Evaluating Cloud & Fog Computing based on Shifting & Scheduling Algorithms, Latency Issues and service Architecture

Evaluating Cloud & Fog Computing based on
Shifting & Scheduling Algorithms, Latency Issues
and service Architecture
Sidra Anwar1
, Ammara Ajmal2
, Faraniya Hayder3
, Sehresh Bibi4
1,2,3,4
Computer Science Department
GC Women University Sialkot
Sialkot, Pakistan
1
sidra.anwar@gcwus.edu.pk, 2
haniya.janjua41@gmail.com,
3
faraniya456@gmail.com, 4
sehrishanwar706@gmail.com
Abstract— In this study, we propose situations where cloud is
suitable and fog is more compatible, also define some services
according to the cloud and fog architecture. We also provide a
comparison of task scheduling algorithms of cloud computing and
determine that fog is a light weight network so which is the best
suitable algorithm for fog architecture on the basis of some
attributes. The implementations of fog computing are challenging in
today’s computational era; we define some reasons in which fog
computing implementation is difficult.
KEYWORDS:
Cloud and Fog computing, scheduling, Internet of Things
(IOT), fog nodes.
I. INTRODUCTION
Today cloud services come with some obvious benefits,
cloud data centers are centralized and, typically far from the
end user resulting in high access latencies (Mell & Grance,
2011). A lot of companies i.e., Google, Amazon, Microsoft
and many other companies increased their speed in
development of cloud computing systems in order to enhance
cloud services that are provided to many users( Zhu, Jiang, et
al. 2013). At this point when information sources are
distributed over numerous areas and low latency is necessary,
cloud information preparing fails to meet this need. The
reason for building such condition is to collect information
from IoT gadgets to perform optimization, pattern detection
and predictive analysis to make smarter decisions
timely(Misra, Prasant, et al. 2015).
The phrase ―Fog computing‖ also termed as edge computing,
is a decentralized architecture unlike cloud computing where
central mainframe is required to access data (Kitchin & Rob.
2014). Fog computing is designed to serve as an extension to
cloud computing by grouping many edge nodes. That provides
sufficient amount of management and configuration, and
localized control. Fog computing serves at the edge of the
network (Cortés, Rudyar, et al., 2015), it permits cloud
computing to extend their services to the edge of the network
to make their services more accessible (Gupta, 2017).
Following key Questions are answered in this paper:
 What type of services need fog computing rather than
cloud computing architecture?
 How Fog will reduce service latency to Improve QOS?
 Either shifting of user data from one fog node to another
face any interruption or not?
 Which efficient scheduling algorithm should be used to
shuffle data among devices without interruption of data
among fog nodes?
Method: Systematic research methodology is undertaken,
Articles enumerating emergence of fog computing and cloud
computing and comparison of services of Fog and cloud
computing as well as the comparison of different scheduling
algorithms are identified.
Literature review:
According to Ivan Stojmenivis Sit, Shen Wen, Fog
Computing isn't a substitution of cloud but a point ahead to it.
As Cloud, as well as Fog computing also provides application
services, data (computation etc.) to end-users (Behrisch,
Michael, et al., 2011). The main variance is Fog computing
services to its nearest or closest user thorough its node that is
close to user geographically that provides mobility, also
termed as edge network (Aazam, Mohammad, et al., 2015).
Low latency and quality of service is achieved through fog
computing (Dsouza, Clinton et al., 2014). Fog figuring is most
appropriate for analytics and real-time big data as per Cisco
because of its wide localized distribution.
