VMware a Dead Duck

[esiebert7625]

My last post on this was deleted so I'll try again.

Interesting opinion by this guy, I think he is over-exaggerating the effects of competition on VMware. Sure their market sure will suffer, but they have been doing virtualization for a long time and are way ahead of their competitors. Competition is good for the customers, it keeps prices down and ensures that vendors constantly update their products to stay competitive. Just look at the effect Firefox has on Internet Explorer.

http://tinyurl.com/2455ho

This comment sums it up well also...

http://tinyurl.com/3y3czk

-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=-

Thanks, Eric

Visit my website:

-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=-

Message was edited by: RDellimmagine to replace links with tinyurls.

[continuum]

Eric - I am surely not qualified to comment on what happens in the enterprise market.

But if I look at what VMware does in the entry-level the comment "VMware a dead duck" is not so far fetched.

What do you answer when someone asks "can you recommend a free product - I am new to virtualisation and just want to test if it makes sense for me ?"

If VMserver 2 release is not very different from the first beta Virtual Box is the best entry to Virtualisation.

IMHO the long-term winner in the virtualisation market will be that company who is able to set a widely accepted standard.

VMware Inc. doesn't seem to be really interested in pushing their standard.

It started promising with the free VMplayer and the Virtual Appliances marketplace.

But without quality control the VA market place will be dead soon.

VMplayer without the ability to create VMs is pretty useless.

VMserver 2 with its bloated webinterface is not attractive for new users.

Workstation is too expensive for new users - compared with the free competitors.

VMware could easily release a free product that blows away Virtual Box or Virtual PC and so make sure that VMware-style VMs can be used on most computers tomorrow.

If you look at new VMserver 2 it looks like they are not interested in this at all. VMserver 2 may bring a couple of future ESX customers but is surely ground-loss in the entry level.

I find this new VMserver beta somehow alarming. It is in no way an answer to the feature requests of users for VMserver 1 - it rather seems to be an attempt to have a free "ESX-feel-alike" that should push future ESX-sales.

Sure VMware Inc. may make the most money with ESX - but if they want to survive they also have to make sure their platform becomes the world-wide standard.

If VMware someday is a "dead duck" then very likely because they are just too ESX-centered.

Just my 2 cents

Ulli

[esiebert7625]

I think the virtualization market still has alot of evolving to do especially with all the new players on the block. It seems like VMware Server is evolving more into the low-end Enterprise market as seen by it's integration with VirtualCenter. There was also a VMworld presentation on Server where they described it as the launchpad to ESX Server. It's tough to compete with free products because businesses like VMware need to generate revenue to keep their company alive and growing. With Server geared more to Enterprises they will probably have to do something with Workstation eventually which is more for the consumer market. It's priced kind of high right now, hopefully it will be reduced in the future to be more in line with Fusion. Maybe they can also spin off a free lite version of Workstation to help compete with the other vendor's free offerings.

-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=-

Thanks, Eric

Visit my website:

-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=-

[continuum]

It's tough to compete with free products because businesses like VMware need to generate revenue to keep their company alive and growing.

A "long run oriented" company may need to go though some lean days if survival is the first priority.

Other companies just exploit a single idea.

I am sure that if VMware-VMs run on any computer out there - selling enough ESX to be able to spend some of that money in the entry segment should be possible.

Also I don't think that pushing the entry level really must be expensive.

Lets have a look at the products:

VMplayer - is expensive for VMware but not really useful enough for users to push their standard

VMserver - is expensive for VMware and not really convincing for companies to use VMware as their virtualisation platform as it is artifically limited

Worksation - here VMware displays their suprior knowhow and experience but still it is artificially limited

ACE - the full Workstation without artificial limitations

GSX - at the moment there is no product in this segment - Workstation technolgy plus network - same control-tool as ESX

ESX - this is the pay - cow

In my layman's view this is not wise: the free products are not really good enough to storm the world and set a standard.

The hosted platform products that currently exist are overprized or do not really supply the needs of the customers.

As the hosted platforms basically all use the same technology setting their feature-list is a matter of a decision.

IMHO they should offer one limited subset of Workstation technology - that is convincing enough to become the most used platform - for free as competitors are free too.

Paying users want the best Workstation - paying companies want the best Workstation running through the same tools like their ESX-servers.

Two of the VMware products don't really make sense to me : VMplayer and VMserver - developement and research in this two are wasted.

This energy would better go back to GSX - meaning Workstation managed in same environment as ESX.

