There are other ways to define cloud computing, as well. It can be defined as computing with virtual servers that can be made available to any resource outside the firewall. Outsourcing could be considered another abstraction of cloud computing: the work is done external to the organization, and we’re not always aware of the exact location where the work is performed.
Many people wonder, “Why a cloud?” In telecommunications, the image of a cloud is often used to illustrate the interconnection of voice-based phone calls. The same imagery is now used to represent the interconnection between thousands of computers.
Virtualized and Dynamic
As the use of the Internet increases, which is happening by leaps and bounds, more and more computing and storage resources are needed. These resources can be physically anywhere on Internet, which is the reason cloud computing is called virtualized. You don’t necessarily have to be able to touch the physical units – they could be doing the work somewhere in Timbuktu.
Because any computer or software can participate in the cloud, common business applications are available online and can be accessed from another Web service or software, such as a Web browser. Again, the software and data may be stored on servers, but we don’t have to know their physical location.
These virtualized resources can be added to the network to keep response time acceptable. This expandable network capability makes cloud computing dynamic. Since IT is often short on resources, cloud computing offers a practical solution by extending IT’s existing capabilities.
And since hardware is a major expenditure, as is buying major software licenses, the ability to rent resources or pay-as-you-go with a subscription makes cloud computing more economical than traditional models.
Four Disciplines
Another way to understand cloud computing is to see it as an amalgam of the following four disciplines:
- Grid computing: This refers to networked computers acting in concert in order to complete large, complex tasks.
- Client/server computing: Client/server computing refers to any distributed application that distinguishes between service providers (servers) and service requesters (clients).
- Peer-to-peer computing: This is a form of distributed architecture without central coordination. That means all servers have the same “say.” One server could play the role of a client requesting a job to be done by another server. Once that is done, the roles could be reversed. The server that performed the job becomes a client and requests the previous client to perform a job as a server. Peer-to-peer computing changes the roles of suppliers and consumers as needed by the task at hand.
- Utility computing: This is a metered computation and storage service, such as electricity. There is a base charge, and after that, the more you use, the more you pay.
Economics of the Cloud
Cloud computing users can reduce capital expenditures on hardware, software and services because they only need to purchase some basic equipment from which to work with the cloud. Consumption is usually billed as a utility service (similar to your electricity provider): the more you use, the more you pay. Alternately, if you pay for a subscription (like you do for cable TV), you are charged a fixed amount per month regardless of how often or how much you use it.
If you are a frequent user of cloud computing, a subscription payment method would be ideal. Otherwise, a pay-per-usage plan is recommended.
As it stands now, IT must use cloud-based computing services separately for each function – such as server computing or software as a service. And, as we all know, when separate pieces are used, it is time-consuming to make them work together, because IT must hire or build resources to accomplish integration. This integration of various pieces is the expensive part of cloud computing. Over time, more and more vendors will develop integrated solutions.
Key Features
The key features of cloud computing from the sample architecture in Figure 2 include the following:
- Expansion capability: Companies can expand their computing resources without having to make capital expenditures all at one time. And if you use less resources than what you’d have in a data center, cloud computing saves you from having “dead” equipment sitting in your data center that is bound to become obsolete over time.
- Mobility: Since cloud is Internet based, users can access applications anywhere in the world. This proves particularly beneficial to international organizations and the mobile workforces at local companies.
- Reliability: The service providers have become increasingly reliable, particularly as they respond to the following issues -
- In-memory analytics: Compared with hard drives, distributed data grids perform much more efficiently. This is because memory is shared between different data centers.
- Security: As data is distributed over a wider area and/or number of devices, the issue of securing stored sensitive data becomes an increasingly complex concern. Physical security, with literally a physical firewall, was one of the most secure environments for sensitive data. But as data distribution becomes more widely implemented, data security has become more and more vulnerable. And with global implementations of the Internet, data can easily fall prey to hackers.
