Application Platform Strategies Blog

Cloud Computing and the S-Word (SOA)

Blogger: Anne Thomas Manes
Follow me on Twitter: @atmanes

David Linthicum wrote yesterday in ebizq that cloud computing proponents seem to be avoiding the S-Word. Although most people recognize that cloud computing needs to leverage SOA, no one wants to talk about it. As Dave says, "The trouble with architecture is that it's almost as popular as the kid in the classroom that reminds the teacher to assign homework."

I guess it's time for me to be unpopular.

Dave's assessment is spot on: Using cloud computing without a sound architectural context is dangerous. Organizations that blithely run off and deploy applications in the cloud without considering SOA principles will find themselves in worse shape than before.

SOA is not a popular topic these days. (Remember? SOA is Dead. Moldering, rotting corpses are never particularly popular.) But just because it's dead, that doesn't mean we don't need it. And SOA is particularly important to cloud computing.

If you want an application to exhibit cloudy characteristics, the traditional n-tier application architecture that we're all so comfortable with is just not going to cut it. If the application isn't designed to scale, it won't scale any better just because you've deployed it in the cloud. It all comes down to architecture:

• How should you factor the application?
• How do you use resources more efficiently?
• How do you ensure a shared-nothing architecture? 
• How do you support multi-tenancy? 
• How do you enable dynamic lifecycle management?
• How do you support consumption-based pricing?
• How do you enable your cloudy application to interoperate with your terrestrial applications?

SOA's core design principles (separation of concerns, loose coupling, and encapsulation) are fundamental to cloudy design. We also need to develop new patterns to support cloudy characteristics. And this is a primary focus of our research in 2010.

If you're interested in learning more about application architecture and cloud computing, come join your peers at Catalyst this year. We're meeting in Prague on 19-22 April, and in San Diego on July 26-30.

Application Platform Strategies on tour: SOA and Cloud Symposium: 22-23 October, Rotterdam

Blogger: Richard Watson

Join Anne Thomas Manes and myself at SOA Symposium, co-located with Cloud Symposium, 22-23 October, in Rotterdam, The Netherlands.

On Thursday 22nd Oct, Anne is delivering the second Opening Keynote - "The Resurrection of SOA" at 9am.  At 11.15, I'm presenting "Service Modeling: Making sure your services deliver value". At 13.15 after lunch, you will have to choose between Anne’s "Proving the Business Value of SOA Investments" and my "Platform as a Service: Application Platform Metamorphosis".  Then I'll be rushing along to the SOA Patterns committee meeting for "Candidate SOA Design Patterns Review & Promotion" Session #1.  To wrap up the day at 17.30, Anne will feature on an Expert Panel, "SOA is Dead, Long Live Next Generation SOA".

On Friday 23rd Oct, I'll be presenting "Cloud Application Architecture: Rebuilding applications for the cloud" at 14.15, then doing more SOA pattern shepherding in the SOA Patterns committee session.  To wrap up the second day and the conference at 17.30, Anne and I will both feature on Expert Panels, "Today's SOA Governance Platforms: Are They Mature Enough?" and "Will the Open Cloud Become a Reality?"

(The session names are hyperlinks to the session descriptions)

So as you can see we'll be kept too busy to check out the post-war architecture which makes Rotterdam such as fascinating place.  Imagine for a moment rebuilding an entire city in the 3 decades of the 1950s-1970s.  What would you come up with?

Pity we can't start building a modular cloud application architecture on such a green field site. Hybrid on-premises/off-premises applications architecture will forever be a reality for enterprises. Some organizations are unwilling or legally not permitted to host every application and its data in public or external clouds. The typical chaotic and brittle applications architecture found in most organizations means that a significant fraction of existing applications are not ready to move to the cloud.

We have a very limited number of free guest passes for the conferences still available, each valued at 1,090 Euro.  If you’d like to attend the conference, leave a comment on the blog, and I'll get back to you.  Once the passes are gone, we can still offer a 25% "friend of the speaker" discount. 