I would simply use a Workstation-lite as the necessary free-product - without any extra research and developement - just Workstation without Teams, Policies, Snapshotmanager.

This doesn't waste any energy - removing some entries in the GUI-menu would be sufficient.

There was also a VMworld presentation on Server where they described it as the launchpad to ESX Server.

Yes - it looks like this. Why does VMware force a "all or nothing" decision ?

Companies - that can't change to ESX yet - maybe be forced to switch to another product. Some of them surely would use GSX happily as long as they have to use Windows/Linux and can't switch to ESX. VMserver forces a decision: ESX or something else - this looks so stupid to me. Sell those companies GSX instead.

[esiebert7625]

Well I pretty much agree with you. They probably would of been better off giving Workstation away for free and charging for Server instead. Still like you said giving a basic Workstation product away for free without the advanced features like snapshots make sense so they can better compete with all the other free offerings. I do think Player does have some value as it can be used in effectively in certain situations. But Server and Workstation are so similar they are better off with only one desktop product like you said. They should give away basic Workstation for free, charge a affordable price for an advanced version of Workstation, change Server into a more Enterprise product and add features to it and sell it is a low end ESX-like offering. I think making Server compatible with VC is definitely a step in that direction.

-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=-

Thanks, Eric

Visit my website:

-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=-

[Jae_Ellers]

This guy gets my goat. I responded here: http://blog.mr-vm.com/2007/11/system-virtualization-market.html

Bottom line: VMware has the line on making money with Enterprise Virtualizaton(tm). Xen et al are making progress in this space, but mostly penetrating the Linux space. It's hard to make money with Linux OSes. So I bet it's really hard to make money virtualizing them.

Please tell me which open source based system you will put your AD infrastructure on tomorrow?

-=-=-=-=-=-=-=-=-=-=-=-

Check my blog:

-=-=-=-=-=-=-=-=-=-=-=-

[TomHowarth]

This XEN hysteria, makes me giggle, it reminds me so much of the much vaunted Open Source assault on Microsoft, I remember late 1990's the OS people stating that we'd all be running Linux desktops and servers as it is better. it is now nearly a decade later and my main work tool is a MS based Laptop, the majority (no Vast majority) of the servers I've installed and systems I've build are MS based in fact on the whole the only 'NIX based environment I regularly utilise is the SC. I believe that this is the case for the vast majority of forum users.

XEN suffers from the same problem, with MS I have one OS Windows (different versions maybe) with ESX I have one OS. simple things are the same, (KISS Keep it simple stupid)

with XEN I have Citrix XEN, Sun XEN, Oracle XEN, OpenSource XEN etc, etc, etc, all different in subtle ways, an administrative and training nightmare, these are the hidden costs to a XEN deployment.

with ESX I have VirtualCenter (a windows based tool) VCB (a windows based tool), these reduce the training overhead as they follow a familiar paradigm for the majority of Windows Admins

XEN based admin is done at a Linux interface (could be wrong here but I don't remember seeing any Windows based admin tools) Another steep learning curve.

True, XEN offers some competition to VMware but I feel it is competition of the style of Linux v Windows. and not competition as in Netware v Windows.

The entry of the Redmond mothership in to the fray will be when the real competition hots up, and it will not be a technical war, but a Marketing/Spin war. this is a battlefield MS are very familar with, they don't loose many skirmishes here. VMware will either have to ramp up their regular forces for a head to head, (not advisable) or learn some tactics of a irregular army. they will need some very good marketeers, currently the best in this arena work at MS, VMware need them, Virtualisation is past the technical sale stage. and into enterprise and board room arena. different skills are needed here. it is less tech more spin.

Kind Regards

Tom,

[mreferre]

I have to agree with Tom here.

Xen is too much of a fragmented game at this point and it's not a good thing (for customers that seek <something> which is stable from a market/roadmap perspective). Also Xen suffer the kid-syndrome as many open source technologies: as soon as it's <done> the attention shifts onto the <next cool thing> (which is now called KVM) and all od a sudden the old thing gets forgottent (well I am oversimplyfying here but you get the point....).

MS might have a different approach even though they are a little bit lost in the wind as well in my opinion .... they have always said that virtualization should be a feature of the OS while now they are coming out with a standalone hypervisor as well etc etc etc ......

I think having a solid / consistent roadmap/strategy/vision is going to be important in this market and currently VMware seems to be king of the hill. But I agree with Tom ... Redmond is going to pose some challenges to Palo Alto ....

