Docstoc

IJETTCS-2013-10-25-081.pdf

Document Sample
IJETTCS-2013-10-25-081.pdf Powered By Docstoc
					    International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)
       Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com
Volume 2, Issue 5, September – October 2013                                    ISSN 2278-6856

            Cloud Computing: Research Activities and
                         Challenges
                         Chetan M Bulla1, Satish S Bhojannavar2 and Vishal M Danawade3
                                                           1
                                                       Dept. of CSE,
                                            KLE’s KLECET, Chikkodi, Karnataka, India
                                                           2,3
                                                         Dept. of ISE,
                                            KLE’s KLECET, Chikkodi, Karnataka, India
Abstract: Cloud computing is a new computing paradigm,            physical server to another. On demand self services: The
which was started in 2007 and it is one of the hottest topic in   computer services such as email, applications, network or
field of IT industry. Cloud computing is based on several         server service can be provided without requiring human
other computing research areas such as SOA, virtualization,       interaction with each service provider.
utility computing and grid computing. Cloud computing is a        Broad network access: Cloud capabilities are available
type of computing that relies on sharing computing resources      over the network and accessed through standard
like memory, processors, database and applications. Cloud
                                                                  mechanisms that promote use by heterogeneous thin or
computing becoming popular because its ease of use and
providing variety of services. In this paper we will survey
                                                                  thick client platforms such as mobile phones, laptops and
research activities and challenges in cloud computing             PDAs.
environment.                                                      Resource pooling: The provider’s computing resources
Keywords: cloud computing, research activities, research          are pooled together to serve multiple consumers using
challenges, and virtualization                                    multiple-tenant model, with different physical and virtual
                                                                  resources dynamically assigned and reassigned according
1. INTRODUCTION                                                   to consumer demand. The resources include among others
Cloud computing is a type of computing that relies on             storage, processing, memory, network bandwidth, virtual
sharing computing resources like memory, processors               machines and email services. The pooling together of the
and applications. Cloud computing is “Internet based              resource builds economies of scale.
computing”, where storage, processing and application
                                                                  Rapid elasticity: Cloud services can be rapidly and
are delivered to requested user or organization through
                                                                  elastically provisioned, in some cases automatically, to
the Internet. Instead of keeping data on own hard drive or
                                                                  quickly scale out and rapidly released to quickly scale in
updating applications for the needs, we use a service over
                                                                  [13].
the Internet, to store information or use its applications
[1].                                                              Pay-as-you-go: Cloud computing resource usage can be
                                                                  measured,     controlled,   and     reported    providing
Cloud computing entrusts remote services with a user's
                                                                  transparency for both the provider and consumer of the
data, software and computation [3] [5]. End users can
                                                                  utilized service [3]. Cloud computing services use a
access cloud-based applications through a web browser or
                                                                  metering capability which enables to control and optimize
a light-weight desktop or mobile app while the software
                                                                  resource use.
and user data are stored on servers at remote locations.
The remainder of this paper is organized as follows.              Reduced Cost: a report by the Brookings Institution finds
Section 2 describes the characteristics of Cloud                  government agencies can save 25% to 50% of their IT
computing, Section 3, the types of Clouds, Section 4, the         costs and increase their business agility by migrating IT
service models and challenges and Section 5 describes             infrastructure to cloud services.
research activities and challenges in cloud environment.

2. Characteristics
Agility: Agility means the ability to respond faster to
changing business needs with IT services and support.
Cloud computing provides services faster and secure [6].
Agility improves with users' ability to re-provision
technological infrastructure resources.
Virtualization: It is an emerging IT paradigm that
separates functions and technology implementations from
physical hardware. Virtualization allows servers and
storage devices to be shared and utilization be increased
[2] [8]. Applications can be easily migrated from one                     Figure 1 Cloud computing Environment
Volume 2, Issue 5 September – October 2013                                                                       Page 206
   International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)
       Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com
Volume 2, Issue 5, September – October 2013                                    ISSN 2278-6856