II. WHY THERE IS NEED OF FOG COMPUTING?
A current study of Endomondo (Yuan, Mugen et al., 2018)
sport activity analyzing application has exposed number of
International Journal of Computer Science and Information Security (IJCSIS),
Vol. 16, No. 6, June 2018
9 https://sites.google.com/site/ijcsis/
ISSN 1947-5500

noteworthy opinions? This review indicates that an activity
produces 170 GPS; maximum tuples can be 6.3 million in a
specific period. Estimated study indicates that till 2020 with
30 million user’s data generated by Endomondo can be 25,000
per second (Cortes, Rudyard, et al., 2015). So with this
rapidity Centralized cloud servers cannot deal in real-time
(Zhu, Jiang, et al, 2013). The users of cloud especially privacy
conscious are not comfy with this. This inspires the need of
alternate the model that is accomplished of more
computational nodes that are nearer to the user than cloud with
internet connection, also can construct local views of data and
can perform further offload computations. Fog computing has
arisen to this end.
III. COMPARISON OF CLOUD COMPUTING AND FOG
COMPUTING
Table 1: Comparison matrix of cloud computing and fog
computing
Attributes Cloud Computing Fog computing
Geo
Distribution
Centralized Decentralized
Latency Predictable and high
Low than cloud
computing
Need of
Internet
connection
Need constant and high
speed internet
connection
Perform offload
computational
operations
Reliability High standard (trusted)
Less than cloud
computing
Repairing
User is not concerned
about repairing
Maintenance is more
difficult and costly
Network
connectivity
Guaranteed but with
heavy latency
Unreachable (all the
time same node is
not accessible
Cost factor
Costly as CAAS
perspective
Auditing of hardware
is costly
Scheduling
Un-necessary latency
un scheduling
complex
Service
Provide service from its
core architecture
Provide service from
the edge of the
network
Maintained or
owned by
Single motivated
organization
Can be implemented
by many independent
agents
Power
consumption
More power
consumption
Power efficient
Heterogeneity
in nature
Heterogeneous in
services perspective
Heterogeneous in
device (node)
specification
Location of
the services
Within the network
(from its core
architecture)
From the edge of the
local network
Client sever
distance
Multiple hopes Single hope
Security Undefined Can be defined
Distance
between
server and
client
Numerous hops One hop
IV. WHAT TYPE OF SERVICES NEED FOG COMPUTING RATHER
THAN CLOUD COMPUTING ARCHITECTURE?
Some characteristics like mobility, cloud integration,
communication protocols, computing resources and
distributed data analytics are supported by Fog computing that
acquire low latency with wide range and in dense geographical
distribution fog improve quality of services, decrease the rate
of latency of services requests (Zhu, Jiang, et al., 2013), that
results in a better user experience.
A. Architecture of Fog computing
There are three logical layers in fog network architecture.
Each device is capable enough to provide storage, host
computations, and provides offload computational operations
(Chiang, Zhang 2016), (working like online- to some extent)
 Device Layer: All the devices that are connected to
the fog network through nodes are included in this
layer i.e., IoT devices e.g., gateways, sensors, and
smart phones like tablets, mobile devices etc. it is
possible that these devices are using peer-to-peer
communication among themselves (Zhu, Jiang, et al.,
2013) or may be exchange data directly with the
network. The word ―data‖ is plural, not singular.
 Fog Layer: All the intermediate network devices that
are placed between end devices and cloud layer are
part of this layer (Gupta1, Chakraborty S. et al., 2016).
These intermediate network devices decrease the rate
of burden on Cloud Layer. The word ―data‖ is plural,
not singular.
 Cloud Layer: In hierarchical architecture of these
three layers, this layer is at apex; (Rehman, Jayaraman
et al., 2017) at this point cloud virtual machines
provide offload computations. Handling intensive
computation, high processing and need of large
storage is possible in cloud infrastructure (Rehman;
Liew, et al. 2014) due to its infinite scalability and
high end performance.
B. Services that fog architecture supports:
 Reduction of network traffic: Billions of devices use
internet to send, receive, and generate data every few
second. It is not sensible and efficient to send all raw
data on cloud. There are 25 billion connected devices,
estimated by cisco. It is the need of modern era to
implement an architecture in which they use devices
that are capable enough of being closer to the
customer to reduce latency that lead towards improves
quality of service (Cao, Yu, et al., 2015). Here fog
computing severely reduces this burden on cloud. Fog
computing provides a platform where generated data is
analyzed and filtered also capable of closed to edge of
the network (Stantchev, Vladimir, et al., 2015).