Massimo.

[admin]

Well, we don't have to look much further than Novell to see how things can go for a company. In the early 90's, Novell was a market leader in the server space. They were arrogant, unfriendly to their partners and customers, etc, etc...and this opened up an opportunity for Microsoft to capture the market. Where is Novell now? Trying to start over and recreate something out of Suse Linux. Microsoft has learned long ago that you don't have to be the best...just the most popular to win at this game. VMware has to capture the SMB market before the Microsofts and XEN flavors out there take it from them. Once they gain the SMB space, it's only a matter of time before they become the defacto standard...and then capture the enterprise space too.

VMware still has time to help themselves before the competition closes the gap. History reminds us of the Novells and Netscapes out there. Having said all of this, the leadership at VMware are certainly alot smarter than I am...and no doubt very keen to the market and their competitors. I am sure they have a plan of attack...so we all have no worries here right?

Chris

[AWo]

This XEN hysteria, makes me giggle, it reminds me so much of the much vaunted Open Source assault on Microsoft, I remember late 1990's the OS people stating that we'd all be running Linux desktops and servers as it is better. it is now nearly a decade later and my main work tool is a MS based Laptop, the majority (no Vast majority) of the servers I've installed and systems I've build are MS based in fact on the whole the only 'NIX based environment I regularly utilise is the SC. I believe that this is the case for the vast majority of forum users.

XEN suffers from the same problem, with MS I have one OS Windows (different versions maybe) with ESX I have one OS. simple things are the same, (KISS Keep it simple stupid)

with XEN I have Citrix XEN, Sun XEN, Oracle XEN, OpenSource XEN etc, etc, etc, all different in subtle ways, an administrative and training nightmare, these are the hidden costs to a XEN deployment.

with ESX I have VirtualCenter (a windows based tool) VCB (a windows based tool), these reduce the training overhead as they follow a familiar paradigm for the majority of Windows Admins

The whole discussion is like a deja-vu for me back in the good old times where Novell was the leading network company and Microsoft started with their directory service, already hunting market shares from Novell. I can remeber how we were laughing at functions that MS announced and that we already had since NetWare 3.11 or their directory implementation compared to eDir (it is not nearly as good as eDir today!). Everbody agreed on that Novell is the technology leader, at least 3-4 yours in advance before MS (later this shrinked to two years). And yes, they really had the lead and the technologie was far more better and adult than AD was.

Everybody knows how the picture looks today.

Virtualization is more and more becoming a commodity product, thereby leaving the technical sector, where the technical skilled people drive the decisions because they were the only one to know. The functions become comparable (even if their technical realisation and the quality of it may differ) and thereby the decisions are made on other criteria: costs, political, support (and MS has the power to steer at this point. Remeber the SAP support for VMware). Marketing and caring about key deciders will do the rest (and MS is very good here). A (financial) controller doesn't care about an easy life for administrators as long as the main purpose to virtualize servers, having high-availability and all the stuff is satisfied. And at least XEN will reach this level soon.

This is not an statement about my opinion regarding VMware or XEN. I just want to state, that technology excellence or a large installed basis doesn't do it. History has shown. And even if Linux doesn't wiped Microsoft servers away, it has a strong installed basis today (doing work on machines were otherwise a MS licensed OS would have run). This could be another entry for the not-so-Linux-bound XEN. It's really hard to get a server without Intel's or AMD's virtualization technology, today. Thereby even hardware is not a barrier anymore for people who want to virtualize MS servers with XEN.

By the way

XEN based admin is done at a Linux interface (could be wrong here but I don't remember seeing any Windows based admin tools) Another steep learning curve.

Yep, you're wrong, there is ONLY a Windows-based tool (Xencenter). The Linux community beatens XEN for publishing a Windows-only administration tool. They say: "Why should I have to install a Windows machine in a pure Linux environment to mange my (not-any-more-then) pure Linux environment." They are right here. So, XEN will release a OS independent Web-based tool.

AWo

[esiebert7625]

Novell, that brings back some memories, I was a Netware admin in a former life, back in the Netware 2.x/3.x days. They are a classic example of the rise and fall of a software company when they don't adapt to changing shifts in the industry. By the time they did something it was too late and their two main products Netware/Groupware went downhill until they almost had no market share.

