Page 280 -
P. 280
Appendix B
Green IT and Cloud Computing 245
enterprises for such diverse applications as drug discovery, economic
forecasting, seismic analysis, and back-office data processing in support
of e-commerce and web services. However, this arrangement is thought
to be well-suited to applications in which multiple parallel computa-
tions can take place independently, without the need to communicate
intermediate results between processors because various processors and
local storage areas may not have high-speed connections. In grid com-
puting, the computers and resources involved can be heterogeneous.
They can run different operating systems and use different server and
storage platforms that may even be owned by different individuals, com-
panies, labs, or universities. Grids may include clusters, individual
servers, or entire data centers as part of the resources that are virtualized
for sharing with other members on the grid. Grids are capable of massive
scaling to share and manage applications, data, and storage resources
among local, campus, regional, and international locations. Google
search engine technology is supposed to use thousands of servers in the
form of grid computing for processing search requests, crawling the web
for new sites, storing documents, and managing advertisements.
ptg
On the other hand, virtualization is about abstracting computing
resources in such a manner that a larger computing resource can be
apportioned and provided as smaller resources. The concept was pio-
neered by the time-sharing concepts of the mainframes of the sixties,
and there are various forms of virtualization today. For example, IBM
System z mainframes are capable of multiple levels of virtualization. A
single machine can be divided into several logical partitions (LPARs),
and each logical partition is a separate virtual machine running a sepa-
rate operating system instance. The same capability is available in mid-
range computers, such as IBM’s System p and others. VMWare is a
frequently quoted name in the arena of server virtualization. VMWare
provides software solutions that partition a physical server into multiple
virtual machines. VMware plans to expand the reach of its virtualization
platform offering to integrate with other aspects of the data center, such
as network and storage. System management, provisioning, workload
management, monitoring, metering, and security are other technologies
that are needed for both grid computing and virtualization.