 Appropriate for IoT (Internet of Things) queries
and tasks: Due to increasing trend of smart devices,
many requests influence their surrounding devices.
This type of requests can be handled without acquiring
Vol. 16, No. 6, June 2018
ISSN 1947-5500

global information at cloud that’s why fog is suitable
for IoT Tasks. For example, here we will discuss the
above mentioned scenario Edomondo (Luan; Gao; et
al. 2015), by using this support tracker application
user locate the similar support in their nearby
locations, as it is clear that these types of requests are
from local area, so fog infrastructure is better to handle
such requests and IoT devices instead of using cloud
infrastructure.
 Low latency requirement: Applications that are
mission critical need real time data processing. Let’s
take an example of robotics, its motion it controlled by
feedback of the control system, and data collected
form sensors. Having this control system on cloud
may make it slower or may be unavailable sometime
that may result in communication failure (Rehman,
Liew, et al. 2014). Here fog computing performs its
role. Fog computing can help by performing real time
processing required for controlling (Rehman, Liew, et
al. 2014). Because fog nodes are very close enough to
the robots. This is how Fog architecture makes real
time response and communication possible.
 Scalability: Due to increased number of devices there
is infinite number of virtual resources. It is overhead
for cloud to handle if we upload all the crude data
created by nodes on cloud (Rehman, Liew, et al.
2014). Rather than that the Aim of fog computing is to
process data from incoming closer device. This
infrastructure reduces the burden from cloud servers
(Zhu, Jiang, et al., 2013). This is how fog overcomes
the scalability issue of cloud computing by using its
end nodes.
 Security and Reliability: In Cloud architecture it is
most challenging task to implement security over it.
Authentication implementation at several levels of fog
nodes is a main task (Trovati, M. 2017). Possible
results that may show to be valuable for this
authentication problem in fog computing are Trusted
execution environment (TEE) techniques (Patty, Penn,
2015) Public Key Infrastructure and also Measurement
based methods also reduce authentication cost.
 Resource Management: For IoT applications that
perform efficient and effective management of
resources offering a service oriented resource
management model for fog computing (Yang, Zhang,
et al. 2014). Fog for competent process of fog
sensible management of resources is important.
Aazam et al. as usual servers of cloud that has
resource capacity, merely fog devices match (Wang,
Lu, et al. 2013).
V. HOW FOG WILL REDUCE SERVICE LATENCY AND
IMPROVE QUALITY OF SERVICE (QOS)?
As the fast growth of IoT applications, the traditional
centralized cloud computing is encountering severe
challenges, such as high latency, low spectral efficiency (SE),
and nonadaptive machine type of communication (Fu, Jiang,
et al. 2012). The evolving IoT presents new challenges that
cannot be effectively addressed by the central cloud
computing architecture, such as resource-constrained devices,
stringent latency, uninterrupted services, capacity constraints
with alternating connectivity, and enhanced security (Dong,
Douglis, et al. 2017). Fog computing is acting like smart
gateway that performs following tasks:
 Local Connectivity: Fog gateway directly contact with
sensors to gain data and to deliver actuations such as
alerts and notifications.
 Computation: for comprehensive analysis Fog gateway
handles the incoming data to produce logs that are sent to
the cloud.
 Onsite Database: It forms a local database encompassing
features that internally and externally can be queried.
 Data Security: To protect the data and user identity, it
provides security layer.
In the context of big data an auspicious tactic is Fog
computing, a lot of possible operations can be possible due to
this (Zerbino & Birney, 2008), Mobile edge devices have
some computational capabilities, that deployed an imaginary
data processing architecture called RedEdge (Lin, Chiu, et al.
