Digital Walkabout

Google Wave, communication/collaboration, and the new mashup

The announcement and demo of Wave, Google's real-time communication and collaboration platform, at the IO conference was met with a standing ovation. Representing the federation, aggregation and mashup of communications protocols and paradigms, Wave illustrated the future of web apps and interoperability.  

Tim O’Reilly wrote the following walk-though of Google Wave, a good way to familiarise onself with it and understand its initial value proposition: http://radar.oreilly.com/2009/05/google-wave-what-might-email-l.html. The creators, Lars and Jens Rasmussen, asked themselves, "What would email look like if we set out to invent it today", a very crisp and concrete way to think of a broad problem.

Yet what is more remarkable is how Google itself has recognised that real-time communication and collaboration extends far beyond the email paradigm. In providing a platform & protocol  based on hosted XML documents (“waves”), a  focus on massive concurrency and low-latency updates on top of a decentralized XMPP architecture, they have really upped the game for dynamic, real-time data on the internet. In choosing XMPP as the foundation for this decentralized, interoperable environment, they’ve accepted what the Web 2.0 community has understood regarding how the potential of the Web can only be realised with a balanced compute paradigm. In taking federation, aggregation and mashups to the next level, they’ve made it clear that email is not a suitable backbone for interoperability, communication and collaboration – it is a remarkable medium, yet suitable only for certain modes.  XMPP is far more suitable interoperability layer, allowing Google’s Wave federation protocol to provide a richer level of interoperability by including additional auto discovery of IP addresses and ports using SRV records and TLS authentication and encryption of connections.

This approach heralds yet another challenge to Microsoft and traditional approaches to both the back office and the private Cloud. For the back office, message based data management will be hard pressed to approach the level of integration and interoperability of Google Wave. This itself enables private Clouds which can more easily integrate and interoperate with one another. In a sense, rather than Google Wave representing “what email might have looked like had it been invented today”, it represents the full potential of the mashup – policy-based dynamic composition of applications on the web. Whilst Microsoft is enabling Office in the Cloud (MSOnline), Google is enabling anyone to assemble their own custom back office in the Cloud out of any SaaS that are available. I see this moving the needle on truly dynamically-composed applications and balanced compute - and as a very crisp and concrete response to "why can't we just do this with email"?

Focus on growth, business processes helps users stay afloat

As companies look to implement new applications in the current economic downturn, some key factors to consider are growth, business processes and the reuse of as many resources as possible, SAP users suggested last week at Sapphire 2009.

Elizabeth Arden Red Door Spas, for one, decided to look at SAP to run its entire business -- with the exception of a point-of-sale and reservation system -- for two main reasons.

"I think we were looking for one platform to have one version of the truth and to really improve our business processes," said Paul Kaczmarek, CIO for Red Door Spas.

In a session titled "Finding Opportunities in Today's Economy", Kaczmarek, along with two other SAP users and Verlin Youd, senior vice president of global trading industries for SAP, discussed how to help drive quicker ROI when making investments in economic downturns, as well as how to be prepared for the economic upturn.

Kaczmarek said that over the past 10 months, he's learned that always having a Plan B in place is crucial when building a business case for new implementations, but where a Plan B once posed as an alternative option, he said, it now means repairing what's in place.

"You've got to come up with a compelling case every single time," he said. "It's been kind of fun to be able to push those things back that don't make sense -- fun in the sense that this really makes you become better."

The path to SAP

But technology buyers should not let the economy deter them from looking ahead, the panelists warned, as they still need to consider future growth. For arts-and-crafts retailer Hobby Lobby, SAP fit the mold.

"For us, something that we could grow into the future with was really important, but also something that wasn't going to break the bank out of the box," said Jeanne Cotter, CIO and vice president of Hobby Lobby. "SAP was kind of a natural fit. We have kind of a Lego approach to everything we do -- we start with the core foundation and then build on it; and, really, enhancement packs were the deal breaker for us."

She added that in order to grow business, the company needed a platform that could support that growth.

"This economy is kind of a double-edged sword, in that there are real estate opportunities that exist today that didn't exist in the past, from a retail perspective," Cotter said. "So I think, foundationally, knowing that we were going to have to invest back and be able to grow was key for us."

Hobby Lobby had outgrown its previous ERP solution, putting it on the hunt for something larger. But payroll was also causing a large amount of grief, she said.

