The Enterprise Server Impact Of Intel/AMD Settlement Remains To Be Seen

As an infrastructure and operations professional, it’s hard not to like the terms of the settlement between Intel and AMD but the implications for your buying strategy and roadmap aren’t that clear yet. The rules of engagement spelled out in the Intel 8K make it clear that server vendors are free to select the CPUs that best fit their design goals and to carry as wide a portfolio of AMD-based servers as they feel the market demands. Server vendors can also market AMD to their heart’s content without fear of retribution from Intel (if the accusations from the past are true — something Intel isn’t admitting through this settlement). It also means that the two companies can stop aiming so many legal weapons at each other and redirect those funds and management brain cells to accelerating innovation. The $1.25B influx to AMD will be a very welcome shot in the R&D arm, too.
That’s all goodness that may lead to greater adoption of AMD-based servers in the future by enterprises. It doesn’t mean we can expect equivalency in performance or purpose between them, though. All it does is spell out the rules of engagement for fair play.

It also doesn’t ensure full compatibility between AMD and Intel processors and this is where one question in particular lies unanswered.

According to the 8K, Section 2.3 on Technical impairments:
“Intel shall not include any Artificial Performance Impairment in any Intel product…mean[ing] an affirmative engineering or design action by Intel (but not a failure to act) that (i) degrades the performance or operation of a Specified AMD product, (ii) is not a consequence of an Intel Product Benefit and (iii) is made intentionally to degrade the performance or operation of a Specified AMD Product. For purposes of this Section 2.3, “Product Benefit” shall mean any benefit, advantage, or improvement in terms of performance, operation, price, cost, manufacturability, reliability, compatibility, or ability to operate or enhance the operation of another product.”

The key word here is “compatibility.” Case in point: virtualization live migration. Intel VT, the set of instruction sets in the Xeon CPU family that helps with the performance of virtual infrastructures and enables live migrations from server to server are not compatible with AMD-V technologies that do the same thing (roughly). What this means today is that with Intel’s implementation you can’t live migrate from an Intel-based server to an AMD-based server. AMD-V actually lets you migrate from AMD to Intel, but you can’t move back. As a result of this you are probably building out your server virtualization pool today purely with Intel-based servers.

So here’s the question: “Is the incompatibility of virtual infrastructure implementations an Artificial Performance Impairment or simply different approaches to delivering Product Benefit?”

The Intel 8K then states:
“In no circumstances shall this Section 2.3 impose or be construed to impose any obligation on
Intel to (i) take any act that would provide a Product Benefit to any AMD or other non-Intel product,
either when such AMD or non-Intel product is used alone or in combination with any other product, (ii)optimize any products for Specified AMD Products, or (iii) provide any technical information, documents, or know how to AMD.”

Hmm. This seems to state that Intel is under no obligation to address this particular example if the resolution provides a Product Benefit to AMD. And it further states in that same “(i)” that such Product Benefit includes cases of the products being used in combination – which describes the case of Intel and AMD servers in a shared virtual pool. It further states that Intel doesn’t have to provide any technical information, documents or know how to AMD through which they could take the initiative themselves to solve this problem.

If AMD can’t penetrate any enterprise’s virtual server pool because the feature of live migration doesn’t work across server families, that would seem to me to be a serious Artificial Market Impairment. But does that fit the terms of the settlement that state Artificial Performance Impairment must be proven? Hmm.

This is just one example of the types of questions and scenarios Intel and AMD’s lawyers and technical experts will have to apply to this agreement. And answers to these types of questions will be imperative for enterprises to understand as they can have significant impacts on buying behavior. 

Until these types of scenarios are played out the impact of this settlement on enterprise buyers remains unsettled.

By James Staten

Check out James' research

You can follow James on  here

Cloud Is Defined, Now Stop the Cloudwashing

This blog post is a response to an article by Alex Williams on ReadWriteWeb. Thanks for the shout out, Alex, and for bringing more attention to the contentious issue of cloud computing definitions. While Forrester research reports are created exclusively for our clients, our definition is freely available:

A standardized IT capability (services, software, or infrastructure) delivered via Internet technologies in a pay-per-use, self-service way.

We first published this definition back in March 2008 in the report, “Is Cloud Computing Ready for the Enterprise,” and have held to that published definition ever since (in fact it has been leveraged in multiple Forrester reports, speeches at industry events, news articles, blog posts and tweets since that original publication). Our definition was also used by NIST and several other Federal government agencies as a resource used to create their definition.

