VMware Getting into PaaS with SpringSource Acquisition

[Reuven Cohen]

Hot on the heals of SpringSource's recent acquisition of Hyperic, VMware today announced their intention to acquire SpringSource. At first glance this move may seem puzzling, why would VMware want to buy an open source enterprise application development platform? Could it be for Hyperic, an open source IT management platform? I doubt it. I'd say it's all about planning for the future, a future where the OS no longer matters, a future where all applications are built, deployed and consumed via the Internet. Yes folks, I'm talking about Platform as a Service.

According to the post by VMware CTO Steve Herrod, he states that since it's founding 11 years ago, VMware has focused on simplifying IT. More to the point saying "VMware has traditionally treated the applications and operating systems running within our virtual machines (VMs) as black boxes with relatively little knowledge about what they were doing."

Moreover I too believe that the operating system seems to get in the way more then it helps. Add in overly complex hyper-visors and you've got several layers too many of abstraction when we all know the real work gets done in the application layer. Everything else just subtracts from the end goal -- Building and deploying scalable applications which at the end of the day is the only reason to have any sort of IT infrastructure anyway.

VMware even has a nice picture to illustrate their the new PaaS initiative:

The announcement goes on to outline "common goals for developers to easily build their applications and move from coding to production execution as seamlessly as possible… regardless of whether they will be deployed to a small internal datacenter for limited use or to a completely external cloud provider for much larger scale audiences (and the hopes of achieving Facebook application stardom!). This end state has a lot in common with what is today referred to as “platform as a service” (abbreviated PaaS). Salesforce.com’s Force.com and Google’s AppEngine are two of the best known examples of PaaS today."

I believe that this is a very smart move for VMware. I find it even more interesting because both Hyperic and SpringSource are open source plaforms. Does this mean that VMware is about to become an open source company, probably not. My read is #1 Hyperic is about to get shut down. It's an unneeded asset for VMware, and #2, SpringSource becomes a focus point in VMware's cloud strategy, a strategy that sees itself becoming the key point of interchange when deploying to the cloud, be it an infrastructure focused offering or platform offering. VMware wants to be in the middle and now they will be. (Cloud Interoperability is now more important then ever)

A few weeks ago Tom Lounibos, CEO of Soasta summed up the opportunity when asked "What's the future for Cloud "IaaS" vendors?"...he replied..."becoming "PaaS" vendors". So true

--
Reuven
CCIF Instigator

[Wil Sinclair]

It's a particularly interesting bit of news for our company, since Zend does essentially the same thing SpringSource does- only in PHP. In fact, it's part of my job to build the compelling platform for PaaS using PHP.

I agree with your thought about the OS's role going forward, Ruv. They seem to be going the way of hardware as another piece of infrastructure that needs to be abstracted out in the cloud. A sandboxed PaaS won't be the solution for everyone, but it will be a good option for the vast majority of enterprises. In addition to providing a more homogeneous environment for sysadmins working in the cloud, it can reduce educational and integration costs for development organizations by providing a common set of API's and technologies to code to.

It will definitely be interesting to see where SpringSource goes under the wing of VMWare.

,Wil

On Monday, August 10, 2009, at 02:09PM, "Reuven Cohen" <r...@enomaly.com> wrote:
>

Hot on the heals of SpringSource's recent acquisition of Hyperic, VMware today announced their intention to acquire SpringSource. At first glance this move may seem puzzling, why would VMware want to buy an open source enterprise application development platform? Could it be for Hyperic, an open source IT management platform? I doubt it. I'd say it's all about planning for the future, a future where the OS no longer matters, a future where all applications are built, deployed and consumed via the Internet. Yes folks, I'm talking about Platform as a Service.

According to the post by VMware CTO Steve Herrod, he states that since it's founding 11 years ago, VMware has focused on simplifying IT. More to the point saying "VMware has traditionally treated the applications and operating systems running within our virtual machines (VMs) as black boxes with relatively little knowledge about what they were doing."

Moreover I too believe that the operating system seems to get in the way more then it helps. Add in overly complex hyper-visors and you've got several layers too many of abstraction when we all know the real work gets done in the application layer. Everything else just subtracts from the end goal -- Building and deploying scalable applications which at the end of the day is the only reason to have any sort of IT infrastructure anyway.

