Today, VMWare and Salesforce.com teamed up to announce VMforce an enterprise Java cloud.  In an effort to productize general-purpose cloud computing services VMware and Salesforces produce a joint product called VMforce.

The new solution and offering is a loud-based service called VMforce for running Java applications and is the combination of a server virtualization player in VMware (owned by EMC a huge storage company) and a customer relationship software company Salesforce.com.

What’s New Here?

This is a new approach to combining the benefits of cloud with real world applications in a completely “rent rather than buy” methodology.  It combines Software as as Service (SasS) with Infrastructure and Platform as a Service (IaaS and PaaS) to deliver a lower cost solution rather than owning all the individuals pieces outright.

What this really means is that companies are combining to vertically integrate the stack from the hardware all the way up to the application thus abstracting away all the hassles for customers.  At least that is the promise of this new world.

Up to now there has been a big movement toward public “open” clouds verses private “closed” clouds.  The distinction has been vendor lock in.  The results to date have been slow on the uptake by customers on public cloud for reasons of complexity and security.  While public clouds have been slow on the uptake, private clouds have taken off for reasons to do with customers favoring ease of solution deployment and better security.

Bottom line:  Customers favor ease of solution deployment and better security over lower cost.   It will be interesting to see what Citrix, Amazon, and Microsoft, and the open source world has to say on this.

A Public “Private” Cloud – VMware SalesForce Solution – VForce – Turn Key Cloud Solution?

Salesforce is leveraging the Java development framework which was acquired by VMware when the company acquired SpringSource for $420 million. According to SpringSource’s CTO the VMware deal is good for the market in that Springsource provides a development platform for engineers to build enterprise Java apps.

According to the VMware on their new VMforce:

“Companies are looking for solutions that deliver the benefits of cloud computing while leveraging existing resources, expertise and infrastructure,” said Paul Maritz, CEO of VMware. “By creating a dramatically simplified solution for modern application development, VMforce is a significant step forward in offering our customers a path that bridges existing internal investments with the resources and flexibility of the cloud.”

“In essence, we’re announcing VMforce, the world’s first trusted cloud for enterprise Java developers,” “This provides Java developers with a simple and trusted path to the cloud. And it’s a technology only Salesforce and VMware could deliver.”

According to the news from eWeek VMforce will include:

Spring Framework: VMforce will use the Spring Framework, the leading Java development framework that is backed by VMware’s SpringSource division. Spring makes it easy for developers to build powerful enterprise Java apps, increasing development productivity and runtime performance while improving test coverage and application quality. VMforce will also use the SpringSource Tool Suite, an integrated, tested and certified development environment offering the most complete set of Eclipse-based tools for creating Java apps.

SpringSource tc server:  VMforce applications will run on the tc Server runtime, the Enterprise version of Apache Tomcat.  tc Server is a wildly popular lightweight application server optimized for virtual and cloud environments.

Force.com Chatter Services: As the world migrates toward Cloud 2 social and mobile applications, developers will be able incorporate collaboration services from Chatter in their applications. These prebuilt services include profiles, status updates, groups, feeds, document sharing, the Chatter API and more.

Force.com Development Platform and Services: Since VMforce will run on the Force.com platform, developers have access to prebuilt business services that can be configured into their apps without requiring any custom coding. These services include search, identity and security, workflow, reporting and analytics, a robust Web services integration API, mobile deployment and more.

My Angle – Stacked Benchmarks

I was talking to some networking executives the other day and the conversation was about the notion of moving up the stack.  The market is all about moving up the stack toward the applications not down in the network.  The problem in the industry is that there are no authorative benchmarks around fully integrated stacks both at the hardware, network, and application levels that tie to business benefits.

What’s worse is that I am seeing company specific benchmarks where companies are trying to pimp themselves around their narrow products with benchmarks tested in silo’s not in real world use case scenarios.

Nice Stack! Moving up the Stack.

I was having a conversation just this week about how important it is for infrastructure providers to “move up the stack” that is deliver business benefits around the application levels.  Many innovative companies are adjusting their strategies to be more application specific or purpose built around application benefits.  The VMware Saleforce.com announcement today highlights this trend of intergrating the stacks.

Companies (customers) don’t want to buy raw infrastructures or platforms anymore unless their is a bundled application benefit.  This is why Infrastructure as a Service (IaaS) and Platform as a Service (PaaS) has been weak on the uptake.

Just today SiliconAngle contributor and industry analyst Dave Vellante highlighted a corner case with Emulex in that their benchmarks are too narrow and meaningless in that Emulex is throwing amazingly high numbers in some benchmark that has no application benefit.  With VMWare’s acquisition of Springsource and most recently acquiring Zimbra VMware is clearly putting a fully integrated stacked solution together for customers who want turn key business benefits not just “speeds and feeds” which require a ton of manual configuration – which means costs more money.

Here is that text and some of the comments back and forth from Dave’s post

But when a vendor starts touting something as absurd as 1M IOPs as a 2X competitive advantage it’s important to understand what is reality and what is hype. As we’ve previously published in our CNA Deep Dive performance in these markets is basically ‘table stakes’ with Emulex and QLogic about comparable. What’s of greater concern is the attention being paid to such meaningless figures as 1M IOPs at 0.5KB block sizes with unsustainable CPU utilization. There is no value in these metrics to IT practitioners; none.

What would be of value is to look at application performance. Specifically, as CNAs are used to virtualize IO, figures that give users confidence that virtualized applications will run at acceptable levels would go much further to advancing technology adoption than meaningless data.

Dave: I love the “Urinary Olympics” line – very good. The core issue today is the issue of moving up the stack. I was just talking about this with respect to Juniper and Arista Networks who have a new model of hardware that focuses on performance and value higher up in the stack. These kinds of benchmarks seem so 1999 to me just speeds and feeds. The core issue today is business benefits. Can you elaborate on the business benefits of 1 million IOPs. Rigging the benchmark with unrealistic block sizes seems like a “head fake”.

Yes John…you are right. The key issue in performance is one of balance– throughout the stack. It’s like a football team. Every component has a job to do and if you ask one player to do too much it becomes a problematic bottleneck. There’s nothing wrong with offloading to the driver, for example, if in fact that’s what Emulex is doing, however the question becomes how does that impact the other parts of the ‘value chain’ – i.e. can the processor handle the load, can the infrastructure support a virtualized environment. I think because new processors are so powerful, this is not going to be an issue but to your point – metrics like 1M IOPS don’t do anyone any good because at that hypothetical level the balance of the system disintegrates.   It’s all about what the application sees– that’s really the only thing that matters to a CIO.

At the end of the day we are moving fast to world where vertically dominated solutions by one vendor is over. No more will companies like Cisco dominate this new order or companies that only view the market from a single product perspective.

Companies like Juniper, Arista Networks, HP, and many others including new startups are betting their future of an integrated world where all things connected are applications specific in their benefits.

Companies who are are either betting with old and/or standalone products or end to end architecture (like Cisco) will continue to lose share.