One of the key values Forrester Research provides to its clients is helping them navigate the technology trends and delineate what is a new type of technology and what is simply last year’s technology in new clothing (what I call “cloud-washing”). Thus it was imperative for us to publish our definition early and we have since been striving to provide clear taxonomy and categorization of cloud services, as shown in our latest Forrester Tech Radar.

In this Tech Radar we lauded NIST for their definition, and contrary to your statement, do not believe we need the “circus” of more or better definitions at this stage. Rather we believe we need broader recognition of what is and what isn’t cloud computing to get past the marketing hype and make it easier for customers to identify and then consume these valuable new service offerings. That’s why we’ve stuck with our definition since 2008 and are glad to see NIST sticking to theirs.

By James Staten

Check out James' research

You should follow James on  here

Cloud Computing Belongs On Your 3-Year Roadmap

Welcome to the fourth quarter of 2009; what we at Forrester call planning season for most IT departments. In a typical year, this is the time that infrastructure and operations professionals spend lots of cycles burning through what remains of the 2009 budget and building plans for investment in 2010 with the hope of gathering a bit more budget than last year. Of course this is no ordinary year. Economists and financial prognosticators, like our own Andrew Bartels are predicting a long recovery from the recession and further delays in IT spending. That means another year of your infrastructure getting older. There’s two ways of looking at this problem and thus your budget proposals for 2010:

• You can either predict when you will get the rights to refresh the systems and return to the infrastructure spending patterns of old, or

• You can realize that a long recovery means a new normal for IT spend is in order.

We think the latter is far more fruitful even if it isn’t entirely accurate. It’s both prudent and different because your go-forward infrastructure strategy likely isn’t the same anymore either. If you are like the respondents to our Q3 2009 Enterprise and SMB Hardware Survey, North America and Europe, you:

• Continue to consolidate infrastructure like a mad clear-cutting logging company.

• Are repackaging workloads into VMs faster than an Amazon.com shipping manager.

• Are standardizing like McDonalds – everything from hardware, golden master server images to change management processes as consistent and repeatable as possible.

• Are racing like Usain Bolt to shorten deployment times from weeks or months, to new world records.

• Are learning to love automation – if only in a small way.

In other words, you aren’t buying, running or managing infrastructure the way you used to. So why plan your budget the old way. And further, why budget for the new now, when you can budget for what’s next in your journey. As we noted in our report, Assessing Your Infrastructure Virtualization Maturity, there is a clear path of improvement for infrastructure architecture and that’s toward customer empowerment, deployment standardization, provisioning and capacity management automation and further consolidation – even across business units. Yes, I said it.

In other words you’re evolving into an Infrastructure as a Service cloud. But despite vendor claims for products such as the VMware vCloud, Eucalyptus and Elastra offerings, most enterprises can’t just drop in an IaaS platform and they have a cloud. You have to prepare your organization and your operations staff for this. But that doesn’t mean you can’t achieve this aim in a short time horizon or that you can’t embrace public or hosted cloud infrastructures in the near term.

Forrester feels that cloud computing is one of the Top 15 Technology Trends and that it warrants investment now so you can gain the experience necessary to take advantage of it in its many forms to transform your organization into a more efficient and responsive service provider to the business. Find small non-critical projects to start with so you can learn how best to apply these services to your business and combine this learning with the advice in our Tech Radar to help plot the timing for these investments. Also read through the many cloud computing case studies like those on USA.gov and CryoPort to see how others are doing this. By investigating these top 15 technology trends and shifting to a strategic rightsourcing approach to IT portfolio investment you can hollow out the MOOSE and become more strategic (and viewed more as a change agent) to the business.

By James Staten

Check out James' research

You should follow James on  here

Assessing The Maturity Of Cloud Computing Services

The number one challenge in cloud computing today is determining what it really is, what categories of services exist within the definition and business model and how ready these options are for enterprise consumption. Forrester defines cloud computing as a standardized IT capability (services, software, or infrastructure) delivered via Internet technologies in a pay-per-use, self-service way.

While definition is crucial to having a fruitful discussion of cloud, the proper taxonomy and maturity of these options is more important when planning your investment strategy. To this aim, Forrester has just published our latest Tech Radar that maps the existing cloud service categories (not the individual vendors within each category) along maturity and value impact lines to help you build your strategic roadmap.