VMware even has a nice picture to illustrate their the new PaaS initiative:

The announcement goes on to outline "common goals for developers to easily build their applications and move from coding to production execution as seamlessly as possible? regardless of whether they will be deployed to a small internal datacenter for limited use or to a completely external cloud provider for much larger scale audiences (and the hopes of achieving Facebook application stardom!). This end state has a lot in common with what is today referred to as ?platform as a service? (abbreviated PaaS). Salesforce.com?s Force.com and Google?s AppEngine are two of the best known examples of PaaS today."

[Geir Magnusson Jr.]

I think this is simply incredible. I've known Rod since before Spring-
the-project really existed, and watched him and his team grow the
company over the years.

Well done!

geir

[Rao Dronamraju]

I disagree.

 

VMware has flourished in a “niche” technology area and also because of the stupidity of the big players not to get into to the software virtualization markets themselves for a long time.

 

VMware is basically running out of “niche” markets. So this is a Hail Mary strategy to get into the traditional markets, considering SpringSource is not going to cost much, what the heck!, just buy something cheap and try it, if you loose you are not going to loose much anyway.

 

They want to get into the PaaS markets?...They will have to compete with Microsoft, IBM, Oracle/Sun and HP just to name a few. You might say they competed with the big wigs in the virtualization markets. Not at all. The big wigs for what ever reasons never got into the software virtualization markets until MS got into it but they left the field open for VMware for a long time and VMWare did an incredibly great job in making use of it.

 

But it is different now. The big wigs are pretty well entrenched in the PaaS technologies. Now the strategic scenario is exactly opposite. VMware is enterring other’s truf with not much strenghts in the PaaS technologies, products and services. What are VMware’s strenghts in PaaS?...all their employees have used Eclipse extensively for developing their software?:-)...does it have any kind of expertise base that is needed for .NET or J2EE type of object technology let alone the future Cloud/PaaS Objectware?..and beyond Objectware, PaaS Patternware!...Even its parent company EMC is woefully inadeqaute in this area, although I do agree that it is not their core competency.

 

I think VMware’s strategy would be much better if they stick to Management of Cloud. They now manage the hypervisor ecosystem pretty well. So if you CONTROL and MANAGE both the hypervisor eco and everything that runs on top also, you may be able to CONTROL and MANAGE the CLOUD. But here too they have a lot of competition. IBM’ Tivoli, HP’s Insight and OpenView, MS with its System Center can get down and manage the hypervisor along with managing the stack above, not to leave out other players like BMC etc.

 

So it appears to me VMware has run out of “free lunch” period and has enterted the era if real competition, although they still maintain considerable lead over MS and Citrix in their core competency.

 

Their acquisition of SpingSource is a weak strategic play. There are much better companies to acquire considering you have EMC as your backer.

 

All in all from a strategy point of view, I am not impressed.

 

But again, who knows, we may have to wait and see.

 

 

 

---

[Reuven Cohen]

The folks at the burton group had an interesting take on the acquisition in a blog post this afternoon at
 http://apsblog.burtongroup.com/2009/08/vmware-fuels-springsources-stratopheric-journey.html

I particularly liked this statement:

Attenuating the feedback loop between VMWare's vSphere and SpringSource's application servers will enable finer grained application footprints and timely elastic scaling. Imagine the ability to dynamically provision a specialized stack (i.e. container, framework, application meta-data and code) to handle a discrete unit of work (e.g. web request, transaction, resource query), intelligently pool the stack, and tear it down when not needed. SpringSource's investment in OSGI, inversion of control, lightweight containers, and application management interfaces establishes a necessary foundation to re-think and shrink run/manage away from monolithic systems (e.g. hardware virtual machines, application servers, and applications) and towards discrete business services.

r/c

[Rao Dronamraju]

I still disagree with burton group completely. I think a much better attenuation/utilization of feedback loop from VMWare's vSphere is in the autonomic computing domain not just application elasticity domain. Although application elasticity is a subset of larger autonomic capabilities of a cloud. So I still feel they are going in the wrong direction. I would much rather prefer the traditional management plane rather than application management plane. There is a natural technological alignment between what an OS does and traditional data center management software does and in the cloud context there is lot of scope for enormous intelligence incorporation to make a cloud 7X24X365 lightsout autonomic and beyond. So where I differ fundamentally is, I prefer infrastructure play rather than an application play of vSphere’s capabilities.

