Scalable Computing and Communications: Theory and Practice, First Edition. Edited by
Samee U. Khan, Albert Y. Zomaya, and Lizhe Wang.
© 2013 John Wiley & Sons, Inc. Published 2013 by John Wiley & Sons, Inc.
1
Scalable Computing and
Communications: Past,
Present, and Future
Yanhui Wu, Kashif Bilal, Samee U. Khan, Lizhe Wang, and Albert Y. Zomaya
1.1 SCALABLE COMPUTING AND COMMUNICATIONS
Scalability is a paradigm that can adapt to the need of computing requirements of
the underlying applications and users. Scalability [1, 2] is also a desirable quality for
a network, process, website, or business model. In terms of hardware, a scalable
computer system may begin with one node, but more nodes can be added as and
when there is a need for more computing capabilities. Scalability, when sold with IT
equipment or software, is a feature to convince high-growth businesses that the
future needs can be accommodated easily and without recourse to expensive
machine replacement or staff retraining. However, a scalable system need not be at
one physical address. The ease of availability of high-speed networks and powerful
computers has led to the emergence of two computing trends: (1) cluster computing
[3, 4] and (2) grid computing [5–9]. Geographically, remote desktop computers,
storage systems, data sources, scientific instruments, and clusters can be combined
into what are known as computational grids.
Cloud computing [10–17] is an emerging paradigm in which users export data
and applications (or computations) to the “cloud” (a remote set of machines) and
then access the data or application in a simple and pervasive way. However, the
aforementioned is a classic example of central processing. Interestingly, about 50
1
COPYRIGHTED MATERIAL