Cloud Computing is "depressingly" similar to SOA

Blogger: Anne Thomas Manes

David Linthicum wins the "grumpy architect" award for today.

He posted a rant on Sys-Con Media entitled, "Getting the Links Straight Between Cloud Computing & SOA". It's well-worth the read. In it he references an ancient (Feb 17, 2009) article by Stacey Higginbotham entitled, "HP Confines the Cloud for Enterprises," in which she wrote up HP's strategy announcement on Cloud Computing.

In particular, Dave was ranting about this comment:

But aside from some vague nods to the benefits of accessing information in the clouds (such as with web-based email) most of HP’s detailed talk of clouds in the first webinar was depressingly similar to the idea of service oriented architecture. HP offered clouds as merely a means to deliver IT as a service inside the enterprise.

Well, Duh!

Alright. I admit it. It's all my fault. I proclaimed that SOA is dead, and people who can't be bothered to actually read the blog post (or any of my follow-on blog posts), took it to mean that we shouldn't bother doing service orientation anymore.

I tried to be really explicit in my last blog post, "SOA: It's Dead, Jim," but perhaps people still didn't get it:

"SOA" as a term has lost its luster, but "SOA" as a practice is essential for all organizations going forward.

As I said in the original obituary:

Service-orientation is a prerequisite for rapid integration of data and business processes; it enables situational development models, such as mashups; and it’s the foundational architecture for SaaS and cloud computing.

The network is a bit overrun with articles talking about the intersection of SOA and Cloud Computing, so it really should be obvious by now: Cloud computing requires SOA.

Note, though, that the exigencies that require we do SOA will do nothing to revive it in the eyes of business people that control its funding. SOA is still dead: business people believe that SOA is expensive, and it doesn't deliver, and they don't want to allocate any more money to it this year. (In case you missed it, this is what I mean when I say, "SOA is dead".)

Our challenge as architects is to figure out how to do SOA without a big pile of money to fund our efforts. Fortunately, SOA doesn't need to cost a lot. It just requires good architectural practices. SOA is about design, not about technology.

Patriot Act vs. EU Data Protection Directive: regulatory death-match in the Cloud

Blogger: Richard Watson

I spent part of last week speaking to European clients about cloud and SaaS.  There's a real thirst here for a conversation about the challenges and risks of cloud and SaaS, free from hype and hysteria.  The architects and executives I spoke to are already fatigued with "SaaS and cloud being pushed non-stop by vendors and analysts" (sic). 

One universal concern about hosting data in external clouds is data privacy.  Heretofore, concerns of EU companies included the fact that storing personal data in "third countries" violated the EU's Data Protection Act.    Of far more of concern now is that local data regulations in the provider's jurisdiction (especially the US Patriot Act), could be prioritized over international Safe Harbor arrangements designed to broker the local and guest privacy regulations.

I think these fears are justified and the recent clarification by the European Commission in its "Frequently asked questions relating to transfers of personal data from the EU/EEA to third countries" does little to allay that substantive fear.   The EU confirmation does indeed clarify which "third countries" have adequate standards to comply with the EU Data Protection Act, thus opening the door for EU data to be stored in (you won't need two hands to count them) the US (with Safe Harbor), Switzerland, Canada, Argentina, Jersey, Guernsey and the Isle of Man.  US companies that have signed safe harbor agreements (including Amazon, Google, Microsoft, and IBM) are bound to comply with European data protection standards, but the relative prioritization with the Patriot Act has never been explicitly tested in court.  Therein lies the problem.  Could you risk your customers' data as the test case?

This interesting study by Galexia challenges the effectiveness of the compromise of Safe Harbor agreement. 