[Rao Dronamraju]

You know how much they are buying for?…whopping $420 million!…Good/Great for SpingSource!!!.

http://www.reuters.com/article/rbssTechMediaTelecomNews/idUSN1047900320090811

 

---

[Diego Parrilla Santamaría]

I agree that the OS has become (or will soon become) the big bottleneck. Morover, if you are familiar with java you know that most of the time the OS looks irrelevant (but it's not...). But being Java a very mature technology, all the efforts to make Java to look like an OS have failed. It seems that SpringSource found the way to succeed in this task. If they have something that can bridge the gap between the hypervisors and the OS, probably they found the way to succeed building a PaaS platform with Java.

SpringSource presence in the industry is massive with the Spring framework. It's opensource, but it has been driven by SpringSource -and that's one of the reason for its success-. If VMware has purchased SpringSource for the framework, they are simply stupid. If they have bought them to build the VMware OpenSource PaaS, then it's a smart move if they succeed -some big players have failed before-.

Another reason to believe that Enterprise Opensource is the right way.

Congrats Rod!

Diego Parrilla Santamaría
Business Development Manager & Product Technology Strategist at Abiquo.
+34 620 57 81 46
mailto:dpar...@abiquo.com
skype:diegoparrilla
www.abiquo.com

[Sam Johnston]

The OS is so yesterday's news... the cloud is the OS now. The sooner
we shed "instance" based scaling for "fabric" based scaling the better
- life is much easier over here.

This is a great move for VMware and something for the other
infrastructure providers to pay close attention to.

Sam on iPhone

On Tuesday, August 11, 2009, Diego Parrilla Santamaría

<diego.parril...@gmail.com> wrote:
> I agree that the OS has become (or will soon become) the big bottleneck. Morover, if you are familiar with java you know that most of the time the OS looks irrelevant (but it's not...). But being Java a very mature technology, all the efforts to make Java to look like an OS have failed. It seems that SpringSource found the way to succeed in this task. If they have something that can bridge the gap between the hypervisors and the OS, probably they found the way to succeed building a PaaS platform with Java.
>
> SpringSource presence in the industry is massive with the Spring framework. It's opensource, but it has been driven by SpringSource -and that's one of the reason for its success-. If VMware has purchased SpringSource for the framework, they are simply stupid. If they have bought them to build the VMware OpenSource PaaS, then it's a smart move if they succeed -some big players have failed before-.
>
> Another reason to believe that Enterprise Opensource is the right way.

> Congrats Rod!Diego Parrilla Santamaría

[Diego Parrilla Santamaría]

Maybe VMware and SpringSource want to develop something similar to:

https://www.projectcaroline.net

One of that amazing projects that Sun starts but never ends...

Diego Parrilla Santamaría
Business Development Manager & Product Technology Strategist at Abiquo.
+34 620 57 81 46
mailto:dpar...@abiquo.com
skype:diegoparrilla
www.abiquo.com

[Derik]

The problem with the PaaS feeding frenzy is that now we have many PaaS
to deal with. PaaS brings a "cloud" programming model with a
programming language. It appears that the programming models and
programming languages will all have some vendor specific proprietary
things. At the same time, we now have a plethora of models and
languages all going to moving the cloud "portability" scale into a
cloud vendor lock in. Once an application is built on a PaaS, moving
it could be a big thing. Thus, we definitely need some architecture to
ensure that separation of the PaaS, call it a PAL or PaaS Abstraction
Layer (hmm, another word just got created) hapens. Then such things
like business logic then remains portable. After all, all this stuff
is ultimately for the business logic. Interestingly, in the beginnng
we had COBOL today we have one of each (all the P langs, J langs, C
langs and many langs).

Anyhow, the reality in enterprise computing is that most IT budgets
are spent on maintaining applications versus building. We do need to
figure out how all those applications being maintained will move into
the cloud (even if that were humanly possible). After all, all the old
stufff can not just go away. Thus, the big challenge is to address the
question of what applications are cloudable for a desired ROI.

Finally, it seems that most IT education systems teach people to write
code. Then they go into the world and spend their time reading code
(and maintaining). Strange how the world turns.