Multi Tenacity: It refers to the need for policy-driven        with other organizations, whether managed internally or
enforcement, segmentation, isolation, governance, service      by a third-party. It can be hosted internally or externally
levels, and chargeback/billing models for different            [28]. There are two variations of private clouds:
consumer constituencies. Consumers might utilize a
public cloud provider’s service offerings or actually be        On-Premise Private Cloud: This type of cloud is hosted
from the same organization, such as different business           within an organizations own facility. These Clouds are
units rather than distinct organizational entities, but          best used for applications that require complete control
would still share infrastructure [14].                           and configurability of the infrastructure and security.
                                                                Externally Hosted Private Cloud: Externally hosted
3. TYPES OF CLOUDS                                               private clouds are exclusively used by one organization,
With cloud computing technology, large pools of                  but are hosted by a third party specializing in cloud
resources can be connected through private or public             infrastructure. The service provider facilitates an
networks [29]. Businesses can choose to deploy                   exclusive cloud environment with full guarantee of
applications either on Public, Private, Community or             privacy. A public cloud is the obvious choice when:
Hybrid clouds.                                                  You need data sovereignty but want cloud efficiencies.
                                                                You want consistency across services.
3.1 Public Clouds                                               You have more server capacity than your organization
Public clouds are made available to the general public by        can use.
a service provider who hosts the cloud infrastructure.
                                                                Your data center must become more efficient.
Amazon AWS, Microsoft and Google are public cloud
                                                                You want to provide private cloud services.
provider. They operate the infrastructure and offer access
over the Internet. With this model, customers have no          3.3 Community Cloud
visibility or control over where the infrastructure is         Community cloud is a is a multi-tenant cloud service
located. It is important to note that all customers on         model that is shared among several or organizations and
public clouds share the same infrastructure pool with          that is governed, managed and secured commonly by all
limited configuration, security protections and availability   the participating organizations or a third party managed
variances.                                                     service provider. The goal of community clouds is to have
                                                               participating organizations realize the benefits of a public
Public cloud customers benefit from economies of scale,
                                                               cloud with the added level of privacy, security, and policy
because infrastructure costs are spread across all users,
                                                               compliance usually associated with a private cloud. Here
allowing each individual client to operate on a low-cost,
                                                               are a couple of situations where a community cloud
“pay-as-you-go model. Another advantage of public cloud
                                                               environment is best:
infrastructures is that they are typically larger in scale
than an in-house enterprise cloud, which provides clients       Government organizations within a state that need to
with seamless, on-demand scalability. A public cloud is          share resources.
the obvious choice when:                                        A private HIPAA (Health Insurance Portability and
                                                                 Accountability Act) compliant cloud for a group of
 Your standardized workload for applications is used by
                                                                 hospitals or clinics Telco community cloud for Telco
  lots of people, such as e-mail.
                                                                 doctors to meet specific FCC regulations Cloud
                                                                 computing is about shared IT infrastructure or the
                                                                 outsourcing of a company's technology.
                                                                It is essential to examine your current IT infrastructure,
                                                                 usage and needs to determine which type of cloud
                                                                 computing can help you best achieve your goals.
                                                                Simply, the cloud is not one concrete term, but rather a
                                                                 metaphor for a global network and how to best utilize
                Figure 2 Types of Clouds                         its advantages depends on your individual cloud focus.

 You need to test and develop application code.               3.4 Hybrid Clouds
 You need incremental capacity (the ability to add            Hybrid clouds are a composition of two or more clouds
  compute resources for peak times).                           (private, community or public) that remain unique entities
                                                               but are bound together offering the advantages of multiple
 You are doing collaboration projects.
                                                               deployment models. Augmenting a traditional private
3.2 Private Clouds                                             cloud with the resources of a public cloud can be used to
Private cloud is cloud infrastructure dedicated to a           manage any unexpected surges in workload. Hybrid cloud
particular organization. Private clouds allow businesses to    architecture requires both on-premise resources and off-
host applications in the cloud, while addressing concerns      site server based cloud infrastructure [28]. By spreading
regarding data security and control, which is often            things out over a hybrid cloud, you keep each aspect of
lacking in a public cloud environment. It is not shared

Volume 2, Issue 5 September – October 2013                                                                      Page 207
   International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)
       Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com
Volume 2, Issue 5, September – October 2013                                    ISSN 2278-6856

