Escolar Documentos
Profissional Documentos
Cultura Documentos
FACTORS
Mr.N.Vijayaraj1 and Dr.SenthilMurugan 2
1 Research Scholar, Department of Computer Science & Engineering in Vel Tech Dr.Rangarajan
andDr.SakunthalaRangarajan Technical University, Chennai, India.
2 Associate Professor, Department of Computer Science & Engineering in Vel Tech Dr.Rangarajan and
Dr.SakunthalaRangarajan Technical University, Chennai, India.
ABSTRACT
A cloud computing is a new age technology it enables users to access
infrastructure resources by using thin clients anywhere and at any time. In this paradigm, multiple users
can share cloud infrastructure resources. As military, academic, and commercial computing systems
evolve from autonomous entities that deliver computing products into network centric enterprise systems
that deliver computing as a service, opportunities emerge to consolidate computing resources, software,
and information through cloud computing. Resource sharing task is quite a challenging in cloud
environment. Usually resources are shared by user needs. These issues are , resource and reputation
management strategies are not well designed and they are not powerful. If the client selects the resource,
then the other resource nodes are neglected and there is no full utilization of resources and it doesnt meet
client Qosdemands. Resource sharing is the key technology in cloud computing. This paper provides the
survey on Resource sharing.
I.INTRODUCTION
Cloud computing has become a needed one by which we can provide services to the clients over the
internet. There are many number of cloud service providers such as Amazon, Google, and IBM etc. These
service providers charge according to the storage that we use, bandwidth and various other parameters.
The service provider cannot provide the services using only a cloud that is one cloud and it is not possible
when the clients are increasing. It also cannot provide resources to an application wholly in some
situations during high time [1]. In order to provide services, the researchers need to connect multiple
clouds having a implicit lab environment in order to provide super-computing capabilities to the clients in
order to fully make use of the resources. Due to this characteristics and developments in cloud computing,
the demand for collaborative cloud computing has grown. Due to CCC, we can provide services to people
where the resources belonging to different organizations are highly pooled. CCC interlinks various
physical resources to empower sharing of resources in the clouds in order to provide implicit perspective
of resources to its clients. This perspective is beneficial when a client requests resources and cloud doesnt
have sufficient resources. It has to discover and use the resources in different clouds [1]
II.RELATED WORK
Alarge-scale resource sharing system builds a virtual supercomputer by provided that an infrastructure for
sharing fabulous amounts of resources above the web. Cloud computing has been proposed to connect a
large number of clouds as an alliance that come together to share resources in order to better respond to
large-scale application requirements. collaborative cloud computing (CCC) can handle the situation when
a single cloud is not sufficient to provide sustainable high-quality service to some applications with
demand for scalable resources or when researchers want to build a virtual lab environment across
geographical distribution of physical hosts.[For example, cloud customer Drop box had around 100
million users in 2012 , and around50 million users in 2011, which is three times the number of 2010. As a
cloud may be overloaded during peak periods and stay idle in time periods with few service requests, it is
promising to integrate many dispersed clouds from different corporations and organizations to fully utilize
cloud resources.
The large-scale resource sharing system is a kinds possible the sharing of a different of resources
comprising software, data (music, video, books) storage, memory, CPU time, network bandwidth and
devices distributed over web. A computing resource (e.g., virtual machine) is stated by a set of attributes
such as CPU speed, memory and OS version and device name. A data resource also can be described by a
few keyword attributes. For example, if a node having a following features like OS name=Linux, Free
memory =1024 MB,CPU speed =1000 MHz [1]
As illustrated in Fig.1, the traditional common cloud its using mobile devices and shared resources in
remote data centers and acts as an agent between the real content providers and mobile devices. The
backbone of network is access the resources/services from data area to mobile devices often need to go
through. In contrast, the lightweight cloudlet can balance the scale of shared resources and the approach
overhead. A cloudlet is a trusted, resource-rich, Internet connected computer or a cluster of computers that
can be utilized by mobile devices via a high-speed wireless local area network (WLAN). In this MCC
architecture, mobile devices function as the clients and cloudlets as the service providers. Logical
interaction between them can be more easily achieved in the cloudlets physical proximity with the low
one-hop communication latency. Due to the spatial distributions of cloudlets and their distinct capabilities
or hosted resources, the mobile devices have different preferences over the cloudlets [2].