Finally, finally there are lots of data out there for which these very
very old applications are needed to read such data. Those will all
have to be looked at for "clouding". Obviously, if such data was
"clouded" then things will be great. The problem is that such very
very old applications (and data) were developed by very very old IT
people.
So, we talk about security of the cloud. Obviously, if the data was
secure and unreadable then secureness of the data does not really
matter because the data needs its application to become information.
In the beginning there was indeed command lines (CLI) and COBOL. It
seems, we can't just leave such things behind. All the same, I like
Python and maybe we should just decide that we will all speak Python
in the cloud. Obviously, the zend people may revolt and speak PHP.
Then, the whole cloud becomes somewhat multi-lingual. I would say we
do have an issue on languages to resolve (maybe, some standardization)
and many other issues to standardize. That then would make up an
Enterprise Architecture technology model.

On Aug 10, 5:09 pm, Reuven Cohen <r...@enomaly.com> wrote:
> Hot on the heals of SpringSource's recent acquisition of

> Hyperic<http://news.cnet.com/8301-13505_3-10232113-16.html>,

> VMware today announced their intention to acquire

> SpringSource<http://blogs.vmware.com/console/2009/08/vmware-acquires-springsource....>.

> At first glance this move may seem puzzling, why would VMware want to buy an
> open source enterprise application development platform? Could it be for
> Hyperic, an open source IT management platform? I doubt it. I'd say it's all
> about planning for the future, a future where the OS no longer matters, a
> future where all applications are built, deployed and consumed via the
> Internet. Yes folks, I'm talking about Platform as a Service.
>
> According to the post by

> VMware<http://blogs.vmware.com/console/2009/08/vmware-acquires-springsource....>CTO

> Steve Herrod, he states that since it's founding 11 years ago, VMware
> has focused on simplifying IT. More to the point saying "VMware has
> traditionally treated the applications and operating systems running within
> our virtual machines (VMs) as black boxes with relatively little knowledge
> about what they were doing."
>
> Moreover I too believe that the operating system seems to get in the way
> more then it helps. Add in overly complex hyper-visors and you've got
> several layers too many of abstraction when we all know the real work gets
> done in the application layer. Everything else just subtracts from the end
> goal -- Building and deploying scalable applications which at the end of the
> day is the only reason to have any sort of IT infrastructure anyway.
>
> VMware even has a nice picture to illustrate their the new PaaS initiative:
>

>  <http://blogs.vmware.com/.a/6a00d8341c328153ef0120a5374e19970c-pi>
>
> [image: Image014]<http://blogs.vmware.com/.a/6a00d8341c328153ef0120a4e05446970b-pi>

>
> The announcement goes on to outline "common goals for developers to easily
> build their applications and move from coding to production execution as
> seamlessly as possible… regardless of whether they will be deployed to a
> small internal datacenter for limited use or to a completely external cloud
> provider for much larger scale audiences (and the hopes of achieving
> Facebook application stardom!). This end state has a lot in common with what
> is today referred to as “platform as a service” (abbreviated PaaS).

> Salesforce.com’s Force.com <http://www.salesforce.com/platform/> and
> Google’s AppEngine <http://code.google.com/appengine/> are two of the best

> known examples of PaaS today."
> I believe that this is a very smart move for VMware. I find it even more
> interesting because both Hyperic and SpringSource are open source plaforms.
> Does this mean that VMware is about to become an open source company,
> probably not. My read is #1 Hyperic is about to get shut down. It's an
> unneeded asset for VMware, and #2, SpringSource becomes a focus point in
> VMware's cloud strategy, a strategy that sees itself becoming the key point
> of interchange when deploying to the cloud, be it an infrastructure focused
> offering or platform offering. VMware wants to be in the middle and now they
> will be. (Cloud Interoperability is now more important then ever)
>
> A few weeks ago Tom Lounibos, CEO of Soasta summed up the

> opportunity<http://twitter.com/Lounibos/status/2951685895>when asked

[Rao Dronamraju]

Here is an interesting thing to note about VMware’s strategic direction to enter PaaS markets.

 

Paul Maritz when he was at Microsoft was the head of Platform Strategy and Developer Group.

 

So you see, any one shoud have guessed easily the direction in which VMware would be taken by Paul.

 

He is basically taking VMware where he has come from….Platform & Developer = PaaS.