your business in the most efficient environment possible.   forecasting and automated business processes where real
A public cloud is the obvious choice when:                  time data and functionality sharing is needed.
                                                            Some SaaS providers have responded to the integration
 Your company wants to use a SaaS application but is       challenge by developing application programming
  concerned about security.                                 interfaces (APIs). Unfortunately, accessing and managing
 Your company offers services that are tailored for        data via an API requires coding and maintenance due to
  different vertical markets.                               API modifications and updates. API have also some
 You can use a public cloud to interact with the clients   limitations.
  but keep their data secured within a private cloud.       Another challenge is related to a data locality. In a SaaS
 You can provide public cloud to your customers while      consumers use the applications to process their business
  using a private cloud for internal IT.                    data. The problem is that the customer does not know
                                                            where the data is getting stored. Due to compliance and
4. SERVICE MODELS                                           data privacy laws in various countries, locality of data is
4.1 Software as a Service (SaaS)
                                                            very important in part of the enterprise architecture.
The most known and the leading service model of more
widespread adoption of cloud computing has been the         4.2 Platform as a Service (PaaS)
SaaS. In SaaS model a software provider licenses a          The difference between SaaS and PaaS is that SaaS only
software application to be used and purchased on            hosts completed cloud applications where PaaS offers a
demand. Applications can be accessed through networks       development platform for both completed and in-progress
from various clients (web browser, mobile phone, etc.) by   cloud applications [28].
application users. The application requires no client       PaaS offers an environment where developers can create
installation, just a browser or other client device and     and deploy applications and do not necessarily need to
network connectivity. SaaS offerings can be classified by   know how much memory or how many processors their
the software and the pricing model. Table 1 gives an        application will be using. In addition multiple
overview of some SaaS providers.                            programming models and specialized services (data
The most common pricing model is pay per use, which a       access, authentication, etc) can be offered as a building
customer pays a static price for units they use. Another    block to new applications [29]. PaaS model provides
pricing model is subscription, which a customer             developers a service that can be used to a complete
subscribes to use a preselected combination of service      software development lifecycle management, from
units for a fixed price and a longer time frame, monthly    planning to design to building applications to deployment
or yearly.                                                  to testing to maintenance [30]. Table 2 summarizes
          Table 1. Popular SaaS Provider.                   popular PaaS service provider.
  Provider         Software              Pricing                        Table 2. Popular PaaS providers
                                         model
  Salesforce       CRM                   Pay per use
  Google Gmail     Email                 Free
  Process          Business process      Pay per use
  Maker Live       management
  XDrive           Storage               Subscription
  OpSource         Billing               Subscription
  Appian           Business process      Pay per use
  Anywhere         management
  Box.net          Storage               Pay per use
  MuxCLoud         Data processing       Pay per use


4.1.1 Challenges
Like any new technology, SaaS model suffers some of
limitations. One of the major challenges with SaaS
                                                            4.2.1 Challenges
applications is integration [11]. SaaS applications
                                                            Cloud data stores provide scalability and high availability
typically provide services for one business area like
                                                            properties for web applications but do not support
Enterprise resource planning (ERP). As a result
                                                            complex queries such as joins. Developers must therefore
companies are facing serious problems with accurate data,

Volume 2, Issue 5 September – October 2013                                                                  Page 208
   International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)
       Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com
Volume 2, Issue 5, September – October 2013                                    ISSN 2278-6856

design their programs according to the peculiarities of       Different types of user interfaces provide different levels
NoSQL data stores rather than established software            of abstraction, the most common being graphical user
engineering practice. In PaaS model the provider gives        interfaces (GUI), command-line tools (CLI), and Web
some control to the developers to build applications on       service (WS) APIs [29]. Usually IaaS providers offer
top of the platform. But any security below the application   multiple user interfaces.
level such as host and network intrusion prevention will
still be in the scope of the provider [9]. The provider       5. RESEARCH ACTIVITIES AND
needs to verify that the data remains inaccessible between    CHALLENGES.
applications. PaaS model offers developers a service that     5.1. Virtualization
provides a complete software development lifecycle
                                                              Most discussions on cloud computing typically begin with
management. Everything else is abstracted away from the
                                                              virtualization. Basic idea of a virtualization is that the
eyes of the developers [28].
                                                              resources of one physical computer can be portioned into
                                                              logical resources and rearranged into multiple virtual
4.3 Infrastructure as a Service (IaaS)
                                                              machines [17]. The operating system, network and
The infrastructure layer focuses on enabling technologies
                                                              storage resources can be virtualized to improve the
[8]. IaaS model changes the way developers deploy their
                                                              performance. Virtualization is critical to cloud computing
applications. Instead of spending time with their own data
                                                              because it simplifies the delivery of services by providing
centers or managed hosting companies, they can just
                                                              a platform for optimizing complex IT resources in a
select one of the IaaS provider, get a virtual server
                                                              scalable manner, which makes cloud computing so cost
running in few minutes and pay only for the resources
                                                              effective. Virtualization is emerging IT paradigm that
they use [34]. From a technology viewpoint the IaaS type
                                                              separates      computing      tasks     and      technology
of cloud offerings have been the most successful [40].
                                                              implementation      from     physical     hardware     [12].
                                                              Virtualization has three characteristics that make it ideal
4.3.1 Challenges
                                                              for cloud computing:
Out of the box IaaS usually only provides basic security
(perimeter firewall, load balancing, etc.) and applications   1. Partitioning: Partitioning supports many applications
moving into the cloud will need higher levels of security        and operating systems (OS’s) in a single physical
provided at the host [29].                                       system.
Security responsibilities of both the provider and the        2. Isolation: Each machine is protected from crashes and
consumer differ between different service providers.             viruses in the other machines.
Amazon’s Elastic Compute Cloud (EC2) includes vendor          3. Encapsulation: Protect each application so that it
responsibility for security up to the hypervisor, meaning        doesn’t interfere with other applications. Using
they can only address security controls such as physical         encapsulation, a virtual machine can be represented as
security, environmental security, and virtualization             a single file, making it easy to identify and present to
security. The consumer is responsible for the security           other applications.
controls that relate to the IT system including the OS,
applications and data [32].                                   5.1.1 Research activities
To be able to improve availability and responsiveness, an     Virtualization is the most emerging technology of this
IaaS provider builds several data centers distributed         decade in which the advancement is taking place
around the world. For example, Amazon Web Services            exponentially. Most of the organizations are now aware of
calls these data centers as availability zones and regions.   the advantages of adopting virtualization and thus are
Currently Amazon is having availability zones in US and       deploying virtualization solutions in order to reduce the
in Europe.                                                    downtime and to make efficient utilization of power [33].
      Table 3: Feature comparison of IaaS providers           Virtualization technology usage is increasing day by day
                                                              and so is the risk associated with it. There are many
                                                              security loop holes in this technology which if
                                                              compromised, can prove to be very expensive in future.
                                                              5.1.2 Challenges
                                                              There are some challenges facing by cloud computing:
                                                              1. Depleted resources: Often created by a move from
                                                                 physical hardware to virtual hardware and VM
                                                                 saturation cause application networking resources to be
                                                                 depleted at a much faster rate.
                                                                I/O intensive operations get bogged down in the
                                                                  virtualization translation layer.


