Cloudware Technologies - Guide to Cloud Computing for Business and Technology Managers - page 303

5. Less space required : Each and every square yard of data center
space is scarce and expensive. With consolidation, the same per-
formance can be obtained on a smaller footprint and the costly
expansion of an existing data center might possibly be avoided.
6. Emergency planning : It is possible to move virtual machines from
one resource pool to another. This ensures better availability of
the services and makes it easier to comply with service-level
agreements. Hardware maintenance windows are inherently
no longer required.
Since the providers of cloud services tend to build very large resource
centers, virtualization leads not only to a size advantage but also to a
more favorable cost situation. This results in the following benefits for
the customer:
1. Dynamic behavior : Any request can be satisfied just in time and
without any delays. In case of bottlenecks, a virtual machine
can draw on additional resources (such as storage space and
I/O capabilities).
2. Availability : Services are highly available and can be used day
and night without stop. In the event of technology upgrades,
it is possible to hot-migrate applications because virtual
machines can easily be moved to an up-to-date system.
3. Access : The virtualization layer isolates each virtual machine
from the others and from the physical infrastructure. This way,
virtual systems feature multitenant capabilities and, using a
roles concept, it is possible to safely delegate management func-
tionality to the customer. Customers can purchase IT capabili-
ties from a self-service portal (customer emancipation).
The most direct benefit from virtualization is to ensure that
MapReduce engines work better. Virtualization will result in
better scale and performance for MapReduce. Each one of the
map and reduce tasks needs to be executed independently. If the
MapReduce engine is parallelized and configured to run in a virtual
environment, you can reduce management overhead and allow for
expansions and contractions in the task workloads. MapReduce itself is
inherently parallel and distributed. By encapsulating the MapReduce
engine in a virtual container, you can run what you need whenever you
need it. With virtualization, you can increase your utilization of the
assets you have already paid for by turning them into generic pools of
resources (see Chapter 17, Section 17.2 “Google MapReduce”).