A further grey area that was raised at our cloud sessions was whether "strategic US business interests" might also be prioritized over Safe Harbor agreements.  I pointed out that an even more concrete concern should be accidental data leakage and inappropriate use through either carelessness or malpractice of providers' employees, not to mention poor isolation practices.  These issues are well-documented in Eric Maiwald's (subscriber only) "Considerations for Risk Management When Choosing Software as a Service".

The Cybersecurity Act of 2009 recently introduced in the US Senate gives European consumers more reasons to think twice before jumping into (especially private) cloud agreements with US based providers.  The measures in the Cybersecurity Act are at the same time potentially wide-reaching and vague, but include giving the US President a lot of power over the security of the Internet.

Despite the recent EU clarification, data privacy raises key questions for European cloud adopters.  How are the clouds looking over Canada ... or Switzerland?

Gearing up for Catalyst: San Diego, July 27-31

Blogger: Anne Thomas Manes

Wow. It's April already. Less than 3 months to Catalyst.

We've been sorting through the plethora of submitted proposals, and we've just about finalized the schedule for event. I'm sorry to report, though, that two of my favorite end-user case studies had to withdraw their submissions last week. Therefore I still have 2 open slots. If you have a great story to tell about SOA or business value metrics, I am once again accepting proposals for these two tracks. (Please send me an email: amanes AT burtongroup DOT com.)

All in all, I'm very excited about the schedule. We have 3 great cases studies lined up on SOA, and 2 for business value metrics. I'm also really excited about the line-up we have planned on Cloud Computing and SaaS. It's a conference you can't afford to miss. Check out the full agenda. Also check out the workshops. I'll be reprising my REST Easy workshop.

By the way, we have a super secret discount available to readers of this blog. We have place a couple of Easter Eggs on the Catalyst Web Site. You can reveal the discount code two ways:

• Hover (but don't click) over the "San Diego" icon for 20 seconds
• Click and hold on the Catalyst logo and then drag your mouse off and release  

Architecture Bailouts and Structural Change

In the economic realm, economists often refer to 'structural change'. 'Structural change' is when long-term game rules have been fundamentally altered, and truisms based on the old model no longer apply (e.g. 'real estate value never decreases', 'over the long term, the stock market always rises in value', 'my children will experience more opportunity', 'owning is better than renting'). If I propose economic and cultural aspects underlying the world around us are radically changing and old models do not provide good guidance, you may not be surprised. Because underlying economic mechanics have changed (i.e. investment yield,counter party risk, credit liquidity, personal consumption habits) economic bailouts based on existing models are not as effective (much to everyone's chagrin). We are forced to re-build the economy and meet new long-term realities imposed by structural changes that occurred during the last twenty years (i.e. financial industry market participation, credit expansion, credit risk management and hedging, asset valuation). To be effective, participants must recognize a new model is in force.

Shouldn't a structural change within Information Technology be recognized today? Cloud computing proponents think so... Disruptive forces (i.e. commoditization, tightening economics, industrialization, business practice acceptance, and a resurgent desire for trust and credibility) are creating a new reality. Organizations can abide by old model rules or gain advantage by adapting.

The normal IT response is to adopt an 'architecture bailout' focusing on products and technology. But many 'architecture bailouts' (i.e. Object oriented programming, Component Based Design, Event Driven Architecture, and Service Oriented Architecture) are pursued as incremental, bolt-on practices. Rarely do organizations transform people, process, and technology to match realities imposed by structural change. But 'What is IT Transformation Really?', Brian Watson of 'CIO insight' furthers the discussion with this call to action: "[Position] your IT shop as the “internal consultant of choice,” versus available consulting or advisory services". How many of us know 'the cost of compute hour', 'the cost of storage per GB', 'the cost of web service development', 'the cost of our application platform'? To determine which aspects of IT to shift into Cloud, we need to fully understand the economics of IT instead of architecture and technology.

