elemental links

Liveblogging Cloud Computing Expo #14, Cloud Computing & EA (Elastra)

Stuart Charlton from Elastra is covering cloud computing and the next generation of enterprise architecture.  I’m in the back of a packed room, hard to see the slides, so forgive what follows.

Elastra is focused on governance, configuration management and hybrid clouds for enterprises. 

Session objectives:

• Does cloud computing change the importance and role of enterprise architecture and IT service management?
• Explore a reference model for the cloud.
• Bridge gap between architectural intent and getting stuff done. 

He speaks of enterprise architecture as dealing with logic for the business.  He claims (wrongly) that business architecture is what used to be called business analysis.  He does (rightly) say that the chief business architect of a large business is the COO.

He’s quoting Weill & Ross’ EA as strategy.  Operating model concept.  No need to explain this to my readers.

Next in the EA tour – you guessed it – Zachman.  He has a chart with rows 1 and 5 with “a miracle occurs” as the transition from 1 to 5.  I can buy that.  He also says cloud computing doesn’t change this.  Architecture doesn’t go away.  The need for architecture doesn’t go away. 

The trick, Stuart shares, is to determine if cloud computing helps to shrink the time “of the miracle occurs”.

Now, he’s talking about how we manage IT.  Talks about role-led techniques, developer-led (code management & controls), manager-led, architect-led (modeling focused), operations-led (ITSM).  Calls out that each trying to solve same problem, but from their own viewpoints, and disconnected.

Current debate on Cloud computing & ITIL – does cloud computing destroy ITIL?  One theory, “Google secret sauce theory” is the recipe for IT has changed – architecture and implementation.  Of course, everything needs to be re-written.  Does this fundamentally change the nature of IT and EA?  [need to think about the degree to which this is true]

Cloud provider continuum from retail ecosystem (force.com) to supplier ecosystem (amazon). Retail is closer to developer and user.  Supplier is closer to sysadmin & operations.  More commonly referred to as platform-as-a-service and infrastructure-as-a-service.

Qualities of an enterprise cloud:

- on-demand, services-oriented computing

- variable cost consumption

- self-service

- elastic scalability

- mandatory network

- governance and compliance

Cloud computing is not just about outsourcing.  “It’s an on-demand, services-oriented environment.  Drastic reduction of lead-time to enable change in my environment.”(I like that)

Calls out Cisco’s unified computing initiative as something to watch.

Enterprises will only work with a handful of cloud providers.  In response to question, believes there will be PCI compliant clouds.  So, characteristics of clouds will include compliance models.

Reference model layers:

- begin with data center; global, possibly cross-organizational, exposes power & cooling information

- add trust, identity & licensing; control point for compliance, auditing, distributed action w/o replicated credentials

- add configuration and resource management software; have more stuff, need tools to deal with it; resource management deals with flip side of scarce resources

- add some visibility; hyperlinked model of metadata; (what uses/contains other things), lifecycle (when and how can things change?)

- add some real world context: governance, constraints & policies, testing, monitoring & operations; gets interesting when applications (business scenarios) span clouds

Slides will be available on Slideshare.

Sees a movement towards Enterprise Cloud Servers.  Need to grab the slide to best explain that concept. 

EA to Implemention Gap – architecture aware system, policy and configuration management.  I see common themes with SOA governance and Service Management.  Because, you know, it’s all services.

This concludes my coverage.  Hope it gave you some things to think about.  I’ll be following up this “reporting” with some “analysis”.  Just not today!

Liveblogging Cloud Computing Expo #13 – Stairway to the Cloud (Freedom OSS)

Freedom OSS is a consulting group with a framework on moving to the cloud, particularly focused on Amazon (AWS) implementation.  Goal is to lower barrier of entry  to cloud adoption (for enterprises).

