CLOUD COMPUTING

Cloud Computing

Cloud Computing

Introduction

Providing software as a service is not a new computing practice. Some companies, known as Application Service Providers (ASPs), were providing businesses with software programs as a service via the medium of the Internet during the 1990s. However, such attempts at “utility computing” did not take off. This was largely attributed to lack of sufficient bandwidth. During that period broadband was neither cheap nor plentiful enough for utilities to deliver computing services with the speed and reliability that businesses enjoyed with their local machines.

Then came Web services (especially those based on the XML-based SOAP1 message protocol) that represented a model of software delivery based on the notion that pieces of software applications can be developed and then published to a registry where they can be dynamically discovered and consumed by other client applications over different transport protocols (e.g., HTTP, TCP/IP, etc.) irrespective of the language used to develop those applications or the platforms (e.g., operation systems, Internet servers) on which they are implemented. This was a dramatic improvement over the services provided by ASPs which relied on proprietary (and hence un-portable) software.

What is cloud computing?

There seems to be many definitions of cloud computing around. A study by McKinsey (the global management consulting firm) found that there are 22 possible separate definitions of cloud computing. In fact, no common standard or definition for cloud computing seems to exist (Clark, 2007 14-15). A more commonly used definition describes it as clusters of distributed computers (largely vast data centers and server farms) which provide on-demand resources and services over a networked medium (usually the Internet). The “cloud” was probably inspired by IT text books' illustrations which depicted remote environments (e.g., the Internet) as cloud images in order to conceal the complexity that lies behind them.

Before proceeding any further at this stage, a word of caution is necessary. One must not assume that cloud products offered by any of the above services are likely to work out-of-the-box. In some cases they might. Google Apps, a messaging and collaboration cloud platform from Google, is probably one good example of those out-of-the-box products (even though it does require some level of configuration nevertheless). Many of the products that are offered by those three types of cloud services will require some degree of programming (by the user or indeed the cloud provider) in order to access the functionality that exists in those services. Cloud providers will have created their own APIs (application programming interfaces) so that software developers can use them to create client applications in order to access that functionality. Currently, some of those APIs are proprietary; an issue which will be revisited later when examining some of the limitations of cloud computing. However, some are based on open source standards such as SOAP or REST.

To demonstrate how those services can be utilized and the processes involved in their utilization (in a very simplified manner), a hypothetical example can be given. Take, for example, a typical university with an IT infrastructure that caters ...