Datasheet

20 Part I  Overview of Commerce
Disparate Systems
Distributed
Computing
P2P
(peer-to-peer)
Grids
Web
Services
Clouds
Clusters
Disparate systems
Resource sharing
Geographically sparse
Within a framework
Heterogeneous
application
interaction
High Performance
Computing
Sharing
Homogeneous
Results in high
performance
Resource sharing
Mainly for file sharing
Geographically sparse
No existing framework
Close to each other
Figure 1-8: Evolution of Grid Computing
Cloud Computing
Cloud computing is a natural evolution of the grid computing paradigm, which
approaches distributed computing from a slightly different perspective. Cloud
computing is a computing paradigm that combines architectural elements (for
example, virtualization or Web services) with business models (for example,
subscription-based and pay-per-use) to deliver IT services across the Internet.
Because the majority of cloud computing business models rely on maximizing
profit and utilizing the computing infrastructure, the infrastructure must be
very efficient. To be efficient, the infrastructure needs to expose, at the very
least, the following properties:
High utilization

Dynamicity — fast turnaround to limit idle capacity

Automation — lower management costs

Implementing the cloud computing business model would also require specific
capabilities such as accounting and chargeback, as well as service cataloging
(enabling self-service capabilities for its users).
In the same way applications have been deconstructed to evolve from mono-
lithic to distributed, to service oriented architecture (SOA)-based models, cloud
624463c01.indd 20 3/29/11 2:26:35 AM