Grid Computing - Guide to Cloud Computing for Business and Technology Managers - page 243

The benefits gained from grid computing can translate into competitive
advantages in the marketplace. Grids enable the following:
• Enable resource sharing
• Provide transparent access to remote resources
• Make effective use of computing resources, including platforms and
data sets
• Reduce significantly the number of servers needed by (25-75%)
• Allow on-demand aggregation of resources at multiple sites
• Reduce execution time for large-scale data processing applications
• Provide access to remote databases and software
• Provide load smoothing across a set of platforms
• Provide fault tolerance
• Take advantage of time zone and random diversity (in peak hours,
users can access resources in off-peak zones)
• Provide the flexibility to meet unforeseen emergency demands by
renting external resources for a required period instead of owning
them
• Enable the realization of a virtual data center
Grid computing emerged in the early 1990s, when high-performance com-
puters were connected by fast data communication with the aim to support
calculation- and data-intensive scientific applications. Since the mid-1990s,
the concept of grid has evolved. Similar to other infrastructure innovations—
for example, the Internet—the grid was first introduced and adopted in sci-
ence for the support of research in various scientific disciplines that require
high-performance computing (HPC) together with huge amounts of data
stored in dedicated databases. Examples of such sciences are earth science,
astroparticle physics, and computational chemistry. They are summarized
under the term e-science . To support e-science, many national and interna-
tional initiatives have been started by governments in many countries in
order to leverage existing investments in research infrastructure and to
enable sharing and efficient use of available computational resources, data,
and specialized equipment.
Cloud computing is frequently compared to grid computing.
Grid computing also has the same intent of abstracting out
computing resources to enable utility models and was pro-
posed at least a decade earlier than cloud computing, and there
are many aspects of grid computing that have formed the basis of the
requirements placed on a cloud. However, there are also very specific