Meanwhile, Jean Luc Vanderbroek, director of finance at Delhaize Group, which oversees such supermarket companies as Food Lion and Hannaford, said 2005 was a key date for the company -- that was when it decided to revamp its organization by going with SAP's business transformation program.

"We switched from a pull to a push organization," Vanderbroek said. "In other words, from a decentralized to a centralized organization."

Delhaize Group selected SAP to replace its difficult-to-maintain legacy systems and improve the quality of the company's information, he said. It also needed to integrate financial processes within the organization.

Vanderbroek said it was difficult to adapt to change in the economic environment at that time (2005) with the legacy systems, which constantly needed updating.

Making a solid business case

Red Door Spas used a value-engineering study it had originally done with SAP as its benchmark to ensure that SAP would meet expectations.

"We actually used that study to get the project approval from our board of directors," Kaczmarek said. "For the past two years, we've looked back at that study and said, 'Did we actually realize the gains we thought?' So far, in general, we're pretty much ahead of where we thought we'd be from the value-engineering, which is a compelling story for our board of directors."

Bumps in the road

Further, training staff and employees already comfortable around an existing environment to something new can also pose a challenge. But while training staff can be challenging, it may not be impossible, as long as everyone's on board.

"We've not had to eliminate any positions because of SAP, and we made that commitment up front," Kaczmarek said.

Still, he said, Red Door Spas does tend to look for outside help if additional training is needed. The company did have to hire a Basis administrator. Kaczmarek said the company's employees were well aware of the options once the decision was made to go with SAP. "Right up front, we told our staff, 'Either embrace this, and if you embrace it, we'll train you; or if you don't want to, then there's not much we can do about that -- there's a door, use it.'"

But the road was not so smooth for Delhaize. Vanderbroek said that when the company switched several of its legacy systems to ERP, it was "very difficult for people to change, because they have a lot of experience with one screen made by the company, and when you organize the go-live, the job is different. That was the main impact in our project, because it was difficult to maintain."

All other things considered

Despite looking for a point-of-sales and reservation system, Red Door Spas is still learning SAP, and that will continue to be the company's main focus in the coming months, Kaczmarek said. But, he added, down the road the company may look more at "selling more products online and integrating that with our warehouse that's running off SAP, so some of those types of next steps as we learn more about SAP and its capabilities."

"Again, we have a legacy system that's old and unobtainable basically," he said, "and we're looking at some newer technologies there, and then integrating that with SAP as well."

The notion of what's to come hasn't escaped Cotter's mind, either. Hobby Lobby, she said, is looking to gain more ground in the cloud computing arena to make way for the large-file storage needs of the creative artwork done by the company. Also, she said, because Hobby Lobby oversees smaller retail units, the company will look to expand on them in addition to further growth in the rest of the company.

Private Cloud may be a preferable option for Ministries

Private cloud offers almost all the benefits of a public cloud, without the security and privacy headaches, per Jim Whitehurst, President and CEO of RedHat.

If an agency is running more than 1,000 servers, it could save money and become more flexible with its processor resources by building an internal cloud-computing infrastructure. A private cloud could offer almost all the benefits of a public cloud, but without the attendant security and privacy headaches. "There is a significant amount of value in a cloud infrastructure for a single entity that is running multiple programs and multiple data centers," Whitehurst said, speaking at the Federal [Information Technology] on a Budget Forum held this week in Washington.

"What is the benefit of a public cloud? For most large agencies in the government, or even medium-sized agencies, there is a not a lot of purchasing cost advantage with going to a third-party cloud," Whitehurst said. "The real benefit is getting high utilization of your existing infrastructure and flexibility around that."

At present, Red Hat has more than 50 enterprise customers with private clouds. "They are seeing huge benefits [by] running clouds themselves," Whitehurst said. He mentioned that one customer he spoke with, a chief information officer of a large organization running 25,000 processors, told him that public-cloud services such as Amazon would not provide much of a cost advantage to his organization, as his company could obtain servers at almost the same cost as those providers.

To set up a private cloud, an agency would pool all of its servers and offer the processing power to each department. It could enjoy the economies of scale that clouds could provide, and not worry about the privacy and security concerns that dog public-cloud providers, according to Whitehurst. It could also sign agreements with other agencies or organizations whose peak processing times are different from its own in order to have extra computing capacity on hand for those periods with the heaviest workloads.

