Patrick T. StingleyJune 9, 2009

An Internet-based or intranet based computing environment wherein computing resources are distributed across the network (i.e.; the “cloud”) and are dynamically allocated on an individual or pooled basis as circumstances warrant, and are increased or reduced as circumstances warrant, to handle the computing task at hand” (Courtesy , Harry Newton, Newton’s Telecom Dictionary, 24th Ed.)

“Cloud computing is a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction. This cloud model promotes availability and is composed of five essential characteristics, three delivery models, and four deployment models.” (Peter Mell and Tim Grance, NIST)

Essential Characteristics include:•On-demand self-service•Ubiquitous network access•Location independent resource pooling•Rapid Elasticity•Measured Service

Delivery Models include:•Cloud Software as a Service (SaaS)•Cloud Platform as a Service (PaaS)•Cloud Infrastructure as a Service (IaaS)(Peter Mell and Tim Grance, NIST)

SaaS

PaaS

IaaS

All of the services shown on this diagram are occurring in government data centers today. (Hence the title)

Cloud architecture uses commodity hardware and virtualization to commoditize these services.

This diagram shows the SRM mappings of the cloud’s associated services.

The black boxes are the holes where the SRM doesn’t match an existing service.

(er, GeoSpatial?)(Not a cloud

problem)

Cloud Architecture is just that, an architecture.

It’s a way of building data centers using commodity hardware and virtualization.

It’s everywhere and nowhere

Everything in this diagram could be done on a single PC, or it could be built using a cloud infrastructure.

This is actually a fairly comprehensive model of how data center functions fit together expressed as services .

A Service Oriented Architecture

These are a collection of IT Infrastructure Services in the tradition of ITIL

One suite of services is GeoSpatial

Data Centers built with Cloud Computing Architectures have the following five characteristics:

Architectural Tenets:

1.Open Architecture 2.Platform Independent3.Scalable and Elastic4.Shared Service Environment5.Built with commodity products

The vision & the dream.

Cloud Computing gives us the potential to devote our lives to working together in the GeoSpatial realm instead of spending the rest of our lives fighting over standards, only to have the fruits of our labors lost to time.

The threat is not between the Cadastre and the data geeks, it’s between GeoSpatial and Temporal.

The worst case scenario is not that a vendor charges us every 18 months for a new version of their product. It’s not the cost of Patch Tuesday. It’s that 5 years from now we won’t be able to go back and refer to our data.

Vendor, hardware and platform dependence, poses a threat to our long term ability to perform our work.

In the Bureau of Land Management, it is common to refer back to land records written hundreds of years ago.

How much longer will we be able to find software to read WordPerfect 5.1? Will we even be able to find a 16 bit CPU?

All the Certification and Accreditation in the world won’t save us from the worst case scenario.

Security is the mitigation of risk.

Risk is based on the value of an asset, the cost of it’s loss and the likelihood of such an occurrence.

There is a security term for the scenario where we can’t read our own data. It’s called a denial of service and it’s every bit as effective as any virus, and potentially more permanent. We spend so much of our time and money preventing people from getting access to essentially public information, we’re failing to see the real threat that we won’t be able to get into our data ourselves.

What’s the cost?

What’s the likelihood?

We are at a prescient time.

The technologies are now mature enough to use.

Cloud Architecture is available

Platform independent languages are available

Open data standards are becoming available

We can migrate the vast majority of our data safely.

We must move now, because every day we delay will translate into data that will be lost forever.

To reiterate: Most agency applications are not designed to be able to be hosted in a cloud environment.

They are vendor, hardware and platform dependent, which poses a threat to our long term ability to perform our work.

In many cases we are on the path to platform independence. This will continue. Re-tooling costs money, so short term savings is unlikely. The long term gain will be the ability to perform our work indefinitely.

There are several programming languages that can be used to develop hardware and platform independent applications. They include:

•PERL•PHP•Python•Java•The .net languages

•All agencies will be re-tooling to migrate to this new architecture.

•Most of the functionality and code in use by an agency will fulfill similar functions at other agencies.

•The key to success is: collaboration

•Collaboration can cut the costs and development time required to migrate to the cloud. At the same time it can improve the quality of the code through cooperative re-use.

Cloud computing offers us a potential to work cooperatively towards a common goal

We can either do this, or we can spend the rest of our careers quibbling over standards and proprietary solutions and we will have

squandered this opportunity to devise a permanent solution.

Another word for collaboration is:

Sharing

So, when you go back to the office and they ask you what’s this cloud

computing is all about

You can answer them in one word

1. Identify best practices that resulted from this work that are relevant to the federal/intergovernmental environment?

1. This work resulted in the identification of the services we may expect from a cloud architecture and the holes in the FEA/CRM that will need to be addressed.

2. A shared user environment will result in encrypted files, ergo better security.

2. What is the level of maturity and viability of the referenced SOA/Cloud solutions or infrastructure within a governmental computing environment?

The functions identified as being associated with the cloud are all being done within the Federal Government today.

I am not sure they are perceived as services or a Service Oriented Architecture.

2. What are perceived impediments to adoption of your highlighted SOA/cloud practices in the government environment?

I think we need to determine the goal for cloud architecture. Is it merely to outsource the government? That can be done without a cloud.

Is it cost savings? Retooling will eat that up.

Coincidentally to this presentation, I believe that the GeoSpatial community stands to gain more than any other group because it will lead to common data standards, which we would not arrive at otherwise.

This slide deck and other materials on Cloud Computing are available on the cloud at:

http://sites.google.com/site/cloudarchitecture/

The cloud community appears to be mainly found on Twitter and Facebook. Feel free to

find me there.

Patrick T. Stingley

The positions in this briefing solely represent those of Patrick Stingley and are not meant to infer any Federal policy statements. This briefing may be shared

at will with attribution in accordance with the Creative Commons.