Cloud Computing: Exploring Edge Computing's Impact and Applications

This report provides an overview of cloud computing and its evolution, highlighting the emergence of edge computing as a significant paradigm shift. It discusses the economic advantages of cloud computing, such as cost reduction through centralization, but also addresses the limitations of relying solely on centralized cloud infrastructure, particularly for latency-sensitive applications in areas like IoT and mobile computing. The report introduces edge computing concepts, including cloudlets, fog nodes, and micro datacenters, emphasizing their role in bringing computing resources closer to the data source. It explores the benefits of proximity in terms of reduced latency and bandwidth consumption, which are crucial for applications like virtual reality, augmented reality, and real-time data processing. The report also touches upon the Radio Applications Cloud Server (RACS) platform and the importance of edge computing in addressing data privacy concerns. References to relevant research papers provide additional context and support the analysis of these technologies.

Running head: CLOUD COMPUTING
CLOUD COMPUTING
Name of the Student:
Name of the University:
Author Note:

1CLOUD COMPUTING
Edge Computing
Cloud computing dominated sector of IT in recent time and has twofold value
proposition. Economics is exploited by centralization of scale for lowering marginal cost for
system operations and administration. Organizations could avoid capital expenditure to create
datacenter by consumption of computing resource on internet from provider of large service.
Such considerations led to consolidation of capacity of computing in several larger datacenters
that are spread all over the world. Cloud computing’s proven benefits in economy make this
likely in remaining permanent feature for future landscape of computing (Satyanarayanan, 2017).
However, Nascent applications and technologies for Internet of Things (IoT) and mobile
computing are taking computing close to dispersion. Edge computing is the new paradigm where
substantial storage and computing resources such as cloudlets, fog nodes and micro datacenters
are placed in edge of internet to mobile sensors and devices.
Research interest and industry investment within edge computing grew in past few years.
Radio Applications Cloud Server (RACS) is introduced jointly by IBM and Nokia, which is a
platform of edge computing for 4G networks. As new use cases and applications are explored for
both IoT and mobile computing, virtues for proximity are increasingly becoming apparent. For
physical world, proximity’s importance was never in doubt. In cyber world, seamless
connectivity has lulled in false sense for disregard of physical proximity. Physical proximity is
dependent upon factors like use of networking technologies, user tolerance, application
characteristics and network contention (Pahalovi, et al., 2017). With sufficient investment in
resource and effort, lack of proximity could be masked partially. However, limits are there for
such approach. Need for using strategy of multihop networking for covering hug geographic

2CLOUD COMPUTING
area having several access points force economic limit over both bandwidth and latency. Every
hop introduces routing and queuing delay along with buffer bloat.
Humans are sensitive in delays to interaction’s crucial path and the performance over
cognitive tasks is quite accurate and quick. It takes speech recognition 300ms to 450ms for the
short phrases. It needs 4ms only for telling that sound is human voice. Applications of VR need
latencies lesser than 16ms for achieving perceptual stability (Rittinghouse & Ransome, 2016).
For recognition of face, person’s identity is returned. For AR, the buildings within image would
be recognized and the labels to such identities would be transmitted to mobile devices.
Performance is impeded by end-to-end latency of network. Increasing time of response increases
also energy consumption per operation on mobile devices.
Cloudlets are the disruptive technology which brings high-end computing in a wireless
mobile device, hence enabling the new applications which are both latency sensitive and
computation intensive. On cloudlet, workflow of the applications includes two phases. Within
first phase, there is analysis of sensor inputs for extracting symbolic representation about
progress of task. The phase should tolerate considerable variability of real world such as
different light sources and lighting levels. Cloudlets could reduce also ingress bandwidth in
cloud (Shabtai, Raz & Shavitt, 2017). Bandwidth and latency advantages of cloudlets are
relevant within respect to automobiles, for complementing approach of vehicle-to-vehicle (V2V)
are explored for accident avoidance and control. Cloudlets help in addressing vexing issue such
as growing concerns of privacy of data arising from system overcentralization of IoT.
Developing innovative solutions is research priority of edge computing.

3CLOUD COMPUTING
References
Pahalovi, M. M. S., Islam, M. R., Adhikary, T., & Razzaque, M. A. (2017, December). Optimal
execution of virtualized network functions in a multi-data-center cloud. In 2017 IEEE
Region 10 Humanitarian Technology Conference (R10-HTC) (pp. 602-605). IEEE.
Rittinghouse, J. W., & Ransome, J. F. (2016). Cloud computing: implementation, management,
and security. CRC press.
Satyanarayanan, M. (2017). The emergence of edge computing. Computer, 50(1), 30-39.
Shabtai, G., Raz, D., & Shavitt, Y. (2017, September). Risk aware stochastic placement of cloud
services: the multiple data center case. In International Workshop on Algorithmic Aspects
of Cloud Computing (pp. 138-156). Springer, Cham.