Whitehurst said that today's public-cloud offerings are mostly in the "vapourware" stage, adding that he sees this as not so usefuk for enterprise users, again due to the outstanding security and privacy issues. Building a private cloud, however, will help agencies get prepared for such time when and if public-cloud offerings become an alternative.

Cloud Computing growth predicted by Coda Research

Scepticism about cloud computing abounds, yet despite the critics, there’s a new prediction that in the next six years (2015), Cloud Computing will make up 17% of all IT expenditure worldwide. A note of caution has been sounded, a warning that IT suppliers who do not begin to invest significantly in cloud computing and “learn from the demands of users” will find themselves increasingly marginalised, with “serious consequences for customer loyalty, revenues and profits.”

This take on the development of cloud computing comes from industry analysts Coda Research Consultancy, who reckon there’s nothing short of a “radical transformation” going on across IT, and, “it’s called cloud computing.” “For many commentators, cloud computing is the future of all computing, whilst for other commentators, much coverage is hype, and cloud computing is not significantly different from current and even older technologies and ways of doing computing.“

Despite strong growth predictions, senior director at Coda, Steve Smith, says many questions remain around revenues, cost, security, vendor lock-in, standards and the roles it has to play among different organisations. “IT users who do not begin to evaluate cloud computing for their businesses risk seeing their competitiveness, productivity, customer loyalty and revenues all decline relative to their competitors. The 17% growth in cloud computing over the next six year predicted by Coda, if achieved, would bring revenues worldwide  to $180 billion by 2015, from a mere $46 billion just last year.

According to Smith, who authored the report on cloud computing, the greatest growth in expenditure upon cloud services will be around collaborative applications, IT management applications, personal and business applications and storage.  “Thus, IT suppliers must develop strong cloud application propositions, especially through providing software as a service (SaaS) platforms and partnerships to such providers.” Smith says that in the short to medium term, growth in public cloud usage will be greatest among small to medium sized organisations, and growth of cloud computing among larger organisations will be slower, with these more likely to employ hybrid and private clouds as the market matures toward 2014.

“Significant vertical market specialisation will begin to take place among cloud service suppliers from 2011, particularly around manufacturing, health, energy, finance and media,” predicts Smith, adding that there is also a significant risk that smaller and more recently launched IT suppliers will “leave behind long established suppliers in the innovation and provision of cloud services.” Smith also says it is vital that IT suppliers work together to increase trust in cloud services among IT users and “act on their concerns.”

“Issues for buyers include security, reliability and transparency, along with the development of common standards that enable interoperability; portability of applications, software and data across cloud services; and management consistency,” observes Smith.

On the question of security risks associated with cloud computing, Smith says these risks “tend to be exaggerated,” and he adds, “rather than increase risks related to security, availability and performance among others, cloud computing is likely to significantly reduce these.”

According to Smith, absence of upfront investment in infrastructure and software, absence of upfront commitments and usage contracts, and what he describes as “few minimum use costs for cloud service buyers” mean that cloud suppliers will become more “susceptible to turnover and revenue fluctuations” and, he says, “the balance of power will shift to buyers as competition, cloud specialisation and ease of migration across cloud services develop. Interestingly, in the introduction to his report, Smith describes cloud computing as big news and points to the experience of Amazon Web Services, which he says has a peak usage of 80,000 work requests per second, 52 billion files stored and thousands of developers using it.

“When Amazon's virtual computing service, the Elastic Compute Cloud (EC2), launched three years ago, it tended to attract low budget start-ups and web 2.0 entrepreneurs. Now it works with organisations from the New York Times to Sun, Oracle and IBM,” Smith says, adding that “organisations spent $383 billion on business applications, infrastructure software, application development and deployment, servers and storage in 2008.”

How Technology is Drastically Changing the Future of Business

Todays announcement of 3Tera's AppStore is just another mounting piece of evidence of how technology is drastically changing the business landscape of our future. 3Tera joins Amazon in delivering "cloud ready" components that enterprises, big and small, can quickly get up and running with very little upfront costs. Take at look at 3Tera's website and you see that in Q3 of 2009 they will release many new components..

The AppStore catalog spans all types of data center infrastructure elements and complete applications, from networking components (such as load balancers, firewalls, routers, traffic managers, content switches), server components (such as web servers, directory servers, databases, application servers) and storage solutions, as well as application software stacks, management and monitoring tools.