Volume 2, Issue 5 September – October 2013                                                                     Page 209
   International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)
       Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com
Volume 2, Issue 5, September – October 2013                                    ISSN 2278-6856

  Saturation of the network card and software switch on         connection, such as NFS, can cause instant bottlenecks,
    the physical host causes performance issues, reduced         flooding and congestion.
    bandwidth, and increased latency.                          • Moving large virtual disk images outside the LAN
                                                                  causes extensive delays.
2. Lack of awareness: Virtual infrastructure platforms         • Unplanned VM migrations resulting from VM sprawl
   include software that can migrate running VM                   bring the network to a standstill.
   instances from one physical device, which often causes
   a lapse in application availability [32].                  6. Management complexity: Managing VMs as part of the
 • When VMware’s Distributed Resource Scheduler                  complete management solution can be a struggle. This
    (DRS) moves an image to a different physical box, the        includes managing the VMs themselves as well as
    app is unavailable during the migration, interrupting        managing all parts of the data center as one delivery
    the customer’s persistence as well as shopping cart          unit.
    transactions.                                              • The hypervisor and the host system are two new
 • Bringing up new virtual images is easy, and so is              components that are not part of existing data center
    bypassing typical work flows and configurations as a            management solutions. It is important to be able to
    result. As VM’s proliferate, IP addresses are lost,           manage these devices and understand their impact on
    images are assigned to the wrong VLAN and                     performance.
    availability is negatively impacted.                       • Built-in management tools for VM platforms only
                                                                  manage the virtual resources and do not take into
3. Additional, unanticipated costs: Additional costs often        account any external information
   result from implementing OS virtualization. New
                                                              5.2. Cloud Aggregation
   hardware and software licenses can be required to solve
                                                              Two key technological challenges for cloud are,
   problems with availability, performance, and
                                                              supporting the expansion of the distributed services across
   management.
                                                              different IaaS providers and enabling the portability of
 • As VMs burden existing infrastructure, requirements
                                                              such services. These challenges require research on
    grow for application and storage networks.
                                                              developing novel architectural architectures and
 • Management of new tools requires increased headcount
                                                              brokering mechanisms for high availability, improving
    and training.
                                                              performance, effective resource utilization, reduce price
 • VM sprawl creates orphaned virtual disk images that
                                                              and defining standards for cloud interoperability and
    take up critical storage space
                                                              portability [15].
4. Non-virtualization: New virtual platforms include          Current IaaS cloud technologies support two main
   many advanced networking technologies, such as             aggregation architectures: cloud-bursting and cloud-
   software switching and support for VLAN                    brokering models. The cloud-bursting architecture (also
   segmentation. However these features are localized and     known as the hybrid cloud), combines existing, on-
   isolated to the VM platforms. They are not integrated      premise cloud infrastructures with remote resources from
   with the rest of the Application Delivery Network          one or more public clouds to provide extra capacity and
   (ADN) because there is no sharing of information           satisfy peak demand periods.
   between VMware and the network.
 • Storage integration issues arise as soon as VMs are        5.2.1 Research Activities
    moved into production as additional storage
    virtualization tools are needed to migrate live machine     One open research line is to develop novel aggregation
    outside the local storage domain.                         architectures that allow a higher degree of coupling in
 • Overrun storage network: OS and data files that             which a provider could :
    typically reside on internal storage in physical server    Have advanced control over remote resources.
    environments are moved to share storage in virtual         Access detailed monitoring information about remote
    environments. OS drives are converted to flat-file            resources; and
    virtual machine disks (VMDKs), which take up 10-           Implement some advanced cross-site features.
    100s of GBs each of networked storage.
 • Storing “gold” virtual images to be cloned as needed       5.2.2 Challenges
    and copied to another part of the storage network         1. Service Portability between different cloud service
    creates a proliferation of rarely used “parked” images       providers involves several challenges due to a dearth of
    on shared storage, adding unnecessary costs.                 standard interfaces for interacting with different clouds
                                                                 and standard interfaces and formats for managing
5. Congested storage network: Data pipes can’t handle the        virtual appliances [24]. Currently, no standard interface
   volume OS virtualization can dramatically increase            exists for interfacing with a cloud, and each provider
   data storage traffic, and passing large amounts of data       exposes its own APIs.
   from multiple guests through one host storage network

