ITWeb: Thinking smarter in the cloud

Alex Kayle posted this write-up post the recent ITWeb Cloud Computing Conference

Organisations are spending up to 70% of their IT budget on simply keeping the organisation running, and this has serious financial implications for the business.

Clifford Foster, CTO at IBM sub-Saharan Africa, explored the synergies between cloud computing, software-as-a-service, virtualisation and service-oriented architecture, during yesterday's ITWeb Cloud Computing conference in Bryanston.

Foster called for businesses to think smarter about managing their technology differently, in order to keep up with the exponential growth of data. “The world is changing and is becoming increasingly instrumented.

“We have about 33 billion RFID tags sold globally. We are gathering information in real-time and everything is becoming interconnected. We have four billion mobile subscribers, two billion Internet users, and we have a trillion interconnected devices.”

Solving problems
Foster pointed out that cloud computing is capable of helping organisations solve many of their business problems: “We have information generated and managed on such a scale so we can prevent problems before they happen, prevent power outages, and water contamination.”

Foster said IDC research reveals that 85% of computing capacity stands idle in computer centres, which results in a waste of energy resources. He added that cloud computing brings cost benefits to IT, as the hosted environment is provided on a pay-per-use basis as a services contract.

“Cloud computing is not new. However, cloud computing involves a paradigm-shift that catalyses many existing technologies and approaches.

“Cloud computing is an infrastructure management and services delivery methodology. We are truly at the point

ITWeb: Cloud computing evolves

A recent write-up done by Alex as an introduction to my presentation "How I explained cloud computing to my wife" next week.


The solution harnesses the synergies of SaaS, SOA and virtualisation.
Cloud computing is enabling technologies such as SaaS, SOA and virtualisation, says Clifford Foster, IBM CTO for sub-Saharan Africa.

Cloud computing is a paradigm rather than a new technology; however, it is driving SaaS, SOA and virtualisation in new ways, says Clifford Foster, IBM's CTO for sub-Saharan Africa.

During the upcoming ITWeb Cloud Computing 2009 Conference, on 24 November at The Campus, Bryanston, Foster will provide an overview of the synergies between cloud computing and software as a service (SaaS), service-oriented architecture (SOA) and virtualisation.

“Cloud computing extends the SaaS model to deliver other services on multiple payment options from the more typical SaaS model to a pay-as-you-use payment model. SOA fits naturally within the cloud paradigm, although it should be noted that it isn't a prerequisite.”

He adds that through virtualisation, applications are separated from physical resources, which provides benefits such as improved resource utilisation and system stability.

Evolving the cloud
According to Foster, cloud computing is evolving from proprietary and monolithic clouds to open standards-based and large-scale clouds.
Click here

“These open clouds will provide a range of standards-based services that will provide individuals and organisations with a range of options, offering differing service levels at respective price points. Cloud computing is evolving to include private clouds (within the organisation's firewall) and hybrid clouds (combination of private and public).”

Foster points out that some of the benefits of cloud computing is improved resource utilisation, reduced costs, ease-of-use and flexibility as users have access to the services they want without needing to own them or know where they are run.

In addition, complexity is reduced as the applications are bundled with the other software components required to execute them.

Foster adds: “However, it must be noted that cloud computing must be carefully considered based on an organisation's specific requirements, as not all services are suitable for delivering in the cloud.”

Original article here

Tying 3 concepts together (cloud, SOA & virtualisation)

Tying these three concepts together is easy, and hopefully clear, when depicted graphically.



Obviously this isn't the only option available, and it is highly simplified, but it works well at a conceptual level.

Cloud Computing vs Virtualisation

Another fairly common question - and worth answering as it is quite topical.



Both have a common objective of abstracting what you want to do from where it is done. There are some fairly clear differences though.

