Cloud Computing (I)

By: Osvaldo Callegari*

The wonderful thing about technology is that man can change his scenario according to his requirements in the global scope, these changes are accompanied by the pulse of the hardware industry, since there is an agreement that is not always visible between computer manufacturers and software developers.

Cloud Computing, how were your beginnings?

"Cloud computing (CC) is a means of delivering IT resources as services."

All IT resources can be provided as a cloud service: applications, computing power, storage capacity, networking, collaboration tools. Its first appearances were through Internet service providers such as Google, Amazon; in the case of others they built their own infrastructure.

A new architecture emerges: a system of horizontally distributed resources, they are introduced as IT virtual services, they are massively scaled, they are managed as continuous configuration resources and they are pooled with each other.

This model received its immortalization in an article published by Wired magazine in October 2006, made by George Gilder, its title was indicative: "The factories of information".

The server farms that Gilder commented were similar to Grid computing, the difference with this model is that grids are used for technical calculations that have several subsystems with autonomy and continuous interrelation with other components.

This new cloud model applies to Internet services. "In this architecture, data resides mostly on servers 'somewhere on the Internet' and the application runs in the meantime, on the 'cloud servers' and in the user's browser," reads one section of the article.

Both clouds and grids are made to scale horizontally very efficiently. Both are built to withstand failures of individual elements or nodes. Both are charged "by-use". But while grids typically process jobs in batches, with a defined start and end point, cloud services can be continuous.

Clouds expand the types of resources available — file storage, databases, Web services and extend applicability to the Web and enterprise applications. At the same time, the concept of utility program computing became the focus of IT design and operations.

Nick Carr commented in his book "The Big Switch", that the infrastructure of computer services was beginning to be comparable with the development of electricity as a utility. Wouldn't it be great if you could buy computing resources, on demand, only paying what you need, when you need it?

For the end user, cloud computing roughly means not having to make investments in hardware acquisition, incorporate software licenses or upgrades, new consultancies, rental facilities, capital costs of any kind or hidden costs. There is only a single metered usage fee or a fixed subscription fee. You use only what you want and pay only for the time of use. Significantly reduces IT onboarding costs.

Cloud computing takes the utility computing model at a higher level. It is a new evolved form of utility computation in which different resources such as hardware, software, storage, communications etc., can be combined and / or modified on the fly within the specific capabilities or service requirements that the client needs.

From processor cycles for large projects to storage capacity, for enterprise-grade backups, when it comes to software development, CC can deliver virtually any IT requirement in real time.

How can an individual or a company take advantage of the CC trend?

CC is not simply the act of uploading images of the machine within the public cloud. There are different ways to exploit this infrastructure and explore the ecosystem of new business models.

Using the Cloud
The number and quality of offerings in public, commercial and private cloud-based services is growing fast.

For research projects, web 2.0 product developers, or simply those who want to use the load-ready and low-cost way, using the cloud is often the best option for start-ups.

If you own a new company in IT investors will ask you to lower costs to a minimum, for this function the cloud is ideal.

There are currently companies that are using public clouds for specific functions or with workload. In some cases it is attractive for development and testing. It is also interesting for a project that is in its concept stage, before buying servers you can test the system with low investment.

Functional data transfers (offloading).

For example, the company SmugMug does its thumbnailing of images as batch processes in the cloud.

"We don't really want to operate data centers anymore. We prefer to invest our time giving our customers great service and writing software that manages physical hardware ." —Don MacAskill, CEO, SmugMug

Features and categories:

* Increase:
Clouds give you a new option to handle peak loads or anticipate spikes in demand for services. This is very attractive to companies but potentially one of the most difficult use cases.
- Success depends on the attributes of the application and its relationship to other datasets that need to be repeated, and its load balancing across both sites.

* Experimentation:
- In the near future the evaluation of the software can be executed in the cloud before the licenses and or supports must be purchased.

* Building the cloud:
- Large companies understand the economic benefit of cloud computing, but want to ensure strict compliance with security policies.
- You are experimenting with private clouds, with the option to migrate applications to a cloud that is capable of delivering the right levels of services.
Some companies want to build private clouds to take advantage of the economics of resource sets and standardize their development and deployment process.

* Be the cloud:
- This category includes both cloud computing service providers and CC companies that offer multiple types of services.
As enterprises and vendors gain experience with the cloud architecture model in security trust and access control technologies that are available, many will decide to deploy external cloud services. The phenomenal growth rates of some of the available cloud public offerings will certainly accelerate the momentum.
- Amazon's EC2 was introduced just two years ago and officially graduated from beta to general availability in October 2008.

* Cloud service providers can:
- Provide new routes to market for startups and developers of Web 2.0 applications.
- Offer new value-added skills such as analytics.
- Become competitive through enterprise-class SLAs (service level agreements).
- Help the company's customers develop their own clouds.

If you're building large data centers today, you should probably think about whether you're going to offer cloud services.

Public clouds: they are managed by third parties and the jobs of different clients can be mixed on servers, on storage systems and on other infrastructure within the cloud. End users don't know the work of who else might be running on the same server, network, or disks as their own.

Private clouds: They are a good choice for companies dealing with data protection and service-level editing. They are in turn in an on-demand infrastructure managed by a single customer who controls which applications to run and where.

They own the server, network, and disk and can decide which users are allowed to use the infrastructure.

But even those who feel compelled to build a private cloud in the short term will want to handle applications on both a proprietary private infrastructure and the public cloud space. This gives rise to the concept of a hybrid cloud.

Hybrid clouds: Combine public and private cloud models.

You own some parts and share other parts, albeit in a controlled manner. Hybrid clouds offer the promise of externally provisioned, on-demand scale, but add the complexity of determining how to distribute applications across these different environments.

While enterprises may be attracted to the promise of a hybrid cloud, this option, at least initially, will likely be reserved for simple, unconditioned applications that don't require any complex synchronization or database.

Tim O'Reilly, CEO, O'Reilly Media defines us for Cloud Computing .

"It's one of the creations of the next generation in computing... It's a world where the network is the platform for all computing, where everything we think of as a computer today is just a device that connects to the big computer we're building. Cloud computing is a great way to think about how we will deliver these computing services in the future:"

In short, as we said at the beginning, man always changes the concept of the structure of the data, it is as if he were never satisfied with the model presented, we can say that the client-server concepts and local applications versus the remote ones will always be inside his mind, always looking for the possibility of having greater benefits and services lowering the costs of ownership. This lies in the countless possibilities we have within this universe of data called the Internet.

Cornerstone Technology
While basic cloud computing technologies such as horizontally scaled and distributed computing nodes have been available for some time, virtualization as an abstraction of computer resources is the cornerstone technology for all cloud architectures. With the ability to virtualize servers (behind a hypervisor-abstract operating system), storage devices, desktops, and applications , a wide distribution of IT resources can now be positioned in demand.

As it is a more extensive concept we will apply one or two more chapters in the next editions of this publication.

Note: The products and brands mentioned belong to their respective authors. Thanks to SUN for its Cloud model, Eric Schmidt in 'Information Factories' by G. Gilder and Tim O'Reilly.

* If you want to expand on this topic you can write to the author at the following email: [email protected]