 

Interesting tid bit…

 

---

[Reuven Cohen]

Agreed, great insight.

R.

[thomask...@gmail.com]

The big question for PaaS vendors is: Is there an easy migration path
from the existing enterprise install base of .net/jee into a PaaS
environment? Force.com does not appear to provide an easy migration
path for the existing install base of .net/jee applications, instead
most of the apps on Force.com are new applications. Microsoft is
working on an answer for migrating .net apps with Azure. None of the
Java vendors appear to have come up with an easy migration process for
JEE apps, although perhaps VMware will.

Tom Plunkett

On Aug 10, 4:09 pm, Reuven Cohen <r...@enomaly.com> wrote:
> Hot on the heals of SpringSource's recent acquisition of

> Hyperic<http://news.cnet.com/8301-13505_3-10232113-16.html>,

> VMware today announced their intention to acquire

> SpringSource<http://blogs.vmware.com/console/2009/08/vmware-acquires-springsource....>.

> At first glance this move may seem puzzling, why would VMware want to buy an
> open source enterprise application development platform? Could it be for
> Hyperic, an open source IT management platform? I doubt it. I'd say it's all
> about planning for the future, a future where the OS no longer matters, a
> future where all applications are built, deployed and consumed via the
> Internet. Yes folks, I'm talking about Platform as a Service.
>
> According to the post by

> VMware<http://blogs.vmware.com/console/2009/08/vmware-acquires-springsource....>CTO

> Steve Herrod, he states that since it's founding 11 years ago, VMware
> has focused on simplifying IT. More to the point saying "VMware has
> traditionally treated the applications and operating systems running within
> our virtual machines (VMs) as black boxes with relatively little knowledge
> about what they were doing."
>
> Moreover I too believe that the operating system seems to get in the way
> more then it helps. Add in overly complex hyper-visors and you've got
> several layers too many of abstraction when we all know the real work gets
> done in the application layer. Everything else just subtracts from the end
> goal -- Building and deploying scalable applications which at the end of the
> day is the only reason to have any sort of IT infrastructure anyway.
>
> VMware even has a nice picture to illustrate their the new PaaS initiative:
>

>  <http://blogs.vmware.com/.a/6a00d8341c328153ef0120a5374e19970c-pi>
>
> [image: Image014]<http://blogs.vmware.com/.a/6a00d8341c328153ef0120a4e05446970b-pi>
>

> The announcement goes on to outline "common goals for developers to easily
> build their applications and move from coding to production execution as
> seamlessly as possible… regardless of whether they will be deployed to a
> small internal datacenter for limited use or to a completely external cloud
> provider for much larger scale audiences (and the hopes of achieving
> Facebook application stardom!). This end state has a lot in common with what
> is today referred to as “platform as a service” (abbreviated PaaS).

> Salesforce.com’s Force.com <http://www.salesforce.com/platform/> and

> Google’s AppEngine <http://code.google.com/appengine/> are two of the best

> known examples of PaaS today."
> I believe that this is a very smart move for VMware. I find it even more
> interesting because both Hyperic and SpringSource are open source plaforms.
> Does this mean that VMware is about to become an open source company,
> probably not. My read is #1 Hyperic is about to get shut down. It's an
> unneeded asset for VMware, and #2, SpringSource becomes a focus point in
> VMware's cloud strategy, a strategy that sees itself becoming the key point
> of interchange when deploying to the cloud, be it an infrastructure focused
> offering or platform offering. VMware wants to be in the middle and now they
> will be. (Cloud Interoperability is now more important then ever)
>
> A few weeks ago Tom Lounibos, CEO of Soasta summed up the

> opportunity<http://twitter.com/Lounibos/status/2951685895>when asked

[Wil Sinclair]

I would say this is up to the applications themselves at least as much as the PaaS vendors. That is, "Was the application written against a higher-level abstraction to begin with?" If an application was written using Zend Framework components (Notice any pattern in my examples yet? ;) ), then the individual PaaS vendors can provide their own implementations of those components to support application portability to their own platform for that PHP application. This will be hard to do- if not impossible- if you use lower-level abstractions such as direct SQL against specific drivers for one particular database flavor vs. an ORM system.

Fortunately for PaaS vendors, many frameworks provide adapter mechanisms to support multiple technologies, so that the PaaS vendors wouldn't have to re-implement entire components to support apps written on these frameworks. There are also few frameworks that have enough traction to really worry about.