Many Cloud computing messages target IT business model adaptation (i.e. pay-as-you-go, externalizing IT services, service consumption) instead of another architecture bailout (i.e. grid, virtual machines, autonomic computing, SOA). The conversation is shifting in the right direction, and Chris Howard is leading with research proposing a process to determine when to engage internal resources (core) or disengage the task and rely on external service providers in the Cloud.

Today's economic players are proposing a transformational response and stepping out of their comfort zone to lead. Meeting structural changes imposed by Cloud computing will require significant action by IT leadership. A recent blog post by Mike Rollings states an incremental responses will often be ineffective and leadership is a critical. Jack Santos has an excellent blog entry on leadership and collaboration (the embedded 2 minute video is uplifting).

Service-orientation and Cloud Events - Feb/March 2009

Join Burton Group’s Application Platform team at live in-person events this month and next. We look forward to collaborating with you during educational opportunities focused on Service-orientation and Cloud.Upcoming events include:

• IASA IT Architect Regional Conference 2009, February 25-26, 2009
• SD West 2009, March 11-12, 2009
• Catalyst North America 2009, July 27-31, 1009

IASA IT Architect Regional Conference 2009, February 25-26, 2009

Location: Atlanta, Georgia

Full-day SOA Assesssment, Planning and Reference Architecture Workshop

Are you ready for SOA? Where should you start? How will you specify actionable steps that will move your organization away from project silos and towards a service-oriented mindset? What projects will bring you the most benefit? What areas of your organization, architecture, infrastructure, or development practices need the most work? This workshop provides guidance and practical advice to help an organization conduct a successful SOA initiative. Every SOA initiative should start with a self-assessment to gauge the organization’s readiness for SOA and to recognize areas that need improvement, identify opportunities, and establish priorities. Once you know where you are, you can then plan a course to get to where you want to go. The workshop will describe the following tools that can be used to define and guide your SOA initiative:

• SOA maturity model framework
• Assessment surveys
• Recommendation templates
• Sample initiative roadmaps

Every SOA initiative should start with a readiness assessment and continuously track progress towards goals. This workshop will give you the tools you need to do so:

• What aspects of the organization should be evaluated?
• How do you measure readiness?
• How do you establish priorities?
• How do you increase organizational maturity?
• How do you measure maturity and demonstrate business value?

You’ve been tasked with designing a SOA infrastructure. Where do you start? What infrastructure technology components must be procured? How will you host services? How will you control access to them? How will you manage them and ensure that service-level agreements are met? How will you ensure that services are properly secured and instrumented? The afternoon workshop session will examine the requirements of a SOA infrastructure from a functional perspective and will discuss the various alternatives available to address those functional requirements. It will provide candid feature/benefit analysis of the various types of products, and discuss methods for upgrading your existing middleware environment.

• Is an ESB a prerequisite for SOA?
• What value does a BPEL engine provide?
• How do you support both SOAP and REST?
• What mechanisms are available to track SLAs and pinpoint failures in the system?
• When is it appropriate to use an XML gateway?
• Is a registry really necessary?
• What aspects of governance aren’t optional?

Presentation: Cloud Now: Infrastructure, Architecture, Service Providers, and Cloud Composition Patterns

The Cloud focuses on re-creating IT into a more efficient and productive delivery organization by implementing cloud principles, infrastructure technology, architecture and composition patterns, and business models. In this presentation, Chris Haddad, Vice President of Burton Group’s Application Platform team, will illustrate principles shaping the Cloud and how organizations adopting Cloud capabilities and composition patterns must navigate through competing tailwinds and headwinds to achieve business benefit and value. The session will answer the following questions:

• What principles shape the Cloud?
• Is the Cloud a business model, a set of technologies, or an IT delivery mechanism?
• What benefits, risks, and value may organizations derive from the Cloud?
• What is the Cloud’s impact on application architecture and infrastructure?
• Who are the players delivering cloud infrastructure and services?
• How should organizations adopt the Cloud?
• When is Cloud the quickest path to deployment and business value?