Volume 2, Issue 5 September – October 2013                                                                     Page 210
   International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)
       Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com
Volume 2, Issue 5, September – October 2013                                    ISSN 2278-6856

2. Clearly, cloud providers must unify their different cloud   2. The improper and unprofessional working force makes
   APIs and adopt a standardized cloud interface. In this         difficulties in cloud. Advanced equipments comprising
   context, several efforts are under way, such as                of software and hardware tools might be good sources
        1. The Open Cloud Computing Interface (OCCI),             for managing cloud of an organization or migrating it,
           which defines a protocol and API specification         but lack of knowledge about them and insufficient skill
           for remotely managing cloud computing                  set for their implementation and installation may prove
           infrastructures;                                       all the efforts of cloud providers in vain.
        2. The Cloud Infrastructure Management Interface       3. Moreover, if data migration is not done systematically
           (CIMI), a work-in-progress standard that defines       and properly, it can give rise to problems concerning
           an interface and a logical model for managing          data and cloud security of company’s assets that
           resources within an IaaS cloud.                        primarily comprise of data. Thus, hiring cloud
3. Important snag for users who is moving their services          providers having sound experience about the field with
   from one cloud to another is a lack of compatibility in        ample knowledge and skill sets becomes vital for
   image-packaging formats and image-management                   managing cloud more effectively and efficiently.
   interfaces. Some standardization efforts also exist in      4. Governing confidential data is not a simple task.
   this area, such as the Open Virtualization Format              Reputed managed cloud providers do it with utmost
   (OVF), which defines a standard for packaging and              care and priority. However, despite equipped with high-
   distributing virtual appliances.                               end tools for installing and incorporating data the cloud
4. Another obstacle for cloud service provider is to share        management professionals also need to take care that
   the resources. There is no standardization for resource        whatever data they are storing in the cloud is well
   sharing in cloud. A lot of research work can be taken to       protected and safeguarded from outside thefts and
   develop standardization for sharing resources between          fallacies. As Internet is housing ground for criminal
   clouds.                                                        acts, services such as cloud management needs special
                                                                  attention in order to remain protected.
5.3 Cloud Migration
The process of transitioning all or part of a company’s        5.4. Cloud Elasticity
data, applications and services from on-site premises          Elasticity in the cloud is the ability of an application to
behind the firewall to the cloud, where the information        automatically adjust the infrastructure resources it uses to
can be provided over the Internet on an on-demand basis        accommodate varied workloads and priorities, while
[8]. While a cloud migration can present numerous              maintaining availability and performance in a context-
challenges and raise security concerns, cloud computing        aware environment [15]. There are other important
can also enable a company to potentially reduce capital        scalable entities in the cloud such as networks, databases
expenditures and operating costs while also benefiting         and important middleware and monitoring tools.
from the dynamic scaling, high availability, multi-
tenancy and effective resource allocation advantages           The main obstacle to utilizing elasticity is that most
cloud-based computing offers.                                  enterprise-application architectures are not designed to
                                                               take advantage of the elastic infrastructure resources
5.3.1 Research Activities                                      currently inherent to cloud IaaS platforms [31].
The research focuses for instance on an advanced model-        5.4.1 Research activities
driven methodology and tools for reuse and migration of
on-premises applications to cloud [24]. To support the         Cloud infrastructures employ virtual infrastructure (VI)
migration, e.g. the REMICS project (EU FP7) [41]               management to dynamically orchestrate the deployment
enhances the Object Management Group’s (OMG)                   of virtual machines, management of storage
Architecture Driven Modernization (ADM) methodology            requirements, and to configure resources to adapt to an
with specific methods, meta-models and tool support,           organization’s changing needs.
including    knowledge      discovery,   patterns  and         5.4.2 Challenges
transformations for SOA and cloud computing.
                                                               Elastic services offer additional management challenges.
5.3.2 Challenges
                                                               1. The components and resources involved in an elastic
1. The cloud managers face extreme problems at the time           service are dynamic and change as demand for the
   of data migration from a company’s server to a server          service changes. This means that the management
   that forms cloud elsewhere. The cloud behaves as an            system must be able to automatically adapt to the
   interface through which organizations can access data          changes in the underlying set of managed resources. It
   in a virtual environment [31]. Thus, smooth                    is also more difficult for a services management system
   functioning of it depends primarily on how well                to ensure predictable performance from the service and
   groomed and knowledgeable cloud providers are in this          to support problem determination since the underlying
   area.                                                          resource configuration changes.