This isn't the full or final answer. Ideally, we would have a "standard" (as an OSS guy, I prefer the phrase "de facto standard") platform for apps to run on. But, on the other hand, choice and a little competition are generally good things, too.

,Wil

[Chris Marino]

Been waiting a while to formulate my thoughts on this.....

My initial, intuitive, reaction was that it was probably a good strategic fit, with an incredibly (insanely??) high valuation.  Then I read several posts/articles that positioned it as a strategic coup for VMware, which to be honest, left me puzzled. In fact, it had me wondering if I really knew anything at all about cloud the computing players. As simple anecdotal evidence, how often was Spring ever mentioned on this forum (let along SpringSource)???

So, I was happy to see that I was not alone in my perspective on this.

Savio Rodrigues at InfoWorld spells out in great detail all the reasons for my confusion.

http://www.infoworld.com/d/open-source/vmware-plus-springsource-more-hype-substance-389

Great read.
CM

[Reuven Cohen]

As an entrepreneur I love the valuation. As an industry observer, I can't help but agree with you. In terms of a cloud application platform, I'd say the work that Alfresco is doing is probably closer to the vision that VMware has outlined.

http://www.alfresco.com/products/platform/features/

r/c

[Tom Deckers]

I believe the reason why Spring hasn't been mentioned on this forum is because Spring in itself isn't cloud centric.  However, if now it's picked by VMWare to move upward in the stack towards PaaS, then it becomes relevant.  For that matter it's kinda coincidental... VMWare might have picked just any software framework.  Admittedly, Spring's one of the leading frameworks, which I guess makes it a bit less coincidental ;-)

So the real buzz in this announcement isn't so much focused on Spring, but on VMWare moving up the stack.

Regards.

Tom.

[Rao Dronamraju]

“VMWare moving up the stack”

 

Yes…it is moving to Vertical Middle Ware….in the infrastructure play it was SOME!ware, with Microsoftization of VMWare, it will end up Noware!.

 

 

---

From: cloud...@googlegroups.com [mailto:cloud...@googlegroups.com] On Behalf Of Tom Deckers
Sent: Wednesday, August 12, 2009 12:12 PM
To: cloud...@googlegroups.com
Subject: Re: VMware Getting into PaaS with SpringSource Acquisition

 

I believe the reason why Spring hasn't been mentioned on this forum is because Spring in itself isn't cloud centric.  However, if now it's picked by VMWare to move upward in the stack towards PaaS, then it becomes relevant.  For that matter it's kinda coincidental... VMWare might have picked just any software framework.  Admittedly, Spring's one of the leading frameworks, which I guess makes it a bit less coincidental ;-)

[Sam Johnston]

On Wed, Aug 12, 2009 at 7:00 PM, Chris Marino <christophe...@gmail.com> wrote:

Been waiting a while to formulate my thoughts on this.....

My initial, intuitive, reaction was that it was probably a good strategic fit, with an incredibly (insanely??) high valuation.  Then I read several posts/articles that positioned it as a strategic coup for VMware, which to be honest, left me puzzled. In fact, it had me wondering if I really knew anything at all about cloud the computing players. As simple anecdotal evidence, how often was Spring ever mentioned on this forum (let along SpringSource)???

Some important background that appears to have been missed altogether is that vCloud is apparently built on top of Spring so VMware have some amount of internal expertise already.

I'm not sure how that translates to spending $$lots to buy the farm though...

Sam

[Jang Younwook]

It may be a good decision, the PaaS service based on application framework like Spring can provide interoperability between legacy(?) spting application and new off-premised application.

Anyone who know whether Spring be remained open-source..?

2009/8/13 Sam Johnston <sa...@samj.net>

[Paulo Calcada]

This kind of discussion always ends on the same point - the vendor-lock-in issue. And in my perspective this is not a real point because, as it was told before, on a no-cloud environment (traditional datacenter's ecosystem: Server, OS, Development Platform) we had already this kind of problem. If a company decided to use the Microsoft's layer, it will definitely end by have problems moving to IBM or SUN, so, is the vendor-lock-in really a  Cloud Computing specific problem?

Paulo Calçada

2009/8/13 Jang Younwook <youn...@gmail.com>