The report identified 11 service categories that fall into three classes of cloud services – software you rent (Software-as-a-service, or SaaS), middleware services and platforms that help developers build cloud-based applications, and infrastructure services and platforms that are places to deploy cloud applications. Note that we did not restrict the second and third categories to Platforms-as-a-service (PaaS) and Infrastructure-as-a-service (IaaS) platforms – because there is value in discrete middleware and infrastructure services in the cloud just as there is in your data center. While Amazon Web Services is clearly an IaaS leader providing a robust set of compute, storage and middleware services, there are also discrete offerings, like Boomi’s cloud integration service that does no more than integration and does not need to be a full platform player to be valuable. In fact there can be significant advantage that comes from this type of focus on just one thing.

Oracle’s Larry Ellison continues to grab headlines for his assertion that the cloud isn’t anything new and to his credit he’s at least half right. Just because applications are delivered from a cloud infrastructure doesn’t mean any of the aspects of application design or components of a service-oriented architecture don’t apply – in fact they have just as much relevance as they did on-premise. But how these services are delivered is what is different. There is a clear difference between an on-premise, single instance deployment of TIBCO and the highly scalable, multi-tenant integration service from Boomi. Tenancy, shared economics, virtualized deployment, and cloud service-to-service integration are game changers for those who are using these services. Not only do they bring potential cost advantage to applications that might otherwise have been deployed on-premise but they create opportunities for new business applications that simply wouldn’t be feasible any other way.

That’s clearly the case for Cryoport, a maker of cryogenic containers for live medical tissue. Delivering these materials safely was very hands-on and expensive until cloud computing came along. Same with genome research; medical research universities and pharmaceutical companies had to make large investments in HPC labs to crunch the massive volumes of DNA data, until cloud computing economics came along.

But cloud computing isn’t the be-all, end-all that many portray and they can’t suit all uses today and may not in the future either. Thus it behooves you to build your roadmap using guideposts to the maturity and applicability of these emerging options. We hope this report helps you bound and plan your cloud investments and look forward to your feedback on how we can assist further.

By James Staten

Check out James' research

You should follow James on  here

When is a VMware cloud a vCloud? Let the API wars begin

Attention enterprises —Pop Quiz: If your favorite hosting provider launches a cloud service that supports VMware vSphere and is part of the VMware vCloud initative, are they providing you with the rich vCloud functionality VMware is touting at VMworld this week?

The answer is no, because vCloud hasn’t been delivered yet and probably won’t be until the first half of 2010. If you answered yes, don’t feel too bad because VMware, in its efforts to prepare the market for vCloud, has made the current state very confusing. First off, the company grandfathered all the members of its VMware Service Provider Partner (VSPP) program into being vCloud partners. This has led many a traditional hoster – who doesn’t even have an infrastructure as a service (IaaS) cloud platform to say they are now a cloud. Sorry, but if you can’t automatically provision VMs upon request from a self-service portal, don’t have pay per use billing and require a 12-month commitment, you’re not a cloud platform. Second, a hosting provider can support the vSphere VM format without letting you manage your VMs in a standard way (they don’t even have to give you access via vCenter). And third, if you want to integrate with their platform programmatically, their APIs are likely their own.

Why does this matter? You use VMware in-house, you just want a cloud that does too. The VM format supported isn’t really the issue (heck converting your VMs from VMDK to OVF, Xen, or Amazon AMI is fairly trivial. And tools like Citrix’ Project Kensho do this very effectively).  Management is where this really matters because the simplest way to consume IaaS cloud resources is when you can use your own tools for deployment, monitoring and reporting, administration, security, integration and life cycle management. And that simply isn’t the case today (thank you Amazon for letting enterprises finally manage cloud security with their own tools).

The same criticism can be levied at all Xen-based clouds today, too. Amazon, Rackspace, GoGrid and several other hosters have published the APIs to their cloud platforms but only Amazon’s, thus far, have a realistic chance of reaching de facto standard status (especially when Eucalyptus and the Sun cloud – if it ever sees the light of day under Oracle – use the same APIs).

