Dell is asserting that cloud computing is raising the danger of vendor lock-in, which the IT industry has been successfully trying to reduce, by "shoehorning" proprietary technology into private clouds.

Well, nothing has really changed: buyer beware still rules the day!

The whole point of virtualization - the core enabler of cloud computing - is to separate hardware architecture from implementation.  Yet there are many cloud solutions that are proprietary.  And it's not limited to private clouds: Amazon, GoGrid, OpSource and others have "innovations" like hardware firewalls, proprietary hypervisors, and unique storage architectures that lock applications into their clouds.   The solution is to go back to basics: stick with hardware that supports standards (like NFS), use a an off-the-shelf hypervisor (paid or free), and implement architectural elements that used to require dedicated hardware (like firewalls and load balancers) with software (paid or free.)  Then, you can replace the underlying hardware (in the case of private clouds) OR cloud provider and keep running.   This isn't rocket science: you do have to swear off the goodies that any single cloud vendor touts as a you-can't-do-without requirement... but it also sets you free of lock-in.

At ENKI we have tried to avoid proprietary technologies that impact our customers' implementations.  Yes, we have branded firewalls and we use Infiniband,  the new wave in datacenter interconnect, as well as other non-standard technologies.  But from our customers' perspective, they are not visible or do not put requirements onto their deployments.   We've avoided exposing our hardware firewalls to our customers since that dependence would prevent moving to another cloud.  We've hidden Infiniband and our SSD-backed storage behind industry-standard NFS.   And so on.   And we always use standard hypervisors (VMWare and soon KVM) since we ourselves learned the hard way that applications stuck in a proprietary hypervisor (like our older cloud technology, Applogic, which has a custom version of Xen) are difficult to support or move around.   

As a result, our customers - for our public or our private cloud services - know that they can easily change providers, hardware architectures, geographies as their business requires, while still enjoying the superior performance and service that ENKI offers.

James Urquhart asks, "Can any cloud catch Amazon Web Services?"

His answer: It's nearly impossible though it may happen eventually.

I couldn't disagree more with this analysis. Certainly, within a limited scope, Amazon shall remain dominant because of the reasons you listed. However, if Amazon were "all that", they would have captured a significant portion of the overall IT market by now, not just early adopters and developers. There are plenty of ways other companies can and will be able to compete with Amazon because as you pointed out, being the market leader pigeonholes you, and because Amazon's continuing and somewhat bizarre focus on an academically pure cloud product rather than a customer-oriented one. The points on which competitors may compete, just for starters, include...
- performance. Amazon's underlying architecture isn't performance-oriented, in part because of pricing
- price. Believe it or not, they're making a large markup on the service which can be undercut
- relationship Being closer to your customers than a web-based vending machine gets them to both spend more and be happier with the service
- flexibility. Hybrid private/public clouds, special-purpose hardware (yes, despite virtualization, hardware has an impact on applications!), custom IT solutions, etc. are all making money for a lot of MSPs, but the Amazon cloud model doesn't allow for them
- reliability. Need I say more? There are plenty of different reliability/price points available for entering the cloud market.
- technology. Amazon's combination of server-for-hire with distributed storage isn't the only way to implement cloud. Ask VMWare. The other alternatives offer advantages that other service providers can exploit.
- market segmentation. Amazon has done little to address the low and high ends of the cloud market, despite inking some large deals. We've won a number of important deals from Amazon because their price point, use model, or instance inventory weren't targeted at the particular customer.
There are many more differentiators, but suffice it to say that the world would be a boring place indeed and cloud would be facing a much longer adoption cycle if Amazon were the only provider, just it would if Amazon didn't exist.

Today, HP warned that cloud computing early adopters faced security risks.

But the timing and motivation for such a warning are suspect: the risk that HP is warning about is already present for for organizations that don't have an adequate security plan.  Where their infrastructure is - virtual, physical, cloud - is of little consequence if there is no application-dependent security plan in place.  Once the application-dependent plan is created, it must be adapted to the deployment strategy.  For example, at the simplest level, there must be a firewall in place to block unused ports, which will be a physical firewall for a physical deployment, or a virtual one coupled with vlan controls for a virtual or cloud deployment.

This is no different than the requirement that there be a business continuity plan available to ensure uptime.  During the failure at Amazon's East Coast datacenter two weeks ago, customers which had a business continuance plan in place barely noticed the outage, while those who had hoped that Amazon's services included solving their business continuance requirements were disappointed by a multiday outage.

So the risk of the cloud is really more that, with its point-and-click deployment ease, it can lull users into ignoring or avoiding their fundamental business-critical IT responsibilities, especially since avoiding these responsibilities can represent a large fraction of the cost savings they see from cloud deployment.   Because these security and in business continuance plans have to start with the application, the cloud-buying organization cannot expect an infrastructure or platform cloud vendor that they interact with purely through a self-service portal to "fix" their security and uptime exposure: they must either retain the IT expertise in-house to use the cloud properly, or find a managed cloud vendor that offers Virtual IT services as part of their management suite.