The Cloud focuses on re-creating IT into a more efficient and productive delivery organization by implementing cloud principles, emerging infrastructure technology, architecture and composition patterns, and business models. In this presentation, Chris Haddad will illustrate principles shaping the Cloud and how organizations adopting Cloud capabilities and composition patterns must navigate through competing tailwinds and headwinds to achieve business benefit and value. The session explores the Cloud’s impact on business models, application architecture, and IT infrastructure.

SD West 2009, March 11-12, 2009

Location: Santa Clara, California

Burton Group will lead the following sessions:

• Cloud Now: Infrastructure, Architecture, Service Providers, and Cloud Composition Patterns
• Cloud Application Architecture: Re-Building Applications for the Cloud
• Panel Moderator: “The New App Server Frontier”

Presentation: Cloud Application Architecture: Re-Building Applications for the Cloud

The Cloud promises to bring infinite scalability, unlimited availability, and increased responsiveness. Can applications realize cloud benefits through a simple off-premise server migration? Does Cloud require developers to re-write applications or port applications to proprietary Platform as a Service (PaaS) environments? In this presentation, Chris Haddad, Vice President of Burton Group’s Application Platform team, will detail cloud application architecture patterns, cloud application frameworks, portability and migration strategies, and deployment topology considerations.

The session will answer the following questions:

• How does Cloud Application Architecture compare to web application, client-server, and desktop architectures?
• When is Cloud Application Architecture the appropriate choice?
• What application frameworks and development environments should teams use to build Cloud applications?
• What architecture roadmap should be chosen to make applications Cloud ready?

Catalyst North America 2009, July 27-31, 2009

Location: San Diego, California

Tracks relevant to Application architects and developers include ‘SOA is Dead: Long Live Services‘  and ‘Cloud Computing’s Business Advantage’.

Application Platform-centric Catalyst Workshops include ‘REST Easy’, ‘Improving the Software Development Process’, and ‘Building Applications for Deployment in the Cloud’.

This XaaS stuff is great, but, wait, what's the catch?

Blogger: JP Morgenthal

It's really difficult to be objective about XaaS when it seems that no matter what technical argument you can make against it is ultimately trumped by cost.  Yet, look at the success of companies like NetSuite and Salesforce.com; how can you argue against the model with so many customers reaping value?

It's probably unfair to lump all "as-a-Service" offerings together in this way given that each has pros and cons that are distinct.  But, I've discussed the concept of business having a Neanderthal brain, and XaaS is a great example of that brain at work.  Great technical positions regarding trust, data ownership, availability issues are responded to with the equivalent of "I'm rubber, you're glue, anything you say bounces off of me and sticks to you!"  Even worse is the vendors playing the parental role in this childish repertoire scolding the technologists for even raising these issues; admonishing them for restraining progress and being blind to the eventuality of life in the cloud. The hard-sell, if ever I've heard one.

The good news is that none of the technical arguments cannot be overcome.  Indeed, most of them can be handled through service-level agreements, legal contracts and the rise of domain-specific XaaS environments.  Still, I believe, the ardent pushback to raising technical issues stems from the fact that XaaS providers have yet to solidify the business models underlying XaaS and fear rejections at this point that may shut off avenues of funding before they even get a chance to figure out how to deliver value and protect the consumer all at a reasonable cost factor that still delivers profitability.

In this CNet piece, "The cost of cloud adoption", the author does a great job of explaining the misconceptions regarding the underlying accounting principles of XaaS, without losing the value for business or the impact of running a large data center.  But, this blog posting applies primarily to IaaS (Infrastructure) more than PaaS (Platform) or SaaS (Software).  However, buying compute time is not a new concept.  It used to be the only way businesses could afford to own a computer.  Now, it's a realization that the cost of running a data center for even slightly above average utilization is not cost-effective.  Moreover, the ability to scale on demand is very useful in shops that see spikes related to specific business-oriented milestones, e.g. end of month reporting.