Interoperability requires consistent APIs – a subset at least – among cloud platforms. And VMware took this step today. It’s a pre-emptive move and a smart one to release its vCloud APIs and hand them over to the DMTF to manage as public APIs, because it lays the groundwork for a promise of better interoperability and compatibility between IaaS clouds (and enterprise internal and hosted clouds) when vCloud finally arrives. To try to ensure that their APIs can deliver this value, VMware also announced its vCloud Express program that asks for a commitment from cloud hosters to support and expose the vCloud API within their API set ensuring not only a clear conveyance through marketing that this cloud is compatible but giving the hoster the ability to differentiate around this API by intermingling it with their own unique functions.  This is a good move by VMware because this guidepost will be key for enterprises that want a cloud provider that can most closely match their in-house deployment. It also gives ISVs some runway to build their solutions and have them ready by the launch of vCloud in 2010.

It’s also competitively important in light of the momentum Amazon Web Services has with its APIs and this week’s announcement by Xen.org to raise its level of cloud compatibility.

This announcement has also moved the hypervisor wars to a new battlefield – programmability. Because at the end of the day, it’s less about how good your virtualization solution is (there’s a clear bar of function, performance and maturity that must be crossed – and XenServer and VMware both are over that bar); it’s how big your market share and how robust your ISV and hoster ecosystems are that matter. Xen has clear hoster ecosystem market share leadership in IaaS clouds. VMware brings the muscle of its enterprise and ISV market share to this battle.

The war is really just a skirmish today as still just a fraction of enterprise buyers are leveraging IaaS cloud services. Forrester survey data shows that while less than 10% of enterprises are leveraging IaaS clouds today, more than half plan to in the next 2-3 years so there’s a lot to be gained from victory. 

  

Let the API wars begin!

By James Staten

Check out James' research

You should follow James on  here

Xen.org: Stop the Forking. Forrester: About Time

 Xen.org, the open source community behind the leading IaaS cloud computing hypervisor finally made a bold move today by stepping up to the plate of delivering a complete open source virtual infrastructure for cloud platforms. Prior to this release, Xen.org had been content to manage and maintain the core Xen hypervisor and let its partners all build solutions around it. The problem with this approach was that while the hypervisor itself was compatible between these solutions the infrastructure and how you managed it were not. Compatibility at these higher layers is what is necessary for the Xen community to attract an ecosystem of ISVs and end customers – no one wants to have to tune their software for each and every Xen variant. Now Xen.org is delivering a complete product that IaaS clouds can more easily deploy, Linux distributions can deliver and that end customers can install knowing that storage, deployment, monitoring, reporting, policy-based automation and most other management tasks are consistently executed. This also makes Xen plus Eucalyptus a more complete cloud-in-a-box, open source solution.

Xen.org even went so far as to make the official Linux kernel the domain 0 for this solution, addressing a contentious point that rival hypervisor KVM had raised as its differentiator. Xen.org vendor partners can still choose to leverage just the hypervisor for integration with their own infrastructure and OS kernel technologies (as Sun has done with xVM) but there will be diminishing value in doing so with products aimed at the mainstream market, as compatibility will be broken at these higher levels. Oracle is one worth watching as its OracleVM uses Xen plus Oracle Unbreakable Linux, which is based on Red Hat Enterprise Linux server. Red Hat is unlikely to add the full Xen infrastructure as it has shifted to KVM as its strategic hypervisor. Could this be the opening Oracle has been waiting for to break away from Red Hat with its Linux (Is Unbreakable SuSE on the horizon)?

Clearly this move benefits Citrix’ aims as well. The company carrying the Xen.org flag will be contributing all its cloud-relevant technologies to this open source project including its StorageLink, virtual switch, virtual appliance, and VMware-to-Xen VM conversion (Project Kensho) technologies. Many early IaaS cloud leaders have their own solutions in these areas but are likely to welcome this IP so they don’t have to maintain their unique solutions long term.

Hopefully with this move the days of cloud incompatibility, at least between Xen-based clouds, will come to an end just a bit sooner.