-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=-

Thanks, Eric

Visit my website:

-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=-

[TomHowarth]

This was the arguement I was trying to make, Novell was taken down by a dominant player, MS were pervasive on the desktop and in the application space, and had got enterprise penetration via thier work on O/S2. but even then they had a hell of a marketing team, they knew that they would not win a pure technical argument, so they went for the board. VMware has to win the Board, they have the technical arguement. they aslo need to be more active in the SMB market space.

Kind Regards

Tom,

[esiebert7625]

Thats another oldie, I used OS/2 1.x - 3.0 Warp extensively back in my Domino admin days.

-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=-

Visit my website:

-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=-

[continuum]

IMHO VMware didn't react very well when the market demanded something for free.

They dropped one existing good tool and replaced it with two half-hearted free ones.

Don't you think a tool above WS that can be managed through the same tools like ESX would be a much better launchpad to ESX than a free - limited tool like VMserver ?

ESX always needs a decision to change - those that don't want to switch right now, should be able to grow into VI still using a professional hosted VMware app that listens to the same management-tools and runs the same VMs as ESX.

If an office has slowly grown into running their servers on top of maybe 2 or 3 hosted-platform "VMserver pro" - replacing the first "VMserver pro" with an ESX is no big jump anymore -it is the logical next step.

Well - folks you are all much more qualified to comment here than me - in my hobbyists common sense I just find it strange that a company invests time and research into artificially limited products they give away for free.

Isn't coming up with the best more convincing ?.

[dpomeroy]

My guess is that we will see a free version of ESX Server (with limitations) at some time, probably replacing VMware Server altogether. VMware has been expanding the amount of hardware "officially" supported by ESX and its no longer limited to just a handful of tier 1 vendors, and I think SMBs would rather have a free ESX version than a free VMware Server. If Microsoft (and everyone else) is going to give away a Hypervisor then VMware will have to do the same to be competitive.

Don Pomeroy

VMTN Communities User Moderator

[AWo]

As a consultant who can act and choose free what the market offers as solutions to customer problems/demands, I would lean back and wait. It is up to the players how they respond to requests and demands from the customers and what they offer at which price. Therefore some more competition is not the worst thing to get. At least it keeps them busy. And if one really goes down like Novell did, there'll be a reason why.

AWo

Message was edited by: AWo

[esiebert7625]

Another response to this article:

http://havemacwillblog.com/2007/11/30/vmware-is-absolutely-not-a-dead-duck/

-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=-

Visit my website:

-=-=-=-=-=-=-=-=-=-=-==-=-=-=-=-=-=-=-=-=-=-=-

[AWo]

If he's referring to Xen as the free software, he is not right, as the enterprise level product Xensource Enterprise is not free software. Only the kernel on which this product is based is free.

In case of ESX 2.5. VMware was close to commodity. We saw that the customers could handle this really easy on their own and they didn't ask for proof or concepts and help any more. With VI3 the customers started to request help again because it is more complex and there are more and new technical function/options available. Of course there are always cases and scenarios where in deep knowledge are needed, but the mass will be served by easy-to-handle products. Of course the latest virtualization techologie won't be commodity because it's new. But over time virtualization will become as normal as processor technologie or operating systems are today. Think about processors, this is high technolgy, complex and you need specialists to design and build one. But the users, just buy them and plug them onto the motherboard.

Regarding the technological advantage VMware has today, this mustn't stay so and this didn't help other former technological leaders in the past (as mentioned above). Famous examples are:

- Novell

- Netscape

- IBM Microchannel

- OS/2

I won't bet on anything in this discussion as this has benn proofed to be useless and as a consultant my job is not to pray for any vendor but to find the technical and commercial solution which helps to address customer needs. Therefore my some of my decision criteria are:

- standard technologie

- support

- proofen technologie

- costs (purchase and ownership)

- trusted vendor

And I can't see why others shouldn't receive this level. Sooner or later.

AWo

[danpalacios]

Three thoughts to throw in here:

1. The legal landscape has shifted slightly since the "glory" days of Netscape and Novell. Microsoft has split (sort of) the hypervisor from the OS - likely to avoid another huge anti-trust beating from the EU.

2. 3i will likely be very attractive to SMB (with low cost and ease of use while still providing an upgrade path). The cost of 3i + 4 std win licenses is equivalent to 1 ent win license.

3. ESX is a platform. is as much akin to blade servers and mainframes as to OSs. This makes our Novell, Netscape analogies a little less accurate. Maybe we should be looking at MSSQL vs Oracle for our comparison.