Increasing Functional Specificity over Increasingly Commoditized Hardware - Guide to Cloud Computing for Business and Technology Managers - page 8

and spreading it across multiple processors. To the user, it made little dif-
ference which CPU executed an application. Cluster management software
ensured that the CPU with the most available processing capability at that
time was used to run the code. A key to efficient cluster management was
engineering where the data were to be held. This process became known as
data residency. Computers in the cluster were usually physically connected
to magnetic disks that stored and retrieved data while the CPUs performed
input/output (I/O) processes quickly and efficiently.
In the early 1990s, Ian Foster and Carl Kesselman presented their concept
of The Grid . They used an analogy to the electricity grid, where users could
plug in and use a (metered) utility service. They reasoned that if companies
cannot generate their own power, it would be reasonable to assume they
would purchase that service from a third party capable of providing a steady
electricity supply. The same should apply to computing resources: if one
node could plug itself into a grid of computers and pay only for the resources
it used, it would be a more cost-effective solution for companies than buy-
ing and managing their own infrastructure. Grid computing expands on the
techniques used in clustered computing models, where multiple indepen-
dent clusters appear to act like a grid simply because they are not all located
within the same domain. A major obstacle to overcome in the migration
from a clustering model to grid computing was data residency. Because of
the distributed nature of a grid, computational nodes could be anywhere in
the world.
The issues of storage management, migration of data, and security pro-
visioning were key to any proposed solution in order for a grid model to
succeed. A toolkit called Globus was created to solve these issues, but the
infrastructure hardware available still has not progressed to a level where
true grid computing can be wholly achieved. The Globus Toolkit is an open-
source software toolkit used for building grid systems and applications. It
is being developed and maintained by the Globus Alliance and many oth-
ers all over the world. The Globus Alliance has grown into a community
of organizations and individuals developing fundamental technologies to
support the grid model. The toolkit provided by Globus allows people to
share computing power, databases (DBs), instruments, and other online tools
securely across corporate, institutional, and geographic boundaries without
sacrificing local autonomy.
In 2002, EMC offered a Content-Addressable Storage (CAS) solution called
Centera as yet another cloud-based data storage service that competes with
Amazon's offering. EMC's product creates a global network of data centers,
each with massive storage capabilities. When a user creates a document,
the application server sends it to the Centera storage system. The storage
system then returns a unique content address to the server. The unique
address allows the system to verify the integrity of the documents whenever
a user moves or copies them. From that point, the application can request