By James Staten

Check out James' research

You should follow James on  here

With DataSynapse, TIBCO Looks To Both Present And Future

TIBCO's acquisition of DataSynapse for a reported $28 million gives the company a development team and mature products with which to expand its financial-services revenue and flesh out its TIBCO Silver cloud computing platform. The acquisition means two things to application development organizations:

1. TIBCO will use DataSynapse's technology to support a variety of existing enterprise applications on its Silver cloud middleware. Silver, which is in beta test, today transposes new applications built in Java, C++, and several other languages to internal and public cloud infrastructure providers. DataSynapse's FabricServer software will help customers adapt a variety of existing applications to cloud infrastructure as well. These include Java applications based on IBM WebSphere, Oracle WebLogic, and Red Hat JBoss, Microsoft .NET Framework 3.0, IIS, and Windows SharePoint Services 3.0, SAP Business Objects, IBM Cognos, Informatica, and SAS Institute, and a variety of vertical-industry specialists. This is a substantial expansion of TIBCO Silver's value to customers.
2. TIBCO will now provide a product for running massively parallel workloads on grids of low-cost Intel servers. Data Synapse's GridServer is used for this purpose by the company's 83 listed customers, most of which are financial-services firms. TIBCO does not currently sell grid solutions, and so this will be add-on business, which is largely not integrated with TIBCO's integration, BPM, BI, and portal products. Financial services, life sciences, telecommunications, and public-sector organizations use such grids to run big models and simulations.

The big promise of TIBCO's acquisition of Data Synapse is obviously in the expansion of Silver its technology will enable. Forrester expects to see a tangle of Silver features from this acquisition during the second half of 2010 at the earliest, as TIBCO plans to ship the first release of Silver by the end of the second calendar quarter of 2010.

Forrester clients who have FabricServer installed today should feel good about this acquisition as it expands TIBCO’s reach for Silver and is thus strategic to its future direction. With stronger financial backing and the expanded sales force, TIBCO should be able to help push DataSynapse products well beyond its financial services beachhead.

The obvious question is around GridServer and if TIBCO will make the investment necessary to maintain as well as expand its presence in the grid computing market — or simply poach the technology purely for fleshing out Silver. We recommend GridServer customers take a cautious wait and see approach.

Customers considering TIBCO Silver should take a look at Fabric Server — making sure to understand its means of rehosting applications and how it provides shared services and scalability benefits as a sign of things to come from Silver.

By John Rymer and James Staten

Check out John's and James' research

You should follow James on  here

Amazon Web Services Is Addressing Enterprises' Top Issues

Attention enterprises — Amazon Web Services LLC is serious about wanting your business. Over the past 6 months the cloud computing leader has made several enhancements to its services that specifically address the security concerns of enterprise infrastructure & operations (I&O) professionals as well as security & risk professionals. With these moves Amazon is slowly knocking down all the barriers to corporate adoption of the Elastic Compute Cloud (EC2) and Simple Storage Service (S3). These moves are likely to take many corporations from test to deploy.

Its latest moves to provide virtual private LAN connectivity to Amazon Machine Images and support multifactor authentication for administration follow a string of moves that have been demanded by enterprise IT shops. The key moves started early this year with the announcement of reserved instances that put a bit more billing predictability in place and make it more palatable to think about deploying an application on EC2 for the long term (Amazon Web Services (AWS) recently lowered the up-front fee for reserved instances making the pricing even more attractive). This was followed by greatly enhanced logging and reporting that lets application development managers and I&O pros better understand what’s happening with their applications and proactively act against potential infrastructure issues. It even enhanced the administrative interface giving users greater control and visibility.

The Virtual Private Cloud (VPC) solution isn’t much more than a supped up VLAN implementation but is one that enterprises will find very easy to use as the solution is an IPsec VPN gateway that speaks Cisco IOS 12.4 and Juniper OS 9.2. And you can bring your own IP addresses so that applications inside your data center and AMIs at EC2 share the same address space and subnet. You can use your normal management, policy enforcement, intrusion detection and other tools across this VPC as well. You can’t fully integrate your S3 volumes into this VPC (yet) but if you highly leverage Elastic Block Store this is less of an issue. There may be some latency impacts to this solution, but it’s a great improvement.

They can’t do anything about the security risks of multitenancy as this is fundamental to the solution — and key to the economic benefits — but these moves, in combination, make EC2 a much more secure and transparent deployment option than ever before. For many Forrester clients, multitenancy may be a show-stopper for some applications but don’t reject Infrastructure as a Service (IaaS) compute clouds outright — a lot of non-critical or non-sensitive applications may fit very well on these types of environments and save your significant costs along the way.