Expect more software vendors to start partnering with cloud providers like 3Tera and Amazon to create their own cloud ready images on these platforms.

Cloud Computing

So what does this all mean? As the cloud matures over the next several years, the playing field for businesses starts leveling off because of the low cost to entry and the quick time to market that the cloud will be able to offer. Huge capital investments in physical infrastructure will be a thing of the past or left for those with huge amounts of assets still depreciating on the books. But it is not all about the hardware. We are seeing much more enterprise software being "componentized" and delivered as a service. As time goes on, expect the industry to start producing more standards and better interfaces. Integration will become less of a burden as the interfaces become simpler and the industry matures.

Less hardware to buy, patch, and administer coupled with less non-core business software to build will enable startups, small and medium sized businesses, and even large businesses with an agile architecture to move quickly to chase new business opportunities, streamline operational efficiencies, or integrate with partners and/or customers. But there is more to it!

I believe the commoditizing of software and hardware that cloud computing will bring over the next few years will change the way businesses think (see next point).

Standardizing business processes

The ability to outsource business processes to the cloud will force companies to rethink how they view their business processes. Business owners will be forced to justify any exceptions or proprietary processes that prohibit the company from leveraging "cloud ready" business processes. As more companies continue to standardize on business processes, the customer experience becomes more consistent. When this happens, the business processes themselves become a commodity and the importance of customer service rises dramatically. Which leads me to my next point.

Brand Recognition

When the underlying infrastructure, software, and business processes become a commodity, products and services are forced to compete at low prices. This puts many products and services on a level playing field. That increases the importance of the brand which is why companies are investing so much in marketing these days. Expect that trend to go up as the the marketing landscape changes, which leads me to my next point.

Social Marketing

Companies will continue to invest heavily in a toolbox full of social networking tools. Viral social marketing tactics, ones that leverage the power and voice of consumers spreading the word through their own personal networks, will be the tool of choice for marketers. People are watching TV less, reading the paper less, and spending way more time on the Internet. Companies are recognizing this and will continue to target communities on Facebook, Twitter, YouTube, and many others. It is already happening. Look at what Pepsi is doing.

Government Influence

The current administration is a huge advocate of social software and cloud computing. Their use of blogs, Flicker, YouTube, Twitter, Facebook, and streaming video is making these technologies increasing popular. The belief in a more open government is the driving force for the use of these technologies. If the government can implement these technologies, surely our enterprises can. The administration went one step further and created Data.Gov. This initiative will lead to more standardization in data and data services and will force various agencies and companies to change their existing practices in order to integrate with this massive public data source.

The government also has a huge cloud computing strategy which should help push the industry to standardize and address the challenges of privacy, security, compliance, and integration. The government also has a ton of money behind this initiative which should help drive much more innovation in the cloud computing industry.

There is no denying it

So as we all argue about the value of Twitter, Facebook, Cloud Computing and the like, the reality is that they are drastically changing the way businesses think and act. More importantly, customers and workers are actually finding themselves more productive at home using new technologies and are applying serious pressure to companies to sacrifice controls and security in exchange for empowerment and agility. All of this is creating more demand for architects and security experts, and driving down the demand for system administrators and development. IT jobs are not going away, but the roles are shifting. It may not be evident today but follow the trend over the next few years. Many companies will outsource more IT functions then ever before. Smart companies will internally focus more on innovation and less on keeping the lights on.

I see it all changing right before our eyes. We are in the early stages of this transformation and many folks may disagree with this assessment. I truly believe that this administration's investment in technology will drastically increase the rate at which businesses transform to a more outsourced, integrated, and socially connected enterprise.

AWS Cloud Data Import/Export

Amazon Web Services (AWS: news, site) has available a limited beta for their new AWS Import/Export service, which provides capabilities to move content in and out of the cloud in short order.

Using Amazon's high speed internal network, one can move data onto, or off of, just about any storage device — as long as it meets AWS compatibility requirements.

It is fairly academic to understand why this capability is desired, but Amazon does offer a few ideas:

·         Data Migration: Loading data into the cloud for the first time is greatly simplified and has never been faster.

·         Disaster Recovery: Using Amazon S3 for storage of websites for disaster recovery purposes? Now one can get their website back up and running faster.

·         Offsite Backup: Simple and straightforward to use S3 to store basic backups, whether full or incremental.

·         Direct Data Exchange: Partners send data on devices on a regular basis? Now they can send that data directly to one's S3 bucket.