2013). The mobile edge devices facilitate with data reduction
platform the RedEdge model. The big data systems enhanced
its Value, velocity and volume, RedEdge architecture is used
to reduce data into streams (Randles, Lamb et al. 2010) before
storing big data, RedEdge reduce the amount of big data. This
reduction of data is based on following factors:
(1) High dimensional datasets are reduced into streams of low
dimensional data-sets
(2) by graph mapping and optimization algorithms in network
theory-based methods;
(3) The volume of network trafﬁc can reduce in order to
applied compression algorithms;
(4) Redundant and replicated data eliminate through data
deduplication methods;
(5) With the help of data filtration methods and feature
extraction, data streams can be reduced at early stages;
(6) Data reduction of Big data can be managed at early stages
with machine learning and datamining techniques
VI. DOES A PROCESS RUNNING ON FOG NODE BE
INTERRUPTED WHEN THE USER MOVES TOWARDS ANOTHER
FOG NODE?
No, fog node cannot be interrupted by another fog node
because from nearby devices the data accumulated in Fog
Data is time-series along with time-stamps (Subramanian,
Krishna et al., 2012). These log files help in debugging the
software programs for certain use cases. Thus, it is known
what data is collected and at what time. Along with other
important factors data collection is accompanied with a log
Vol. 16, No. 6, June 2018
ISSN 1947-5500

file that stores the time-stamps of data acquirement. The log
files are .xls files
VII. WHICH EFFICIENT SCHEDULING ALGORITHM SHOULD BE
USED TO SHUFFLE DATA AMONG DEVICES WITHOUT
INTERRUPTION OF DATA AMONG FOG NODES?
There are many technologies that are developed to meet the
need of ―busy users‖ in the modern era of competition. In
order to reduce latencies and offer better QoS, fog computing
architecture executes workload of computations on the edge
nodes of network, that are nearer to the user (Kokilavani &
Amalarethinam, 2011). This architecture of for computing
results in reduction of latency in communication between edge
nodes and users (Varghese, Wang, et al. 2017). Modern
technologies have more users if they can address user’s
demands without adding more workload on these technologies
(devkota, Ghimire, 2017). Task scheduling algorithms directly
affect the efficiency of resource utilization and also to the
proficiency of users’ tasks. There are a lot of task scheduling
techniques or algorithms that is used in cloud to balance the
load (Raghava & Singh, 2014). A comparison matrix of
different scheduling algorithms is given below with respect to
different attributes.
Table 2: Comparison matrix of different Scheduling Algorithms
Algorithms Attributes
HoneyBee
Foragging (Mao,
Chen, et al. 2014)
No
Throughput
No
Overhead
No Fault
Tolerance
No
Migration
Time
No
Response
Time
Resource
Utilization
available
No
Scalability
No Performance
Active Clustering
(Wickremasinghe,
2009)
No
Throughput
Overhead
available
No Fault
Tolerance
Migration
Time
No
Response
Time
Resource
Utilization
available
No
Scalability
No Performance
PALB (Kaur,
2012)
Throughput
available
Overhead
available
Fault
Tolerance
available
Migration
Time
available
Response
Time
available
Resource
Utilization
available
No
Scalability
No Performance
Round Robin
(Wang, Yan, et
al. 2010)
Throughput
available
Overhead
available
No Fault
Tolerance
No
Migration
Time
Response
Time
available
Resource
Utilization
available
Scalability
available
Performance
available
Min-Min (Nine,
SQ, et al. 2013)
Throughput
available
Overhead
available
No Fault
Tolerance
No
Migration
Time
Response
Time
available
Resource
Utilization
available
No
Scalability
Performance
available
Max-Min (Lin,
Liu, et al. 2011)
Throughput
available
Overhead
available
No Fault
Tolerance
No
Migration
Time
Response
Time
available
Resource
Utilization
available
No
Scalability
Performance
available
Active Monitoring
(Kaur & Bharti
2014)
Throughput
available
Overhead
available
No Fault
Tolerance
Migration
Time
available
Response
Time
available
Resource
Utilization
available
Scalability
available
No Performance
OLB+LBMM
(Bonomi, Flavio,
et al. 2012)
No
Throughput
No
Overhead
No Fault
Tolerance
No
Migration
Time
No
Response
Time
Resource
Utilization
available
No
Scalability
Performance
available
Biased Random
Sampling (Li,
Zhao, et al. 2015)
No
Throughput
Overhead
available
No Fault
Tolerance
No
Migration
Time
No
Response
Time
Resource
Utilization
available
No
Scalability
Performance
available
Throttled (Li,
Zhao, et al. 2015)
No
Throughput
Overhead
available
Fault
Tolerance
available
Migration
Time
available
Response
Time
available
Resource
Utilization
available
Scalability
available
Performance
available
Vol. 16, No. 6, June 2018
ISSN 1947-5500

Dynamic Round
Robin (Hromic,
Phuoc, et al.