Volume 2, Issue 5 September – October 2013                                                                      Page 211
   International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)
       Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com
Volume 2, Issue 5, September – October 2013                                    ISSN 2278-6856

2. Adding or removing resources, such as a virtual            5.6.2 Challenges
   machine, to a service can involve significant delays.
   The services management system must be able to             Cloud computing security challenges:
   predict when a change in the makeup of the service will    1. Protection: Securing your data both at rest and in
   be necessary with enough lead time to minimize the            transit
   impact of the change.                                      2. Authentication: Limiting access to data and monitoring
5.5 Cloud Interoperability                                       who accesses the data
                                                              3. Access Control and Accounting: access control services
The ability of applications to move from one cloud               should be flexible enough to capture dynamic, context,
environment to the next and for the applications to work         or attribute- or credential-based access requirements
exactly the same in both places Cloud interoperability           and to enforce the principle of least privilege
refers to the ease of migration and integration of            4. Trust Management and Policy Integration: Cloud
applications and data between different providers’ clouds        service providers need to combine multiple services to
[8].                                                             enable biggest application services. Therefore,
                                                                 mechanisms are necessary to ensure that such dynamic
5.5.1 Research activities                                        and huge application collaboration is handled securely
The requirement of interoperability is mostly addressed          and that security breaches are effectively monitored
with regard to specific dimensions, indicating that there        during the interoperation process [27].
is still room for a more holistic approach with the regard    5. Secure-Service Management: The cloud issues such as
to the issue at hand. The solutions focus on providers that      quality of service, price, and SLAs are critical in
use the same technology, denoting that there is a long           service search and composition. These issues must be
way ahead until global interoperability across the cloud is      addressed to describe services and introduce their
achieved.                                                        features, find the best interoperable options, integrate
                                                                 them without violating the service owner’s policies, and
The business value of mashing up applications to develop         ensure that SLAs are satisfied.
interoperable cloud services, tailored to different users     6. Privacy and Data Protection: Cloud service providers
have to be further highlighted and promoted within               must assure their customers and provide a high degree
research endeavors.                                              of transparency into their operations and privacy
                                                                 assurance.
5.5.2 Challenges
                                                              5.7. Resource management
1. How applications and services can be combined, while
   maintaining some level of resiliency with regards to       Resource management involves resource scheduling,
   changes in connectivity and location.                      dynamic resource allocation, resource mapping, resource
2. The APIs that a management application can use to          provisioning and load balancing [34].
   control multiple cloud environments. This explores
   how images are delivered between providers; how            5.7.1 Research Activities
   servers/applications are started or stopped; and how       There is plenty of work undergone on above areas. The
   storage can be manipulated.                                load balancing is achieved through some sort of load
3. A single sign-on for users who access multiple cloud       balancing among the participating nodes. There are some
   platforms, enabling a service hosted on one platform to    metrics that will help to evaluate the efficiency of each
   automatically call a service hosted by another, and        load balancing method. Load balancing techniques in
   having a private cloud application seamlessly obtain       cloud environment, consider various parameters like
   resources from a public cloud when necessary [24].         performance, scalability, response time, throughput,
                                                              resource utilization, fault tolerance, migration time and
5.6. Security and Privacy
Cloud computing set up a area of opportunities for            associated overhead time, throughput, resource
businesses, but mixed in with these opportunities are         utilization, fault tolerance, migration time and associated
numerous security challenges that need to be considered       overhead.
and addressed prior to committing to a cloud computing        5.7.2 Challenges
strategy [9] [10].
                                                              There are lot of work is undergoing with resource
5.6.1 Research activities                                     management, but we are taking main challenges of
Many security and privacy incidents are also observed in      resource management. The main challenges faced by
today’s cloud computing systems. The thousands of             cloud computing are:
research works is undergone and still not getting             1. Dynamic allocation: The resources are allocated
complete solution because its market based and market is        dynamically on the arrival of workloads according to its
always varying. So need good research work is needed.           requirements. It is complex task because dynamic