Framework Steps:

1. Business case – ROI, TCO

- application portfolio analysis, assessments for cloud readiness, ROI, TCO, proof of concept, educate/win over CFO, Capex vs. Opex, pre-build collateral to shorten the cycle

2. Large Data Set transfer – large as in terabytes – service offering cloudIngest

- reliable and secure data transfers; FTP will take weeks or months, need to rethink how

3. Enterprise integration in the cloud – sync, integrate with partners

- cloudMQ offering; cloud + event-driven architecture = Internet Scale SOA

4. Enterprise IDM for the cloud – bridge cloud based ACL with IDM

5. Cloud operational excellence – how to make the cloud an extension of your data center – assumes data center has operational excellence

- monitoring & management tools, security, enterprise backup system integration, enterprise data center to elastic cloud bridging

6. Cloud disaster recovery

7. Cloud high performance computing

- non-invasive HPC application migration, integration with enterprise operations management, HPC elastic computing, readily available AMI for HPC (MapReduce via Hadoop or GridGain), clusteredFS over EBS or S3, data synchronization, pre-configured management & monitoring

8. Training

Enterprise Use Cases (unnamed)

- Major Hedge Fund – P&L Stress test / risk exposure

- Prime brokerage – EOB Trade Processing – spill-over/bursting strategy

- Major Investment Bank – HPC for equities modeling ; disaster recovery

- two more, missed them, but mostly bursting strategies

Q: does methodology have step to revisit what was done after a given time period?

A: yes.

Liveblogging Cloud Computing Expo #12 David Snead, P.C. – Virtualization & Legal Concerns

David Snead on Virtualization and Legal implications.  David is a practicing attorney focusing on web infrastructure concerns.  From a legal perspective, David shares that from a legal perspective, virtualization and cloud computing are similar.

Three aspects to be considered:

Software or Operating System

Expectations – your own, that of your users, that of your customers

Contract Review

Goal of session to give information on how to parse legal risk.  Lawyers who say “no legal risk” are misstating the truth. 

What is Virtualization (CC) from lawyer’s perspective, operating system or network distributing things. (Didn’t catch that exactly).  Can look at virtualization as server based, application based or desktop based.

Application virtualization, this is the holy grail.  This is what his clients want.  Want to put applications on Salesforce.com

Server virtualization risk areas: virtualization software, individual operating systems.  For example, do individual have the ability to interfere with other users on the server?  What happens when a user brings down a shared environment.

Application virtualization brings highest risk and reward.  Keys to ask, who is providing the underlying application infrastructure?  Is the application part of an ecosystem?  An application ecosystem?  A platform (infrastructure) ecosystem? 

Concerns: will the infrastructure be up when required?  what happens if it isn’t?  next, your data.  where is it?  is it safe?  can you get it back?  finally, virtualization software.  what is the provider running?  how might that impact you from a risk perspective?

In response to question on fiduciary responsibility, contract law is going to govern 98% of scenarios.  So, understand and comb those contracts!  Need to approach all relationships as though the provider doesn’t have your best interest at heart.  Admittedly, a lawyerly response, but important to say/hear.

Desktop virtualization concern – no surprise – is you want to be in compliance with your Microsoft desktop license agreement.  Also, need to look at security provided, beyond the application layer.

Common issue is understanding virtualization software and understanding if that software will work with type of virtualization – desktop, application or server.  Is software appropriate for task at hand.  Sounds simple, but has seen issues here.  You can’t get out of contract just because application doesn’t work, especially if you didn’t check compatibility.

Different risks for Vmware view of world vs. Parallels (container) view of universe.  In container view, it is easier for one user to destabilize another user.  If that’s a concern, consider the vmware view.

Need to determine if your licenses will work in virtualization environment. 

Warns that anytime a human touches the system, there is a chance to break it.  Need to understand the degree of automation done by the service provider.  (Reminds me of my Joyent briefing earlier, automation is critical)

Understand that the data doesn’t go to “the cloud”.  It goes somewhere.  Understand cloud provider’s warranties, SLA, privacy, skin in the game.  Suggests going out to read Amazon’s terms of service.  Go to company that drinks own Kool-aid.

Also want to understand potential channel conflict.  [Example I just read about in HBR (I think) is Toys-r-Us and Amazon.]  David cites Salesforce & Zoho recent litigation.

Don’t do find/replace contract changes, such as other vendor name/your name or dedicated host/cloud host.

In response to question, avoid CA, UT and most of South in interpreting technology contract law.

End of day, who owns the data?  Especially critical where channel conflict might exist.  Don’t want to accidently give your customer list to a competitor.

360 degree contract review:

Infrastructure: bandwidth, upstream, AUP (SLA)

Virtualization: type, IP protection, end user interaction

End User: customer expectations, ultimate use, expertise

Read the contracts!  Not just data center.  All contracts, current and new.  Don’t forget SLA / expectations of your ultimate end-users.

Don’t forget choice of law and regulatory issues.