2015)
Throughput
available
Overhead
available
Fault
Tolerance
available
Migration
Time
available
No
Response
Time
Resource
Utilization
available
No
Scalability
No Performance
FAMLB (Vaquero
& Rodero-
Merino, 2014)
Throughput
available
Overhead
available
Fault
Tolerance
available
Migration
Time
available
No
Response
Time
Resource
Utilization
available
Scalability
available
Performance
available
Fog Computing is also balance the load between the edge
nodes but it is light weighted than cloud so it also required an
algorithm which is also light weight and satisfy all above
mentioned attributes. So, to fulfill the needs of the fog
computing there must be an algorithm which will be the
combination of two algorithms that perform all the above
mentioned features (Wang, Yan, et al. 2010). According to
the above comparison an algorithm which will be the
combination of FAMLB and Throttled algorithm can fulfill
the requirements of fog computing. The frame work of fog
computing architecture is not yet available that’s why its
implementation is difficult and still it is infancy. Fog
architecture should be able to fulfill the requirement, i.e., we
need such a technology that is capable of processing requests
in real-time. Implementation of real-time data processing in
fog nodes is still a question mark (Lin, Liu, et al. 2011).
Although, current framework of cloud computing i.e., Google
App Engine, Microsoft Azure, and Amazon Web Service
(Gupta, Chakraborty et al. 2016), are capable of supporting
data intensive applications. Furthermore, there will need to be
well plain how to develop a framework, that is capable of
handling workload on fog nodes. Secondly, how to deal with
different types of nodes on which such applications will
deploy, i.e., connection policies that presents when to use
heterogeneity and edge, and Deployment strategies that show
where to place a workload.
VIII. DISCUSSION
This paper reviews the use of fog computing as an extension
of cloud computing by analyzing the comparison of fog and
cloud. Define some services which fog computing supports
and which methods of reduction used to better the Quality of
service and control the latency of services. Different
scheduling algorithms learn and compared them with each
other to determine which will be the best suitable for fog
architecture and also define what challenges will face in the
implementation of fog computing.
IX. CONCLUSION AND FUTURE WORK
We determine fog computing is essential to facilitate better
end user environment to get better and quick real-time
response of requests especially for sensitive applications in
current era of computations. Also investigate that fog
computing provide better quality of service at the edge of the
network and the comparison of different scheduling
algorithms define some innovations in compute, storage and
scheduling may be inspired in the future to handle data
intensive services between different fog nodes. In future we
will work on the implementations of the algorithms that will
be the best suitable for the fog environment and fulfill all the
perspectives of attributes that mentioned to compare
algorithms
References
[1] Mell, P., Grance, T.: The NIST deﬁnition of cloud computing. NIST
Special Publication 800-145 (2011)
[2] Zhu, Jiang, et al. (2013), Improving web sites performance using edge
servers in fog computing architecture. Service Oriented System
Engineering (SOSE), IEEE 7th International Symposium on.
[3] Misra, Prasant, et al. (2015). Towards a Practical Architecture for the
Next Generation Internet of Things. Retrieved from arXiv preprint
arXiv:1502.00797.
[4] Kitchin & Rob. (2014). The real-time city? Big data and smart urbanism.
GeoJournal, 79,1. Pp. 1-14.
[5] Cortés, Rudyar, et al. (2015). Stream Processing of Healthcare Sensor
Data: Studying User Traces to Identify Challenges from a Big Data
Perspective. Procedia Computer Science 52: 1004-1009.
