization = compartmentalization of memory and permissions. (e.g. OpenVZ, *BSD jails, User-mode Linux, Kernel-mode Linux, etc)

I figured I'd let you know that since you seem to think that hypervisors should be defeated by compartmentalization...

Re:Enough competing hypervisors already!!

[Alex Belits]

I do not think virtualization means what you think it means.

I can assure you, your thoughts have nothing to do with reality.

Virtualization = FULL emulation (e.g. VMWare, VirtualBox, XEN, Parallels)

And this technology has no place in OS design. It's a completely spurious level of indirection used as a crutch for broken operating systems. Last time it was legitimate, it was used on IBM mainframes to run nested instances of operating systems, because consistent kernel/userspace interface was not invented yet.

Basically, virtualization system's developer says: "I don't care what is at the level below hypervisor -- hardware must SERVE ME! I don't care what is at the level above hypervisor -- OS must SERVE ME! I don't care if your OS has millions times better scheduler, virtual memory, filesystems, anything -- I will put my ham-fisted interface into every path between hardware and software, so you all can just as well write it as if you are retarded monkeys -- BECAUSE I AM THE PERPETUAL BOTTLENECK, AND YOU CAN DO NOTHING, NOTHING ABOUT IT!!! BWAHAHAHAHAHAHAHAHA!!!"

Naturally, OS that is already designed worse than any hypervisor imaginable, does not suffers from this at all.

Re:Enough competing hypervisors already!!

[codepunk]

If you are doing it right it does not matter what you pick, at the end of the day these are nothing more than environments in which to run virtualized instances. If you have a good deployment model then the environment in which it runs means little.

I know at least in my shop we could potentially switch from one to the other nearly overnight.

Re:Enough competing hypervisors already!!

[Anonymous Coward]

Virtualization is NOT emulation. Virtualization and emulation are different things. Think of virtualization as a layer of abstraction to the hardware you have so you can actually have multiple different types of devices such as network adapters from different manufacturers but they all appear as one standard model to an OS so you don't have to worry about drivers. On the other hand, think of emulation as a simulation of hardware you may not have or is not possible to have.

You virtualize x86 on top of x86 since abstract one hardware type on same/similar hardware, but you have to emulate mk86 (or SPARC, VAX, Alpha) if you run it on top of x86 since the two are incompatible so you simulate an mk86 environment.

OS compartmentalization doesn't help running different OSs on the same hardware and I don't know of any implementations that allow for live migration between physical hosts. Compartmentalization solves the problems in a less complete manner and will not seriously compete with virtualization for solving real-world problems.

Re:Enough competing hypervisors already!!

[Alex Belits]

Virtualization is NOT emulation.

Hardware emulation. Hardware emulates different hardware (in this case, set of processors with peripherals). OS design started from this, then the idea of virtual memory that uses model radically different from the hardware memory model, killed it. Similarly, time-sharing turned into scheduling based on state of processes and i/o operations (modern virtualization uses "cooperative" scheduling where OS still has something to say, however it's still clumsy), and storage partitioning was replaced with permissions model in filesystems.

If someone has any illusions about the time I am talking about, let me spell it out: IT WAS LATE SIXTIES. Modern virtualization is a regression toward pre-minicomputers state of computer technology. And no, it's not any better this time around, it has exactly the same benefits and exactly the same flaws as then.

Re:Enough competing hypervisors already!!

[Alex Belits]

OS compartmentalization doesn't help running different OSs on the same hardware

The only reason for running multiple OSes on the same hardware is some software being sabotaged to not run on OS where everything else runs. There is no benefit in encouraging this kind of behavior.

and I don't know of any implementations that allow for live migration between physical hosts

It's such a rarely needed capability, no one bothered implementing it since Condor (that exists, but no one uses it). Any infrastructure for state persistency on the application level serves the same putpose better and provide other benefits unachievable by a simple checkpointing/transfer. However if it was necessary, it would be implemented with or without compartmentalization (it is not always necessary to combine hosts, when the goal is migrating applications).

Getting on the inside

[Anonymous Coward]

To quote from Yes Minister "We 'had' to break the whole thing [the EEC] up, so we had to get inside. We tried to break it up from the outside, but that wouldn't work. Now that we're inside we can make a complete pig's breakfast of the whole thing: set the Germans against the French, the French against the Italians, the Italians against the Dutch. The Foreign Office is terribly pleased; it's just like old times."

VMWare are already there, you just didn't notice

[Anonymous Coward]

Check out Cloud Foundry (http://cloudfoundry.com). A VMWare creation which is completely open, and non-proprietary, and is open sourced on github.

VMWare also recently bought SpringSource, the biggest Open Source company in the Java landscape. So yeah, VMWare are actually all for the open source movement.

Won't matter

[Anonymous Coward]

Won't really matter until OpenStack is no longer an obfuscated and poorly documented mess, such that it's essentially vaporware. Even attempts to make it usable (StackOps) are dismal.

Why is there so much hype around OpenStack compared to other, more viable, 'cloud-based' solutions?

Re:VMWare are already there, you just didn't notic

[Anonymous Coward]

"But it's not open enough!"

Certainly, anyone with at least one brain cell would know a company should gift its competitors all of the hard work it put into R&D over years and billions in future revenue just for someone on Slashdot that probably has never looked at the source code for anything on purpose is happy.

How many actual VMware customers care about seeing its source code for some sort of legitimate reason and not just out of interest? I'm willing to bet 0.