This new service basically works by putting data on a storage device, then shipping that device to Amazon Web Services to process in their secure data center. Of course there are a few other steps to complete prior to shipping off the device.

The beta is currently limited to importing data to AWS for US buckets. The export capability and support for EU buckets will be available later.

Mashups, Automation and the Rich Client Cloud

9 Amazing Software Mashups – Killer Free Apps that Work Better Together

This is the sort of thinking driving users of both Software-as-a-Service and Service-as-a-Service – mashing up full-on apps or finer services  to develop/compose/orchestrate more capable applications.

Automation/BPM has been driving this for enterprise applications for the last 6-7 years. What the Cloud brings to this is co-opting the same protocols and transports, whilst opening up their meta-processes through the distributed architecture of Rich Client Cloud.  The same mashup concept has appealed to automation, yet without breaking free from the prevailing (although not entirely dominant) host-bound architecture, this next-gen Automation could not reach beyond silos within the enterprise. The Rich Client Cloud breaks down these silos and enables this next-gen Automation – Cloud Automation –  to achieve mobility, now appealing to the consumer as well as the enterprise.

The better known Mashups have focused on Service-as-a-Service in the Public Cloud, yet there are just as many Software-as-a-Service and Service-as-a-Service Mashups in the Private Cloud. Software- and Platform-as-a-Service providers such as SalesForce.com (Force.com), Microsoft (Azure + MSOnline) and Intuit (IPP) have focused on Mashups as a pattern for developing value-added applications based upon their offerings, whether Public or Private Cloud. With the growing convergence of these efforts with Cloud Automation, this will highlight the value proposition of Cloud for application and process development.

Balanced Compute: Is RIA still relevant?

Reading Chad Udell's entry The RIA is Dead, Long Live the RIA, I could not help but reflect on whether balanced compute has gained enough traction to be considered the prevailing metamodel for Cloud and Web applications. After last week's Google outage, and the minor row it started regarding Cloud and RAS, it should be clear that there is still a long way to go.

The fact that the Google downtime raised so many questions about Cloud reliability, is on its own indicative that a majority of both IT and Web pros still consider the Thin Client, Hosted Services model to be representative of the Cloud. It should be clear by now that with RIA, Cloud and Web aps have advanced beyond Thin Client technologies. Whilst there are a considerable number of successful Hosted Services projects (in particular those targetting task-orientated, always-connected applications), there are an equal number of failures (in particular those targetting complex applications, graphically intensive applications, mobility, hot-desking, contract/offshore, etcetera). Whilst there continue to be architects who pitch JEE as a sole platform for Cloud and Web applications (e.g. Servlets/JSPs pitched as viable presentation technology, or EJBs as as viable framework, for consumer or enterprise apps), RIA still need to be distinctly identified and held up for comparison. 

This may be hard for many of my colleagues to grasp, given how we collectively have been a part of the growing success and adoption of RIA among Web and now Cloud developers. When one is in the midst of a new metamodel and technology direction which gains traction and successes at a rather quick pace, it is hard to see why that movement might not yet be truly 'mainstream'. Any new metamodel and companion technologies take time to build enough critical mass to be considered prevailing. Cobol was the prevailing development language well into the 90s, and Java tooks three and a half years to gain critical mass in new development. JEE had to wait for 1.3, released 5 years after Java's debut and 3 years after EJB's. The Web and Enterprise development now has a large body of JEE developers, of whom it is estimated the majority have experience solely with Sun's technologies in one aspect or another (e.g. either EJB or Servlets/JSP to the exclusion of, say, anything else). That's a body of EJB developers still wrestling with Web Services, REST and a body of Servlet/JSP developers wrestling with AJAX. If we think carefully about this, we should reasonably expect, as a very rough estimate, that it would take anywhere from 3-5 years for this to likewise happen with a new metamodel which successfully drives companion technologies.

The list of Hosted Services project failures helps to complete this picture. The primary reason these projects fail is that they "put on the wrong trousers" - that is, focus on technology rather than objectives, scenarios and architecture. The result is that they apply the wrong architecture to address their requirements, and cut-and-paste requirements without understanding the fundamental differences, say, between Web and Desktop architectures. Were the various Cloud and Web architectures were well understood in these cases, it would be difficult to ignore how these are applied to various problem sets and scenarios.