[6] Behrisch, Michael, et al. (2011). SUMO–Simulation of Urban MObility.
The Third International Conference on Advances in System Simulation
(SIMUL 2011), Barcelona, Spain.
[7] Aazam, Mohammad, et al. (2015). Dynamic resource provisioning
through Fog micro datacenter. Pervasive Computing and
Communication Workshops (PerCom Workshops), 2015 IEEE
International Conference on.
[8] Dsouza, Clinton et al. (2014). Policy-driven security management for
fog computing: Preliminary framework and a case study. Information
Reuse and Integration (IRI), 2014 IEEE 15th International Conference
on.
[9] Yuan A., Mugen P. et al. (2018). Edge cloud computing technologies for
internet of things: A primer. Digital Communications and Networks.
4(2), Pp. 77-86.
[10] Chiang M., Zhang T. (2016), Fog and IoT: An Overview of Research
Opportunities, IEEE Internet Things J.3, 854–864.
[11] Gupta1 H., Chakraborty S. et al. (2016). Fog Computing in 5G
Networks: An Application Perspective retrieved from
https://pdfs.semanticscholar.org/12d8/13f14166578dea8aa6aacc945102d
ddfd05d.pdf
[12] Rehman M. H. Jayaraman P. P. et al. (2017) A Novel Architecture for
Big Data Processing in Mobile Edge Computing Environments, 6(3), 17,
retrieved from http://www.mdpi.com/2224-2708/6/3/17/htm
[13] Luan, T.H.; Gao, L.; et al. (2015). Fog Computing: Focusing on Mobile
Users at the Edge. arXiv 2015, arXiv:1502.01815.
[14] Rehman, M.H.; Liew, C.S. et al. (2014). Frequent pattern mining in
mobile devices: A feasibility study. In Proceedings of the International
Conference on Information Technology and Multimedia (ICIMU),
Putrajaya, Malaysia, 18-20. pp. 351–356.
Vol. 16, No. 6, June 2018
ISSN 1947-5500

[15] Rehman, M.H.; Batool, A.; et al. (2017). Execution Models for Mobile
Data Analytics. IT Prof., 19, 24–30.
[16] Rehman, M.H.; Liew, C.S. et al. (2014). UniMiner: Towards a unified
framework for data mining. In Proceedings of the 4th World Congress
on Information and Communication Technologies (WICT), Malacca,
Malaysia, 8–10, pp. 134–139.
[17] Trovati, M. (2017). Reduced topologically real-world networks: A big-
data approach. Int. J. Distrib. Syst. Technol, 6, 13–27.
[18] Patty, J.W.; Penn, E.M. (2015) Analyzing big data: Social choice and
measurement. PS Political Sci. Politics, 48, 95–101.
[19] Yang, C.; Zhang, X.;et al. (2014) A spatiotemporal compression based
approach for efficient big data processing on cloud. J. Comput. Syst.
Sci., 80, 1563–1583.
[20] Wang, W.; Lu, D.; et al. (2013). Statistical wavelet-based anomaly
detection in big data with compressive sensing. EURASIP J. Wirel.
Commun. Netw, 1–6.
[21] Fu, Y.; Jiang, H.; et al. (2012). A scalable inline cluster deduplication
framework for big data protection. In Middleware 2012; Springer: New
York, NY, USA; pp. 354–373.
[22] Dong, W.; Douglis, F. et al. (2017). Tradeoffs in Scalable Data Routing
for Deduplication Clusters. Retrieved from
https://www.usenix.org/legacy/events/fast11/tech/full_ papers/Dong.pdf
[23] Zerbino, D.R.; Birney, E. Velvet: (2008). Algorithms for de novo short
read assembly using de Bruijn graphs. Genome Res., 18, 821–829.
[24] Lin, M.S.; Chiu, C.Y. et al. (2013) Malicious URL filtering—A big data
application. In Proceedings of the IEEE International Conference on Big
Data, Silicon Valley, CA, USA, 6–9; pp. 589–596.