• Cloud Computing is a deployment architecture as I previously mentioned. From a consumer's perspective it means that the user is not aware (philosophically speaking) of where the service is executed (i.e. could be in any data centre, on any server, anywhere in the world as long as the SLA is met).
  

• Virtualisation by contrast is a term describes the decoupling of software from hardware (at its simplest level). The term can be employed more widely to refer to any level of abstraction, including data centres. Decoupling software enables a business to make deploy-time and run-time decisions about where to deploy a software stack. Benefits include resource maximisation, eg server consolidation, and resilience by moving a software stack between machines as and when required.
  

These terms and concepts will intersect as a service deployed within a cloud could be part of an application software stack that is virtualised on a hardware/resource stack.

Ideally the cloud platform itself is virtualised, in order to scale quickly, which many argue is a key characteristic of a cloud platform.

Cloud Computing vs SOA

I am often asked what is the difference between these architectures or whether Cloud Computing is replacing SOA (Service Oriented Architecture).

Let's remind ourselves:

• Cloud Computing is a deployment architecture that abstracts what we want to do from where it gets done. In addition a cloud is a pool of virtualised resources that ideally offers exceptional scalability.
  
• Service Oriented Architecture is an architectural style at the application architecture or even enterprise architecture level. SOA is fundamentally about building services that when composed deliver a rich application or business process.

These two worlds can and will intersect ... when a service is deployed in a cloud it becomes a cloud service.

Cloud Computing

The exact definition of cloud computing is being hotly debated.

However, I include the following key concepts:
(1) A Cloud - no surprise there
(2) Cloud Services
(3) Cloud Applications
(4) Cloud Computing Platform



A cloud is a pool of virtualized computer resources (servers, storage etc). The Internet is the most obvious example and some say that this is the only cloud. I prefer to think of this as a public cloud.

A cloud service is the interface to cloud application's functionality delivered using open standards. The cloud application has therefore been extended to be accessible through the cloud.

A cloud computing platform dynamically provisions, configures, reconfigures, and de-provisions servers as needed. Servers in the cloud can be physical machines or virtual machines.

Enterprise Cloud Computing
We differentiate this from cloud computing to denote the difference between the public cloud (aka Internet) and how cloud computing can be leveraged by enterprise clients within their own IT infrastructure.

The basic value proposition for the enterprise client is that they maintain control over business critical data, applications and processes while providing cloud-like user experience to their employees, partners and customers and while leveraging their existing legacy IT investments.

Cloud Computing vs Grid Computing
Cloud computing environments support grid computing by quickly providing physical and virtual servers on which the grid applications can run.

Cloud computing should not be confused with grid computing. Grid computing involves dividing a large task into many smaller tasks that run in parallel on separate servers. Grids require many computers, typically in the thousands, and commonly use servers, desktops, and laptops.

Characteristics and Benefits

• Capital expenditure is minimised as infrastructure is not owned by the user; rather it is shared by multiple users.
  

• Services are paid for typically as a subscription, usage-basis or other mechanisms like advertising

• Potential to reduce the total cost of ownership but this assumes that users understand their current financial position.

• Skill requirements are reduced as the service is provided as a fully-managed offering that is service level driven according to required performance, reliability and scalability.
  

• Environmentally-friendly as shared resources amongst users allows for improved utilization and efficiency of the infrastructure.

• Device and location independence enabling users to connect wherever and using their device of choice.

• Reduced time-to-market as businesses will be able to adopt new services quickly for pilot usages and then scale quickly to a global scale.

Risks and concerns

• Shortage of skills could limit future growth. Students need to understand principles even as they experiment with the technologies of today. That way they’ll easily adjust to the technologies of tomorrow.
  

• We still need to define and develop certain standards and technologies to support on-demand provisioning and configuration of integrated “virtual systems”. This will require protocols that allow users and service providers to discover and hand off demands to other providers, to monitor and manage their reservations, and arrange payment.

• Data and security concerns and a general fear in relinquishing their data and information to the cloud.