I thus see that it will take some more time for RIA to gain the level of traction needed for it to become the dominating model in the Cloud and Web architecure consciousness. Then I think it no longer needs to be referenced distinctly. Is that important? Indeed it is - once it becomes integral to this consciousness, the industry will cease to think of Cloud and Web models monolithically. That day those of us driving RiA can wear (yet never rest on) our laurels.

Microsoft Azure seks to leverage the best of Cloud architectures

Microsoft Azure looks to leverage best of Cloud architectures

 

There’s much to be said about Microsoft’s Windows Azure, in particular, regarding how it represents a balanced compute model and seeks to maximize closest resources. At the same time, Microsoft is always keen to ease the development process for any of its initiatives and it has done no different with Azure, integrating well with Visual Studio, providing APIs to support plugins and extensions for IDEs such as Eclipse, and extending language support beyond .NET (PHP support went to CTP last week).

 

The blog entry from XConomy covers the impact of Azure and how it leverages the best of Cloud architectures:

“The idea of computing “in the cloud” is certainly, well, in the air—to the point where I think people are starting to become allergic to it. It seems that “cloud computing” has become the new “nano.” Putting the term in front of whatever your business is doing doesn’t necessarily make it a good (or even viable) strategy. And the concept—making data storage, computing power, and software applications available over the Internet via remote servers—certainly isn’t new.

But if Microsoft is betting heavily on it, you know it has gone totally mainstream. Last October, Ray Ozzie, Microsoft’s chief software architect, announced Windows Azure, the company’s upcoming cloud-based operating system, and Azure Services, which is meant to allow software developers to build and run software applications hosted on Microsoft servers. The platform is still in the “community technology preview” stage, with a more general release planned for later this year. There has been a lot of speculation about what Microsoft’s cloud-based products can and can’t do—and so far the top brass, including Ozzie (who just spoke at Friday’s Technology Alliance luncheon in Seattle), have kept mum on the details.

Enter Doug Hauger, general manager of cloud infrastructure services and former chief operating officer of Microsoft India. Hauger gave a talk last Thursday evening, organized by the Washington Technology Industry Association (WTIA), in which he spelled out Microsoft’s world view of cloud computing. Hauger said his goal was to convey, “What are we thinking about cloud, and what should people be thinking about?”

It was the final event in a four-part WTIA cloud computing series, which included talks by Amazon, Google, and Terremark. Thursday’s event also had a demo from Ian Knox of Seattle-based Skytap, showing how companies can run Microsoft’s development software in the cloud, and use Skytap’s cloud-computing platform to set up virtual environments for doing testing and debugging; Microsoft is one of Skytap’s partners. But what most people were wondering was how Microsoft is positioning itself versus Amazon and Google in the competition for cloud services.

Hauger, a Boston University alum, didn’t say anything particularly earth-shattering, and he didn’t give very many specifics about Windows Azure—which is fair enough, since the product isn’t out there yet. He did clarify that it’s pronounced “AZH-ure,” with the emphasis on the first syllable (sort of rhymes with “badger”)—at least that’s how Hauger says it. Here are my top five takeaways from his talk:

5. “CIOs are incredibly confused” about the cloud, Hauger said. “IT vendors are saying, This is an evolution,’” from the mainframe computers of the 1970s, to client-server systems in the 80s, to Web development in the 90s, to cloud services in 2009 and beyond. “But all this stuff still exists. Mainframes are still here today. There are shared pieces of all this together, and you have to take all this into account as you think about cloud services.”

4. For chief information officers and chief technology officers, all of this boils down to four top-of-mind issues that Microsoft is trying to address: portability (hosting data and applications on-site or off), interoperability (putting data storage on Amazon Web Services, say, and computing power on Windows Azure), manageability (everything has to be consistent and coherent), and security (corporate customers have compliance and audit requirements).

3. Cloud customers will trust Microsoft—once the company delivers its product. “What is interesting and a huge opportunity is, [companies] are advancing very quickly into this space,” Hauger said. “We’re very, very early, but people are starting to get educated.” Cost is the key driver, and trust is no longer an issue. “Customers are telling me, ‘There are only one or two companies we’d trust.’ They just want us to get our technology act together and deliver something,” he said. Hauger cited Microsoft’s medical-records platform, HealthVault, as an example of Web privacy and security done right. “We’ve been through that hoop.”