Other IaaS clouds have similar capabilities to these just added to EC2. Some provide much deeper reporting, direct control over hardware firewalls and support for different MFA solutions so you definitely should shop around for the right deployment platform, but EC2 just got a lot more enterprise friendly.

Forrester recommends clients at least experiment with IaaS clouds today and that I&O publish a policy endorsing the use of these clouds for developer functional testing at a minimum. Now it is much easier to specify exactly how your use of these clouds should be configured to comply with corporate security policies.

By James Staten

Check out James' research

You should follow me on  here

VMware's Surprising Acquisition Of SpringSource

VMware, usually known for virtualization software, has come out of left field with a shrewd move purchasing app dev. company SpringSource, which is likely to be paired up with part of their VCloud and VSphere offerings as part of their ongoing construction of a new kind of operating system. A move like this suggests VMware is hoping to increase visibility and earn additional OS-relevance.

Hear more about VMware's new acquisition in the podcast featuring myself and fellow Forrester Analysts Mike Gualtieri and Jeffrey Hammond — download at the link below.

Click here to download the full podcast

By James Staten

Check out James' research

You should follow me on  here

VMware Completes The PaaS To vCloud

If you’ve been wondering why an infrastructure leader would acquire a developer framework, the answer is a bit more complex that what shows on the surface — and a lot more strategic. As stated in the press release and in the blogs by VMware CTO Steve Herrod and SpringSource CEO Rod Johnson, the acquisition helps by, “creating a single, integrated, build-run-manage solution for the data center, private clouds, and public clouds.” For the developer they will be able to use SpringSource tools to fully describe their application as a VMware vApp “a deployment blueprint that describes how the various machine images, middleware, and management components fit together and then we can take that blueprint and ‘make it so’ with a single click,” Johnson added in his blog. Infrastructure & operations professionals benefit because there will be less mystery to capacity planning, deployment and configuration, performance tuning and SLA creation for these applications, presumably making them easier to manage.

But that’s just the basics of this story. VMware has a bigger agenda SpringSource helps to fulfill making vCloud bigger than simply an Infrastructure as a Service (IaaS) alternative and keeping Microsoft at bay. Enterprises are already demanding that cloud environments and internal cloud solutions support their hypervisor standard VMware. So it wasn’t going to be a stretch to get vCloud adopted, assuming it delivered as promised. But the battle isn’t IaaS, it’s becoming the equivalent of the operating system for the next generation data center and you can’t achieve that aim without applications; and you can’t become application-relevant without being relevant to developers. While tools such as vCenter AppSpeed help to analyze the performance of applications so you could tune the infrastructure below them, merging this tool with SpringSource’s Hyperic gives visibility up into the application itself so performance experts from I&O and development can work together from the same information. There’s also room for presumption here that VMware will use the knowledge they will gain through this visibility (and the insights from the rest of the portfolio) to make vSphere and vCloud optimized for Java workloads. An integrated build-run-manage-deploy solution for Java sounds an awful lot like a strategy from Paul Maritz’ former company.

Microsoft Windows Azure is clearly the optimized cloud for .Net. Hook this together with Visual Studio, Hyper-V and System Center in the enterprise and you have a similar story. Both company’s offerings are incomplete today but the similarities are now clear.

The real question is who can spread their relevancy beyond their installed bases most effectively and since every enterprise has both, which you should you bet on. VMware clearly has dominant share in virtualized Windows workloads but that will wane over time as Microsoft cranks up its roadmap and marketing machine that story has already played out on VMware’s stock price. So VMware needs to drive up application stickiness. It’s been successful thus far in staying ahead of Microsoft with its roadmap and through building out its management portfolio. What it needs next is to get the OS out from between itself and the applications. Its first effort at this was its Virtual Appliance play with just enough operating system (JeOS). Step two will be tying SpringSource’s tc Server to vApp so that the OS in between is irrelevant for Java applications. Third is a play SpringSource has been driving for a while making its framework relevant to more than Java developers (or at least keeping Java developers from moving to other languages). These efforts thus far have included Groovy and Grails (a Ruby on Rails lookalike atop Spring), Spring .Net (for .Net developers) and plays toward ColdFusion, Adobe Flex and Javascript. Couple this with VMware’s application certification efforts with the majority of commercial applications and you have the beginnings of a very sticky platform.

Is it a platform worth betting on? Tell us your thoughts.

By James Staten

Check out James' research

You should follow me on  here