Volume 2, Issue 5 September – October 2013                                                                    Page 212
   International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)
       Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com
Volume 2, Issue 5, September – October 2013                                    ISSN 2278-6856

  resource allocation involves resource selection,            [10] Cloud Security Alliance. Security guidance for
  mapping of appropriate resources [5].                            critical areas of focus in cloud computing (v2.1).
2. Workload prediction: Predicting the workload is one             Decemeber, 2009.
  of main challenging task in cloud computing because         [11] D. Catteddu and G. Hogben, “Cloud Computing:
  it’s difficult to predict future load of the cloud. To            Benefits, Risks and Recommendations for
  predict the workload, first we classify the workloads             Information Security,” ENISA, 2009.
  into homogeneous, heterogeneous workload and then           [12] Paul Hofmann, Dan Woods, "Cloud Computing:
  apply either statistical analysis or neural network for           The Limits of Public Clouds for Business
  load prediction.                                                  Applications",IEEE,1089-7801/10 © 2010.
                                                              [13] Philip Dawson, Amit Singh, Virtualization Key
                                                                    Initiative                                 Overview,
6. CONCLUSION                                                       http://www.gartner.com/it/initiatives/pdf/KeyInitiati
Cloud computing is a developing computing technology                ve Overview_Virtualization.pdf, 2010.
that uses the Internet and multiple remote servers to         [14] Hugos, M., and Hulitzky, D., Business in the cloud:
maintain data and software applications. We have                    what every business needs to know about cloud
identified the characteristics, service models along with            computing, Wiley, 2010.
the challenges. We have also identified            research   [15] Dillon T., Chen W., and Chang, E., Cloud
activities related to various technologies used in cloud            Computing: Issues and Challenges, Proc 24th IEEE
computing environment. According to the challenges                  International Conference on Advanced Information
noted above, lot of research work can do in different areas         Networking and Applications (AINA), Perth,
to overcome these challenges and potential users need to            Australia, April 2010.
evaluate QoS and security measures such as firewalls, and     [16] Subashini S., and Kavitha V., A survey on security
encryption technique and make sure that they will have              issues in service delivery models of cloud
access to data and software.                                        computing. Journal of Network and Computer
References                                                          Applications, July 2010.
[1] Broberg J., Buyya, R., and Goscinski A., Cloud            [17] Anandasivam A., Blau B., Stosser, J., and
    Computing: Principles and Paradigms, Wiley Press,               Weinhardt C., Business Models in the Service
    USA, 2011.                                                      World, IT Professional, September/October,2010.
                                                              [18] Jinzhu Kong, “A Practical Approach to Improve the
[2] Mark F. Mergen, Volkmar Uhlig, Orran Krieger,
                                                                    Data Privacy of Virtual Machines” 2010 IEEE 10th
    Jimi Xenidis, "Virtualization for High-Performance
                                                                    International Conference on Computer and
    Computing", IBM T. J. Watson Research Center
                                                                    Information Technology (CIT), pp. 936-941 June 29
    Yorktown Heights, 2007.
                                                                    -July 1 ,2010.
[3] Greg Goth, "Software-as-a-Service: The Spark That
    Will Change Software Engineering?” , IEEE, vol. 9,        [19] Pearson, S. and Azzedine Benameur, “Privacy,
    no. 7, 1541-4922 © 2008.                                        Security and Trust Issues Arising from Cloud
                                                                    Computing” in 2010 IEEE Second International
[4] Bechtolsheim, A. “Cloud Computing and Cloud
                                                                    Conference Cloud Computing Technology and
    Networking.” talk at UC Berkeley, December 2008.
                                                                    Science (CloudCom), page(s): 693-702,Nov 30-Dec
[5] N.M. Mosharaf, Kabir Chowdhury, "Identity
                                                                    3,2010
    Management and Resource Allocation in the
                                                              [20] Esteves, R.M. and Chunming Rong, “Social Impact
    Network Virtualization Environment", Waterloo,
                                                                    of Privacy in Cloud Computing” in 2010 IEEE
    Ontario, Canada, 2008.
                                                                    Second International Conference on Cloud
[6] Foster I, Zhao Y, Raicu I, "Cloud Computing and
                                                                    Computing Technology and Science (CloudCom),
    Grid Computing 360-Degree Compared", IEEE,
                                                                    pp. 593-596,N ov. 30-Dec. 3 ,2010.
    2008
                                                              [21] Hassan T, James B.D.Joshi, Gail-Joon, "A Security
[7] RightScale (2008) Define Cloud Computing.
                                                                    and Privacy Challenges in Cloud Computing
    RightScale        Blog,      26      May        2008.
                                                                    Environments", COPUBLISHED BY THE IEEE
    http://blog.rightscale.com/2008/05/26/define-cloud-
                                                                    COMPUTER               AND            RELIABILITY
    computing/,2008
                                                                    SOCIETIES,1540-7993/10/$26.00 © 2010.
[8] Amit Goyal and Sara D., "A Survey on Cloud
                                                              [22] Grance T., and Mell P., NIST definition of cloud
    Computing" University of British Columbia,
                                                                    computing, National Institute of Standards and
    Technical Report for CS 508, December 2009.
                                                                    Technology, January 2011.
[9] N. M. Mosharaf Kabir Chowdhury and Raouf
                                                              [23] L.Arockiam,       S.Monikandan,      G.Parthasarathy,
    Boutaba, "Network Virtualization: State of the Art
                                                                    "Cloud Computing: A Survey",(IJIC), ISSN No:
    and Research Challenges", IEEE, 0163-6804 ©
                                                                    2231 – 6965, Volume-1, Issue-2, 2011.
    2009.
                                                              [24] V.Krishna Reddy, B.Thirumala Rao,Dr.L.S.S.
                                                                    Reddy, P. Sai Kiran, "Research Issues in Cloud