2. Microsoft thinks about its cloud-based products in terms of finished services (like Windows Live or Exchange Online) and a developer platform—and Azure will do both. Hauger described the “four pillars” of Azure as scalable hosting, automated service management, durable storage, and a rich developer experience. The latter will enable engineers to “build applications in a seamless way and publish them in the cloud.” Which all sounds a lot like what Amazon is already doing, which brings me to…

1. What’s special about Azure? In his talk, as I understood it, Hauger hinted that Microsoft’s advantage over Amazon Web Services was the former’s extensive partner ecosystem and its service management. Its differentiator versus Google App Engine is that Azure will be more geared towards big corporate customers and will work with many different programming languages.

“The business model is disruptive,” Hauger said. “The technology is not as disruptive.” When I spoke with him after his talk, he stressed that Azure will provide “complete integration across all platforms.” He also stressed that customers should spend time learning about Amazon’s offerings, and see how Azure measures up. Asked to sum up specifically how Azure will compare with Amazon Web Services and Google App Engine, Hauger said it will combine the best features of both—the infrastructure of Amazon and the scalability of Google—and will work across many different operating systems and programming languages.

The proof, of course, will be in the pudding. Hauger reiterated that Azure is still in its preview stage, where it is gathering feedback from free trials with customers around the world. “We’ll start charging later this year,” he said.”

I would add that: 

5. The McKinsey report highlights the confusion surrounding Cloud, which extends beyond integration and legacy systems. Until CIOs firmly understand that Cloud is more than Hosted Services, this confusion will be faced at every Cloud opportunity. With Microsoft Azure and like architectures, this will be easier to accomplish, since these exemplify the concept of balanced compute and closest resourcing.

4. What Microsoft terms “portability”, the industry terms “mobility”. I would also add that Consumerisation and Replaceability would provide the fifth top-of-mind issue.

3. This is the critical issue – Microsoft (and others) have to deliver to build trust in their products and confidence in the model/architecture.

2. & 1. Offering a complete PaaS to IaaS solution – development, provisioning/deployment, management/monitoring – bridges the gap present with both IaaS- and SaaS-only providers: How to get up and running with Cloud Apps in as scalable and smooth a manner as possible, which include re-purposing/re-packaging apps for the Cloud platform. This also bridges the gap present with SaaS-only providers: How to get “my” apps in the Cloud, and manage that Cloud as utility computing (application resources on demand) rather than utility applications (applications on demand).

 

These points are key to a) enterprise adoption of CC and b) consumer services realizing the potential of CC. What is most encouraging about Azure is that it is further along in terms of availability than like platforms/architectures. I am keen to see a number of vendors providing Rich Client Services platforms which enable users, developers and IT to maximise best processing - i.e. processing is distributed, and handled where and when optimal.

MMS: Microsoft outlines strategy for Public and Private Cloud

Microsoft’s strategy for public and private Cloud:

http://www.networkworld.com/news/2009/050109-microsoft-cloud-management.html

http://sev.prnewswire.com/computer-electronics/20090429/SF0737029042009-1.html 

At MMS, Microsoft laid out its its vision of corporate IT of public and private Cloud computing, on the heels of VMware’s announcements around vSphere. Microsoft’s focus is aligning its existing array of infrastructure, management and applications with Cloud, with private Cloud built on the Dynamic Systems Initative (DSI)/Dynamic IT and Windows Server. "Think about the private cloud as your internal data centre optimised for performance and cost and with the compatibility with your existing applications all built on tools you know today - Windows Server and System Center," said Bob Kelly, Microsoft's corporate vice president for infrastructure server marketing, at his MMS keynote. "We believe that business enterprise customers will demand the same levels of reliability and predictability [found in the public cloud] for their internal data centres."

Whilst little of this is new, Microsoft is clearly focusing its marketing and positioning on Cloud, and bridging Windows Azure for customers already committed to Windows infrastructure internally. "We think customers will not standardise on one cloud," Kelly said. "We think it will be critical to deploy applications on premises and federate across two clouds." To this end, Microsoft is building cloud federation features into Virtual Machine Manager to enable management from a single console of provisioned environments spread across the public and private Cloud. Together with its Geneva identity management platform for the cloud, Forefront tools integrated with System Center for federating public and private clouds, Microsoft is pushing to deliver on this promise with the simultaneous rollout of Windows 7 and Windows Server 2008 R2 – aligned with Windows Azure - in Q3 2009. .