[25] Randles, M. Lamb D., et al. (2010), A comparative study into distributed
load balancing algorithms for cloud computing. in Advanced
Information Networking and Applications Workshops (WAINA), 2010
IEEE 24th International Conference on, pp. 551–556.
[26] Galloway, J. M. Smith, K. L. et al. (2011). Power aware load balancing
for cloud computing, in Proceedings of the World Congress on
Engineering and Computer Science, vol. 1, pp. 19–21, 2011.
[27] Subramanian, S. Krishna, G. N. et al. (2012). An adaptive algorithm for
dynamic priority based virtual machine scheduling in cloud.
International Journal of Computer Science Issues (IJCSI), 9(6).
[28] Kokilavani T. & Amalarethinam, D. (2011). Load balanced min-min
algorithm for static meta-task scheduling in grid computing.
International Journal of Computer Applications, 20(2).
[29] Mao, Y. Chen, X. et al. (2014). Max–min task scheduling algorithm for
load balance in cloud computing. in Proceedings of International
Conference on Computer Science and Information Technology,
Springer, pp. 457–465.
[30] Wickremasinghe, B. (2009). Cloudanalyst: A cloudsim-based tool for
modelling and analysis of large scale cloud computing environments.
MEDC Project Report, 22(6), pp. 433–659.
[31] Kaur, J. (2012). Comparison of load balancing algorithms in a cloud.
International Journal of Engineering Research and Applications, 2(3),
pp. 1169–173.
[32] Wang, S.-C. Yan, K.-Q. et al. (2010). Towards a load balancing in a
three-level cloud computing network. in Computer Science and
Information Technology (ICCSIT), 2010 3rd IEEE International
Conference on, vol. 1, pp. 108–113.
[33] Nine, Z. SQ, M. et al. (2013). Fuzzy logic based dynamic load balancing
in virtualized data centers. in Fuzzy Systems (FUZZ), 2013 IEEE
International Conference on, pp. 1–7.
[34] Lin, C.-C. Liu, P. et al. (2011). Energy-aware virtual machine dynamic
provision and scheduling for cloud computing. in Cloud Computing
(CLOUD), 2011 IEEE International Conference on, pp. 736–737.
[35] Kaur M. & Bharti M. (2014). Securing user data on cloud using Fog
computing and Decoy technique. International Journal of Advance
Research in Computer Science and Management Studies, 2(10).
[36] Bonomi, Flavio, et al. (2012). Fog computing and its role in the internet
of things. Proceedings of the first edition of the MCC workshop on
Mobile cloud computing. ACM, pp. 13-16.
[37] Li, W. Zhao, Y. et al. (2015). Mechanisms and challenges on Mobility-
augmented Service Provisioning for Mobile Cloud Computing, IEEE
Communications Magazine, 53(3), pp. 89–97.
[38] Hromic, H. Phuoc, D. L. et al. (2015). Real Time Analysis of Sensor
Data for the Internet of Things by Means of Clustering and Event
Processing, in Proceedings of the IEEE International Conference on
Communications, pp. 685–691)
[39] Vaquero, L.M. & Rodero-Merino, L. (2014). Finding your way in the
fog: Towards a comprehensive definition of fog computing, ACM
SIGCOMM Computer Communication Review 44(5): 27-32
[40] The Future Is in Fog Computing - DZone IoT. (n.d.). Retrieved July 4,
2018, from https://dzone.com/articles/future-is-in-fog-computing
Vol. 16, No. 6, June 2018
ISSN 1947-5500

Evaluating Cloud & Fog Computing based on Shifting & Scheduling Algorithms, Latency Issues and service Architecture

More Related Content

What's hot (20)

Similar to Evaluating Cloud & Fog Computing based on Shifting & Scheduling Algorithms, Latency Issues and service Architecture (20)

Recently uploaded (20)

Evaluating Cloud & Fog Computing based on Shifting & Scheduling Algorithms, Latency Issues and service Architecture