Page 133 -

The Greening of IT
98 How Companies Can Make a Difference for the Environment

better and easier sharing of resources for a mix of lightly and heavily used
servers (as in the Server A/Server B preceding example), but also tends to
spread out the utilization over 24 hours on the large physical server that
houses the virtual servers.

Server Virtualization—the Reason

Current Asset Utilization (Stand-Alone Servers)
Peak-Hour Prime-Shift 24-Hour Period
Utilization Utilization Utilization

•Mainframes •85-100% •70% •60%

•UNIX •50-70% •10-15% •<10%

•Intel-based •30% •5-10% •2-5%
ptg

•Storage •N/A •N/A •52%

Source: IBM Scorpion White Paper: Simplifying the Corporate IT Infrastructure, 2000
Source: IBM White Paper
Figure 6.4 Server virtualization—the reason

The Ultimate in Server and Data Storage Virtualization

Grid computing is a major evolutionary step that virtualizes an IT infra-
structure. It’s defined by the Global Grid Forum (www.gridforum.org) as
distributed computing over a network of heterogeneous resources across
domain boundaries and enabled by open standards. Although the industry
has used server cluster technology and distributed computing over networks
for nearly two decades, these technologies cannot in themselves constitute
grid computing. What makes grid computing different is the use of open
source middleware to virtualize resources across domains. See Appendix B,
“Green IT and Cloud Computing,” for details.