Liveblogging Cloud Computing Expo #11 – Martin Ingram (Appsense)

Martin Ingram of AppSense is presenting to a general session on Why VDI?  If you are like me, what VDI? VDI is desktop virtualization.  Session abstract:

“The technology of desktop virtualization is moving forward rapidly but what is the real goal? This session discusses the overall objective of desktop virtualization and how technologies available now or in development contribute to that objective. By looking at the future directions of the principal virtualization technologies (hardware, operating system, application and user) and where they can be used (hosted, client and cloud) it equips you to evaluate all the technologies, their relevance for your users and the benefits you should expect.”

Talking about Henry Ford and the Model T.  How standardization played a role in keeping the car price affordable, 1/3 of competitive products.  Standardization as a lack of options, no driver’s side door, and standardization in component design and what became assembly production.

Three forms of desktop virtualization: hardware, operating system and applications.  Where application virtualization is separate from operating system virtualization.

Trick is to remove user environment from OS and application.  This includes user configuration, user applications and user data.  Then, you can mix and match the layers of hardware, os and applications.

Keys to successful VDI are standardization, personalization and componentization.

Liveblogging: Cloud Computing Expo #10, David Linthicum

Dave is talking about Winning with Cloud Computing Step-by-Step.  The presentation is up on slideshare. [link added 4.1.2009]  Dave’s presentation is based on work from a forthcoming book.

The basic idea is you can extend your SOA to the cloud, utilizing external resources, either business or informational services, or infrastructure resources.

Cloud & SOA lets us mix an enterprise architecture cocktail.  [works for me]

You can’t replace enterprise architecture with cloud computing.  You can’t replace SOA with cloud computing.  You always need an architectural strategy.  Adding cloud computing allows you to cash-in on SOA. Saving money because you are using other people’s work.

If you want to boil cloud computing down, it’s switching the use of resources that are currently available in your enterprise, out in the cloud, but in a virtualized, reliable, secure manner via a subscription model.

Cloud computing is neither completely evil or completely good.  Lots of hype at this conference.  Lots of skeptism in IT.  Cloud computing has tendency to be cost-effective, but need to consider “all-in” cost.  Still need to maintain your systems (applications, databases etc), do integration, deal with security on the cloud. 

On-premise, you’ll always be under utilizing your hardware and software investments.  Many on-premise systems are only at 3% capacity.  Cloud delivered, you pay as you go.  So, your entire investment is productive – running business transactions.

Warnings:

- Not all computing resources should exist in the cloud

- Not always cost-effective – all-in cost

- Don’t move crap to the cloud (not what he said today, but I heard him say it last week)

When it’s a fit:

- processes, applications and data are largely independent

- integration points well defined

- lower level of security is ok

- core enterprise architecture is healthy

- web is desired platform

- cost is an issue

- applications are new

When it’s not a fit: essentially the opposite for 1-6, plus: application is legacy, application requires native interface

Start with the architecture: understand your business drivers, information under management, services under management and core business processes.  Use whatever EA framework works for your organization.

Getting ready:

- accept that it’s ok to leverage external services in your SOA

- create a strategy for the consumption and management of cloud services

- create a proof of concept now.  get a login.  try stuff.

Dave has his 17 point stepping into the clouds list up now.  I’ve blogged on this before.  Items include, assessing culture, business, value, data, services, processes, cloud resources.  Then, identify candidate data, services, processes.  Create governance strategy.  And some more.

Dave’s 17 points could be called, “stepping over the hype to architect your way to the cloud”.

Man!  Dave just crashed a lightening bolt through my fluffy clouds.  He shares, we’ll never, ever, have a magic console that allows us to drag and drop icons and suddenly have cloud applications and cloud deployments.  He implies, we’ll actually have to do some work.  It’s almost as though he’s been through a paradigm shift or two.

Other thoughts from Dave:

- external cloud services should function like any other enterprise resource – should work the same

- evaluate cloud providers using similar validation patterns as you would with business and data center solutions

- consider private clouds; right now, that’s really a virtualization story, recouping that unused capacity

Q&A

Discussion on Vertical clouds versus private.  A hedge fund cloud is not private, because there would be multi-parties involved.  Private cloud is for an organization and their data.

Discussion on SLAs, compensation and risk.  If you get (expect) compensation when the public cloud doesn’t meet the service levels, you’ve increased the cost, and therefore the value of the public cloud.  [This point is interesting, I’ll be exploring it further.  Certainly, there needs to be some penalty for not meeting an SLA, otherwise, why have one].