Multitenancy in Cloud Computing

This report discusses the concept of multitenancy in cloud computing, including its benefits, challenges, and possible solutions. It emphasizes the importance of data isolation and security in a multi-tenant environment. The report also explores different models of multitenancy and their implications.

Multitenancy in cloud computing
Authors Name[] Authors Name[]
Abstract— With the changing computing paradigm and
emergence of cloud computing, the concerns related to this
are also increasing. Most important risks related to the
multitenancy is the threat to the confidentiality of the data
of the clients. The following report, contributes to the
different aspects of the multitenancy, benefits, challenges
and different possible mitigation technique for those
issues/challenges.
Keywords: Militancy, cloud services, architecture,
deployment.
I. INTRODUCTION (HEADING 1)
With the improved availability of internet, the cloud
computing is trending to be the most favorable computational
model for the organizations as well as individual users [5]. In
this cloud computing architecture one of the most important
aspect is the multi-tenancy [10]. The Multi-Tenancy is often
considered as one of the benefits to the Cloud service
providers as well as to the clients using the services.
Nevertheless, this benefits comes along with associated
security risks. When security is considered then a natural
proposal is the elimination of risk [8]. Which leads to the
proposal of eliminating the virtualization layer in order to
increase system security of the multitenant cloud. However,
cost to make this kind of change in the existing systems is
considered to be very high. In this architecture, the resources
sharing among the different consumers inside clouds (whether
private or public) is known as the multi-tenancy. According to
the standard, each tenant's data inside a cloud needs to be
isolated and should not be visible to the other tenants in the
same cloud.
In the following report, architecture of the multi
tenancy, challenges in the multi tenancy and solutions are
discussed in different segments. In addition to that, the
security and isolation aspects are also discussed.
II. ARCHITECTURE OF THE MULTITENANCY IN CLOUD
In this architecture, multiple number of
tenants/consumers shares same application which is executing
on same OS, hardware as well as same data-storage
mechanism in them. The division among the customers who
are using the application is accomplished at the time of the
application design. In this way it becomes difficult for the
customers to share or view other user’s data [2]. In any
Multitenancy based cloud service there are multiple models
are available for the users/ clients to choose from which
includes shared hardware, shared nothing, shared database,

shared OS, shared everything as well as custom sharing based
multitenancy.
For Shared Hardware multitenancy, it is used in
order to create backward compatibility for the legacy
applications of the organizations through the use of the Virtual
Machines.
On the other hand, for shared everything
Multitenancy environment this is important to for quick
deployment and off-ramping of the applications used by the
tenant’s such as individual users as well as enterprises.
III. LITERATURE REVIEW
The computing resources as well as the application
codes are shared among the users on the same server. In case
of the data, every tenant on the server has their own set of data
which is logically isolated from the other tenant’s data. In
case SaaS or Software as a Service in cloud architecture
applications are delivered as service by the Provider [6]. In
this scenario, the customer is not able to control or monitor the
underlying infrastructure running under the application which
they are using. Here, Multi-Tenancy implies that two or more
than that customers have the ability to use the application at a
given point of time without having any knowledge about the
underlying resource.
Different scholars or experts has their different
opinion about the efficiency of the Multitenancy [4].
According to developers, it is an opportunity and helpful in
order to get the required infrastructure without actually
acquiring them and just by paying according to the usage. On
the other hand, according to the security experts, this sharing
environment is very vulnerable to the attacks that can lead
to the exposed confidentiality of the consumer data [11]. For
the service providers, share of the resource leads to the
enhanced utilization of the resources that helps in reduction in
operational cost while making resources available for the
multiple users.
IV. CHALLENGES IN MULTITENANT PLATFORM
Identified from the user’s perspective the main issues
are Performance isolation, and inconstancy for all the
resources, Scalability with the quantity of tenants, Support for
esteem included operations for every application type and
Security for information. Development engineers confronting
the principle issues are get to control, customizability [8].
Customizability includes in the database, UI, work processes,
tenant provisioning, and usage based metering. Identified with
the specialist organizations focus on information sharing,
backup of tenant information, enablement of multitenancy
with no code changes, scalability, improved usage, operational
costs and required skills [8]. Tenant specific nature of
administrations Multi-tenure help, occupants to tweak their
very own administration in runtime without affecting others in
the same cloud.
As every one of the tenant shares a similar
application example, when the customization is accomplished
for a specific tenant on the cloud, the administrations for all
tenants will be influenced, and potentially interfered with amid

