Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: THEend8_
(i) Explain two (2) benefits of using this approach.
(ii) Explain one (1) drawback of using this approach.
(b) Analyse the following statements, and give your opinion.
(i) “Grid computing is dead, cloud computing is the future.”
(ii) “Hadoop Distributed File System (HDFS) and MapReduce are
essential components of big data applications.”
2. (a) There are four types of cloud deployment models – public, private, hybrid
and community. Show which cloud model is the most suitable for the given
scenario.
(i) Company A needs highly controlled storage and access to their
databases as well as managing the infrastructure for web front ends
and other applications. They have a large existing IT infrastructure
and they are continually expanding the capabilities. They want to
satisfy all their current needs as well as to reduce cost.
(ii) Company B would like to leverage cloud computing to provide
advanced collaboration services (i.e. video, chat, and web
conferences) for its employees but does not have the IT resources
to deploy such an infrastructure.
(iii) Company C is considering a cloud environment to improve the
operating efficiency for their data and applications. It is part of an
industry where strict security and data privacy issues are of the
highest importance.
(iv) Company D wants to use virtualization in their computing
infrastructure. Since their data is highly sensitive, they want to
restrict access to within company premises only. They also want to
build and manage it themselves, without third-party involvement.
(b) Explain whether a static or dynamic scheduling algorithm is more suitable
for most big data applications
(c) Assume a program takes N seconds to execute running on a single
processor. If the same program is executed using X processors, we
expect its execution time to reduce to N/X seconds. However, this does
not happen. Discuss the reasons behind it.