So, where's the catch?  Personally, I practically live out in the cloud now.  I can set up shop at pretty much any computer and get to my email, feeds (e.g. RSS, Twitter, etc.), social networks, instant messaging and business documents.  With the exception of my business documents, which are hosted by my company, I realize that I don't pay anything for all the other services; they are supported primarily by advertising revenue or venture funding.  In a recent dialogue with another analyst we both admitted, if these companies needed to charge to survive, would we pay for these services?  Both of us admitted we would not.  I suppose that means that I would revert to using desktop client applications and Microsoft Outlook as the primary means of communication.  Sadly, if this did occur, I believe many recently rekindled friendships on Facebook would once again dissipate; collateral damage of a bad economy.

While there is a major difference between my personal life in the clouds and business in the cloud, I do believe the one attribute they share is the response to failure by XaaS providers ability to thrive.  As I would be returned to using traditional desktop-oriented applications that I pay for once versus pay-by-use or pay-by-subscription, so would organizations that have made bets on the cloud be returned to alternative means of delivering IT functions.  This does not have to be devastating, but it certainly can be disruptive and, for some, terminal.  While fiscal responsibility is important to the lifeblood of business, betting on XaaS to save money now may mean paying more money in the future as you need to be rescued from a failing platform.

As my Uncle Louie liked to say, "There's no such thing as a free lunch ... unless you pull it out of the garbage can." ;-)

What I've Learned From Anne's Blog Entry

Blogger: JP Morgenthal

Anne's blog entry SOA is Dead; Long Live Services has drummed up many opinions and emotions from both technical and non-technical camps.  The intellectual gyrations are phenomenal, but at the end of the day, if it only served to break us out of our doldrums, create drama and generate traffic for Twitter, then we have a net sum gain of zero.

Prior to joining Burton Group I was heading up a software company that developed Web Services for Supply-Chain Management.  We used all the key buzzwords in our marketing literature, SOA, Supply-Chain-as-a-Service, etc.  What we developed definitely leveraged SOA principles.  We focused on granularity of the services and ensured we didn't introduce tight-coupling between services, which, by the way, was a royal pain in the a$$! But, to Anne's point, the desired outcome was achieved simply by focusing on the services.  When the resulting services were layered on top of monolithic, home-grown SCM applications, we created an environment where new capabilities and business processes could be added in weeks verses months.

Lesson #1:  Services is the new components.  Call them what you will.  I am not a fan of the term "service" because I believe it is too overloaded, but all the good terms were already applied in years past and are now associated with failed initiatives.  Still, the key here is tactical works and it's okay to be tactical; especially if you are not prepared to commit to the costs, resources and time to take a holistic look at your information technology assets and structure them in a way that will benefit your enterprise long-term.

Lesson #2:  The concept of word-of-mouth needs to be totally re-evaluated in the age of social networking.  I actually noticed re-tweets pointing to Anne's blog entry in multiple languages.  The speed that this news traveled across the World was incredible.  And along with this lesson a reminder that what you post on the Internet can, and most likely will, be seen by an audience you never anticipated.

Lesson #3:  There's a lot of smart people out there, too!  I saw some really well-thought-out responses to Anne's blog entry and some great comments.  There was also various perspectives on the issue from practitioners and vendors alike.  Being in the midst of the game, it's easy to forget that the poor schnook on the sidelines doesn't have a clue who to listen to.  Hence, my advice to them ... listen to me!  (that was a joke for those humoristically disabled).  In all seriousness, it illustrates the importance of trust that can only be seriously built face-to-face and cannot be replaced by all the Tweets in the world (yes, that was a pun).