Deployment
Ownership
Location
model
Resource
Advantages
Disadvantages
Example
Amazon,
Sharing
Off-site
Low
End-users
part
of
the
resources
Private
Organization
Cloud
or third party
On-site
High
Resource
dedicated
Less flexible
Nebula
to
organization,
more secure
Hybrid
Organization
On-site or Medium
Maximum
need additional IT
Cloud
or third party
off-site
flexibility,
skills to operate
dedicated
and manage
resources
Community
Organization
On-site or High
Cloud
or third party
off-site
A. Virtual machines
TASK
Different kind of resource sharing mechanism are SCHEDULINGQ
proposed
in cloud . in the work done by Sheng Di[3],
dynamically optimize the performance in both competitive and non-competitive situation also plan to
ACCESS CONTROL
BAND WIDTH
RSC
Efficient
,Fidelity,
Flexibility
Multi Tenant
Bandwidth
allocation,
Traffic flow,
Network sharing
Resource
discovery,
DHT,
LSH & LIS
PRICING
SOCIAL
CLOUD
HYBRID
&Heterogeneo
us CLOUD
Economic,
Energy
Preference
based resource
allocation,
GATA
Resource
Pricing
,Clabacus,
Financial
Option, Fuzzy
logic
H-CRAN,
CRRM,
SPECTRUM
SHARING
Preference
based resource
allocation,
GATA
B.Risk
Risk is one important parameter in cloud computing paradigm. Abdurahman Almutiri[4] has proposed an
efficient risk aware virtual resource responsibility mechanism for clouds multitenant environment. In this
paper introduce impressionable in datacenters and minimize the data leakage in high sensitive data center
to low sensitivity data center.
Abdurahman Almutiri[4] had implemented the three assignment heuristics like security aware scheduling
,access Control in cloud data center and vulnerability models. In organization , define the permission for
Role based access control model (RBAC) policy. The policy like a set of Roles( R), set of user(U) and set
of permission (P). this work does not detailed knowledge in data leakage in sensitive data center.
C.Bandwidth
Resource sharing in cloud computing over the web is based on bandwidth sharing. In bandwidth allocation
among tenants based on different requirements like network proportionality, min-guarantee and high
utilization. Network sharing policy to achieve both min-guarantee and proportionality, while prevent
tenants earning unfair bandwidth. Bandwidth sharing based on the analysis pricing model, foreign link
transmission , bandwidth allocation enhancement and traffic flow arrangement policy. Haiyingshen
proposed [5] bandwidth sharing and pricing policies to transform the competitive environment to a winwin cooperative environment for tenants strive to increase the utilities of cloud provider. Haiyingshen
further will consider rewarding tenants for reducing demand to maintain the uncongested link states.
D .Efficient
Efficient resource sharing method based on Distributed Hash Tables(DHT) maintain for number of
resource information service systems have been put forth based on that offer scalable key-based lookup
functions. Effective resource sharing based on three challenges. The first challenge it achieved high
efficiency in an environment characterized by dynamically, geographically in large-scale scattered
resource. The second challenge is resource sharing with high fidelity. Fidelity defined by the ability to
locate all resource in the system that satisfy a resource request. The third challenge is flexibility it defined
by the ability to allow nodes to specify unlimited expensiveness with desired resource and similar resource
searching based on exact matching searching.
Haiyingshen[6] proposed the resource discovery mechanism , which all attributes into a set of indices
using locality sensitive hashing(LSH) with DHT and this paper shows an LSH-based resource information
service(LIS) that combine with efficiency, fidelity and flexibility. Alpha LIS and Hilbert-LIS provide
attribute likeness search that can attribute with similar character. Haiyingshen will further study the
effectiveness with Alpha LIS and Hilbert-LIS in such feature similarity search and their side effects
caused by the alphanumeric transformation and develop an effective and robust resource information
service.
E.Pricing
Resource pricing in one the key parameter in resource sharing in cloud computing. BhanuSharma[7]
proposed the cloud compute community(C3)pricing architecture called clabacus(cloud abacus) with
moores law, that captures the technological advances of the resources. Mapping C 3and contribution with
three major parallels between financial option and C3.The first major mapping is cloud resource pricing to
well established financial option pricing models and apply them to compute prices for the cloud resources.
Second mapping is estimating the risk associated with investment and its requires intricate mathematical
formulation. The third mapping techniques is to use the principles of collateral service agreements (CSA).
Pricing algorithm based on five parameter pertinent to pricing cloud resource like capital investment,
contract time, rate of depreciation, quality of service, age of resource. Value at risk(VaR) is based on two
different techniques like fuzzy logic and genetic algorithm. BhanuSharma[7] will next step is blended
effect of the parameter with resource pricing and multi objective price optimization.