So in a sense, HP is belaboring the obvious: large enterprises already have enough IT expertise to use the cloud safely, and small enterprises are well warned not to ignore the risks of skipping business continuity and security planning.   A possible reason for HP's announcement is that HP is late to the cloud game.  While HP's competitors were fielding cloud services, HP was focusing on providing hardware and software to cloud vendors, and it still does not have a credible cloud offering, according to outsourcing experts.  And there's nothing wrong with that: HP makes fine hardware and software.  But this sounds more like a stalling tactic to me: "stay away from the cloud until we have a better solution to your security problems than our competitors!"

ENKI believes the future of cloud is increased usability, availability, and performance that will enable true enterprise computing.  The key is combining application-based operations services that companies have had to provide for themselves until now, with on-demand computing.  We call this Cloud 2.0.

When ENKI first entered the pay-as-you-go, on-demand, managed computing services market 4 years ago, the term "Cloud Computing" wasn't even in common use.  Since then, Cloud Computing has come to be commonly accepted to denote on-demand computing delivered over the Internet.   Businesses using applications in the cloud (SaaS, or Software-as-a-Service) have learned to expect that they will not need to add any staff to manage those applications.  However, companies that want to use cloud computing to replace their in-house or colocated datacenters for production computing (IaaS, or Infrastructure-as-a-Service) have not been seeing the dramatic cost redutions promised by cloud vendors who only replace physical hardware with remote virtual hardware.  Instead, they have simply replaced staff who manage physical servers with staff who manage cloud servers.

ENKI has four years of experience delivering cloud computing that begins with outsourced virtualization, but adds operations services including system administration, security services, incident response and software release management.  By offering SLA-focused management for their cloud deployments, ENKI enables our customers to enjoy improved service levels while reducing their management costs using the same on-demand model as for their computing.

We call the combination of cloud computing with service-level focused management, Cloud 2.0, as it fulfills the original promise of cloud computing to not only deliver improved services but also overall cost reductions.  At its essence, Cloud 2.0 is Virtual IT, which is the tag phrase we have started to use to communicate the value we provide our customers.

Great article in Network World. The article shows why Cloud Computing with a service component is the wave of the future. Not every company is going to have someone who is equipped to manage the internal infrastructure in the cloud. However in order to compete in today's business environment, virtualization is important. The companies that are succeeding are doing more with less. Which then points to a virtual infrastructure with a little added bonus, support.
The new breed of cloud provider offers the service component which enables any company to run as efficiently as a nimble web 2.0. Enjoy the article. By Denise Dubie , Network World , 05/20/2009
LAS VEGAS -- Despite the reported benefits of virtualization, a majority of IT managers polled at Interop this week say they experience problems with the technology and don't always realize the cost benefit.
Network Instruments polled 120 network managers, engineers and IT executives at Interop to learn how IT organizations are putting server and desktop virtualization technology to use. Fifty-five percent reported they virtualized mission-critical servers, including e-mail and Web servers, and another 50% said they run DNS and DHCP servers on virtual machines. And nearly 40% have already extended virtualization to their desktop environments.
Yet 55% told the network analysis vendor they experience more problems than benefits with the technology, while the remaining 45% said they had realized the benefits of virtualization. Among the problems were a lack of visibility and tools to troubleshoot performance problems in virtual environments for 27% of respondents. More than one-fourth of those polled at Interop cited a lack of training on virtual infrastructure and 21% expressed concern over an inability to secure the infrastructure.
For nearly 60%, the primary problem with virtualization was a lack of experience to appropriately manage the technology and nearly 50% said that technology implementation costs were too high, according to Network Instruments.
Managing the technology doesn't seem to have become less challenging for network managers. Last year, nearly 40% of 117 network managers polled at Interop also listed virtualization as the emerging technology that represents the "greatest monitoring challenges," according to a joint survey conducted by Network Instruments and NetQoS.
"Not surprisingly, a high number of companies have deployed critical network services on virtual machines," said Charles Thompson, product manager at Network Instruments. "The number of organizations without appropriate monitoring tools, however, definitely caught my attention. Without proper tools, application performance can unnecessarily degrade and network teams waste hours troubleshooting."

Today I read a blog from David Linthicum in which he said that the corporate datacenter is safe from cloud computing because issues around compliance, privacy, fear, and control will cause companies to want to keep their computing in-house.   However, I'm not sure that this is an us-vs-them issue, and there is a good reason that the corporate datacenter will stick around for a long time: it's cost-effective, though only if well-run, and large enough to provide the economies of scale that make it so. What cloud computing will do is provide a benchmark for the cost-effectiveness of the internal datacenter by providing an alternative.

I think the discussion on whether cloud computing will kill the datacenter should be reframed: it's more like "The king is dead. Long live the King!"  Once the hype of cloud computing has been overcome, I think it will be shown to be a set of *technologies* that enable a set of *services*, rather than a specific set of branded products as it is being marketed now.  And those technologies are just as useful inside the corporate datacenter in a "private cloud" as they are in a service provider's datacenter serving up a public cloud.
The issues around compliance, privacy, fear, and control will have to be played out inside the enterprise just as well as they are in the public forum as these technologies do actually provide advantages no matter where they are applied and will create revolutionary change in any case. What are these technologies?  Virtualization of compute, I/O, and storage; pay-per-use microbilling; IT workflow automation; and application template libraries.  If IT staff think that their jobs are safe from radical change just because their company isn't outsourcing its computing to the cloud, I think they'll be in for a surprise because these technologies - collectively "cloud computing" - will impact them anyway.