Lesson #4:  A group of smart people can be ignorant.  There was a lot of responses directed at, and rightfully so, "the next shiny thing" that IT will move toward.  Will it be Cloud? SaaS? Mashups? To Anne's point, "The latest shiny new technology will not make things better."  As I like to say, business has a Neanderthal brain--it lives in the "now", cannot incorporate past experience into current situations and likes shiny things.  There's a great quote, that I could not locate, about the lack of utility of people operating in a group that is very apt here.  If you know it, please share.

Lesson #5:  It was Marshal McLuhan that said, "Societies have always been shaped more by the nature of the media by which men communicate than by the content of the communication."  I found this statement remarkably relevant in this situation.  It seems many individuals felt the need to make a statement irrelevant of the original thesis.  McLuhan's message can be applied at many levels, but most simply applied, I saw quite a few postings that incorrectly addressed either the issue or the original content.  For me, these individuals were seeking to leverage the medium regardless of the content to gain attention.

Lesson #6:  Shakespeare was a smart dude! I quote, "What's in a name? That which we call a rose by any other name would smell as sweet."  I don't know if I'd go so far as to say good architecture is sweet smelling, but the statement holds.  Whether you call it SOA or call it good architectural principles, the act of designing for longevity, agility, reuse is a good thing and will reduce total cost of ownership over time.

Lesson #6a:  If you're flipping properties, don't get caught holding any when the bubble bursts!  Many businesses treat their business like flipping properties in the real estate market.  For them, software just has to be good enough, not great.  It's a disposable mindset in which they believe that business changes very rapidly and they want to invest just enough in IT to keep the ship running.  This mentality limits perspective to months versus years, but this may be okay.  I'm not critcizing this approach as it seems in many cases to be viable.  The key, just like in real estate, is don't get caught holding in a downturn.  As the business climate slows, like we are seeing now, the cycles of change get further apart and you will be left with IT investments that cannot support continued execution without more investment in a climate where being frugal would be best.  Still, I wouldn't be trying to sell management in these businesses on the virtues of SOA governance anytime soon.  Moreover, you can bet these will be the first businesses trying to run 100% on Cloud infrastructure.  That is, in the land of the blind, a one-eyed architect is ... essentially out of work!

As for any advice I have on SOA futures, its death, its life, etc., I will defer (happily!) to Anne, Richard, Chris and Mike at Burton Group and revel in the fact that I get to cover other amorphous topics like BPM and PaaS!

Do-It-yourself SaaS, a first step into the cloud?

Blogger: Richard Watson

This article on Bechtel’s strategy for reinventing its IT as an internal SaaS provider over at NetworkWorld is thought-provoking in many respects.  Leaving Bechtel’s infrastructure aside for a moment – I’ll leave that one for Drue Reeves in Data Centre Strategies to comment on - the exercise they went through in application portfolio management alone is astonishing, and surely some of their expected savings in IT budget is attributable to their leaner application portfolio.

I recently spoke to Annrai O’Toole, VP of Integration at the ERP SaaS vendor Workday.  We discussed walking his customers through a thought experiment like the process Bechtel went through: If you had to start again, would you start with your “development platform”: Java, Application servers, databases, etc.  He makes the case for an “application platform” based on ‘SWAG’ (Salesforce.com, Workday, Amazon and Google).

In this era of divestiture & enterprise reinvention – typified by the financial industry – that’s not too farfetched.   The landscape of investment banking is almost unrecognizable from the one I left 9 months ago.  The credit crunch has accelerated the process of disaggregating the banks into global holding companies (e.g. Citi), boutique hedge funds (e.g. funds have fallen by 20% on average) and specialized back-office service providers (e.g. Thomson/Reuters, ADP, etc.)  I’ve blogged previously on my skepticism for the success of migrating enterprise applications to a SaaS model. When an organization (or an industry) is reinvented, that issue becomes moot.

If enterprises feel they can fund such extensive portfolio management activities from OPEX, my punt is that we’ll see lots of it going on.  Over to Mike Rollings in Executive Advisory for his thoughts - Mike?