F.Social cloud
Social cloud network platform have rapidly changed the way that people communicate and interact. It
enabled the establishment of participation in digital communities as well as representation, documentation
and explore the relationship. Following challenges proposed in social cloud like technical facilitation to
enable edge users to provide resource to, and consume resource from , one another it need to traverse
network address translation , handle non static IP address with QoS.
Simon caton[8] was implemented & proposed different types of algorithm like polynomial time algorithm,
deferred acceptance(DA) algorithm, welfare optimal (WO) algorithm, heuristic algorithm(HA), genetic
algorithm(GA), GATA. Simon caton[8] will include additional way for users to provide their preferences,
as well as method to detect them automatically from their social network.
G.Hybrid And Heterogeneous Cloud
Resource sharing in hybrid and heterogeneous cloud is a one primary parameter. In hybrid cloud, sharing
the resource in secured & authorized manner it was reduced the duplication using different types of
encryption and decryption algorithm. Jin Li[8] was implemented in symmetric encryption, convergent
encryption in private cloud and public cloud and also achieved the differential authorization and
authorized duplicate check.
Sharing the resource in cloud computing based on Economic and Energy. Manjinder Nir[9] was proposed
computation offloading by using the task scheduler based on centralized broker node and it also produced
the optimal solution. In resource augmentation providing the paying for resources, squeak of resources
and accessibility of resources.
Manjinder Nir[9] will also extend the work in scheduler model to consider network congestion based on
task
priority
and
task
execution
redundancy
while
scheduling
task
offloading.
V. CONCLUSION
Cloud computing expertise is more and more used in business markets. In cloud computing paradigm, an
effective resource sharing strategy is required for achieving people satisfaction and higher the profit for
cloud service providers. This paper summarizes the dissimilar factors of RSC and its impacts in cloud
system. Some of the ways are discussed above mainly focus on bandwidth, memory, pricing, economic
&energy resources but are lacking in some determinant. Hence this survey paper will hopefully motivate
future researchers to come up with smarter , protected
algorithms and
Networks.Simon Caton, Member, IEEE, Christian Haas, Member, IEEE, Kyle Chard, Member,
IEEE,Kris Bubendorfer, Member, IEEE, and Omer F. Rana, Member, IEEE, 2014
[11]. Resource Sharing In Heterogeneous Cloud Radio Access Networks Marcelo Antonio Marotta,
Nicholas Kaminski, Ismael Gomez-Miguelez, Lisandro Zambenedetti Granville, Juergen Rochol,
Luiz Dasilva, And Cristiano Bonato Both-2015
[12]. CRESS: A Platform of Infrastructure Resource Sharing for Educational Cloud Computing
[13]. A Fair and Efficient Resource Allocation Scheme for Multi-Server Distributed Systems and
Networks PeymanTeymoori, KhosrowSohraby, Senior Member, IEEE, and Kiseon Kim, Senior
Member, IEEE -2015
[14]. Auction-Based Resource Allocation for Sharing Cloudlets in Mobile Cloud ComputingA-
Long Jin, Wei Song, Senior Member, IEEE, and WeihuaZhuang, Fellow, IEEE-2015
[15]. Efficient Modeling and Demand Allocation for Differentiated Cloud Virtual-Network as-a
Service Offerings BassemWanis, Student Member, IEEE, Nancy Samaan, Member, IEEE, and
Ahmed Karmouch, Member, IEEE-2015
[16]. Performance and Energy Efficiency Metrics for Communication Systems of Cloud Computing
ManjinderNir, Ashraf Matrawy, Senior Member, IEEE, and Marc St-Hilaire, Senior Member,
IEEE-2015
[20]. PriDyn: Enabling Differentiated I/O Services in Cloud Using Dynamic Priorities Nitisha Jain,
IEEE, Xumin Huang, Jiawen Kang, Jiefei Ding, SabitaMaharjan, Member, IEEE, Stein Gjessing,
Member, IEEE, and Yan Zhang, Senior Member, IEEE-2015
[22]. Decentralized and Optimal Resource Cooperation in Geo-distributed Mobile Cloud Computing
Rong Yu, Member, IEEE, Jiefei Ding, Student Member, IEEE, SabitaMaharjan, Member,
IEEE,SteinGjessing, Senior Member, IEEE, Yan Zhang, Senior Member, IEEE, Danny H.K.
Tsang, Fellow, IEEE -2015
[23]. Fair Network Bandwidth Allocation in IaaS Datacenters via a Cooperative Game Approach
JianGuo, Fangming Liu, Member, IEEE, ACM, John C. S. Lui, Fellow, IEEE, ACM, andHai Jin,
Senior Member, IEEE, Member, ACM-2015
[24]. CPU Provisioning Algorithms for Service Differentiation in Cloud-Based Environments