Volume 2, Issue 5 September – October 2013                                                                    Page 213
   International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)
       Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com
Volume 2, Issue 5, September – October 2013                                    ISSN 2278-6856

       Computing", GJCST, Volume 11, Issue 11, Version
       1.0, July 2011.
[25]   Patrick Martin, Andrew Brown, Wendy Powley,
       Jose      Luis      Vazquez-Poletti,   "Autonomic
       Management of Elastic Services in the Cloud",
       IEEE, 978-1-4577-0681-3/11/$26.00 ©2011.
[26]   Google AppEngine, http://code.google.com/intl/fi-
       FI/appengine/. [10.2.2012].
[27]   Pradeep Kumar Tiwari1, Dr. Bharat Mishra2,
       "Cloud Computing Security Issues, Challenges and
       Solution",ijetae, ISSN 2250-2459, Volume 2, Issue
       8, August 2012.
[28]   Eeva Savolainen, "Cloud Service Models" Seminar
       – Cloud Computing and Web Services,
       UNIVERSITY OF HELSINKI, Department of
       Computer Science, Helsinki 10.2.1012.
[29]   Amazon Web Services, http://aws.amazon.com/.
       [9.2.2012].
[30]   Claus Pahl, Huanhuan Xiong, Ray Walshe, "A
       Comparison of On-premise to Cloud Migration
       Approaches", IJSCT, 2 (2), 278-286,2012.
[31]   Amazon. Amazon Elastic Compute Cloud (Amazon
       EC2).http://aws.amazon.com/ec2/.
[32]   "Advanced Network Virtualization: Definition,
       Benefits, Applications, and Technical Challenges",
       Network Virtualization Study Group (NVSG) White
       Paper,v1.0, 2012.
[33]   Patidar, S. Rane, D, Jain, P,"A Survey Paper on
       Cloud Computing", Advanced Computing &
       Communication Technologies (ACCT), 2012
       Second International Conference, IEEE, P.394 -
       398, 2012
[34]   Alexandre Kandalintsev, Trento, Italy, "Advanced
       Cloud Resource Management For Performance and
       Power Saving",2012.
[35]   Jyotiprakash Sahoo, Subasish Mohapatra, Radha
       Lath, 2010, Virtualization: A Survey On Concepts,
       Taxonomy And Associated Security Issues, 2nd
       International Conference on Computer and Network
       Technology, IEEE,2012.
[36]   Understanding       Cloud     Security Challenges,
       cognizant 20-20 insights , nov 2012.
[37]   Pooja Kedia, Renuka Nagpal, Tejinder Pal Singh,
       "A Survey on Virtualization Service Providers,
       Security Issues, Tools and Future Trends",
       IJCA(0975 – 8887),Volume 69– No.24, May 2013
[38]   Vikas Kumar, "Survey Paper on Cloud Computing",
       IJEAT, ISSN: 2249 – 8958, Volume-2, Issue-6,
       August 2013.
[39]   Arto Ojala, "Software-as-a-Service Revenue
       Models",IEEE,1520-9202/13/$31.00 © 2013.
[40]   K. Rasmi and V. Vivek, "Resource Management
       Techniques in Cloud Environment - A Brief
       Survey", IJIAS, ISSN 2028-9324 Vol. 2 No. 4 Apr.
       2013, pp. 525-532, © 2013.
[41]   REMICS Project:
       http://cordis.europa.eu/fp7/ict/ssai.

Volume 2, Issue 5 September – October 2013                                          Page 214

				
DOCUMENT INFO
Description: International Journal of Emerging Trends & Technology in Computer Science (IJETTCS) Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com Volume 2, Issue 5, September – October 2013 ISSN 2278-6856, Impact Factor 2.524 ISRA:JIF