the update. As the number of tenants extends, the interferences
become increasingly regular and lead to intense administration
accessibility issues [11].
Figure 1: Architecture of Multitenant cloud
At the point when issues occur happen or when
certain administrations create unusual burdens the
administration conveyance can be hindered – yet business
customers will frequently demand high-accessibility. Along
these lines, observing the administration conveyance and its
accessibility is basic to guarantee that the administration is
appropriately conveyed.
Side Channel attack to another tenant: a side
channel attack is any in light of data picked up from the
physical execution of a framework [10]. There are many side
direct attacks known in the field; a portion of the well-known
side channel attacks are timing attacks, control utilization
attacks and differential deficiency examination.
Network Probing: This mechanism is helpful in
order to discover the physical topology of any selected
network system that comprises of IPs and servers associated in
the system. Such data can be used to distinguish potential
targets and to structure the attack to the selected target.
Brute Force attacks: This attack technique or
instrument which could be connected over any sort of attack to
any system. It is one of the least complex procedures so as to
assemble an attack yet it is a standout amongst the most
widely recognized utilized methodologies. For example, if an
attacker needs to discover out a password of a framework by
using this attack the assailant will attempt each conceivable
mix until the right password is found [4]. Hence, this
technique can be characterized as running the attack
repeatedly until an effective breach is accomplished. The
brute force is recognized as one of the best attacks by the Data
Breach Investigators where it shapes 22% of information
rupture attacks.
POSSIBLE SOLUTIONS FOR THE ISSUES
This view incorporates the information they get to,
the setup, the client the board and specific usefulness. A multi-
tenant’s application has a similar example among various
clients to decrease overhead cost. Taking care of various
occupants inside one application occurrence requires a few
adjustments as each tenants need its own view [5]. Multi-
occupancy can likewise be accomplished through a devoted
innovation that gives assets to tenants exclusively. This is
known as multi-occupancy by means of physical partition.

This choice is by a long shot the costliest. In spite of the fact
that CSPs offer this as a discretionary design to uncommon
clients, who need to utilize the Cloud administration however
would prefer not to impart equipment assets to other clients.
Multi-tenancy in Cloud registering could prompt a
circumstance whereby attackers and the unfortunate victim
share a similar PM. This presents another security challenge
which customary system centred methods cannot alleviate.
When an objective VM has been recognized, a brute force
attack would then be able to be propelled [2]. This is one of
the potential deficiencies of multi-tenancy and shows how by
spending only a couple of dollars, an attacker can separate
information from exploited people utilizing a side-channel
attack when facilitated on the equivalent PM. These attacks
are regularly time not perceivable by the PM's hypervisor or
occupant working framework.
VM Introspection (VMI) empowers data assembling
about virtual machines, virtual systems, security and virtual
condition setting without the utilization of operators. The
capacity of malware to conceal or debilitate from security
operators is a security issue that has tormented the security
industry for quite a long time [10]. VMI offers a fascinating
way to deal with influence the hypervisor for a positive review
of VMs. VMI is fundamentally a hypervisor based
administration that analyses the inward condition of a running
for all intents and purposes machines. Later advancements
have been marketed that influence VMI to give abnormal
amounts of division and confinement for visitor VMs or then
again Cloud-administrations tenants.
Placement prevention: In request to decrease the
impact brought about by shared infrastructure, a couple of
recommendations to guard the attack in each progression are
given in [12]. For example, cloud service providers may
jumble co-living arrangement by disabling traceroute, or
potentially by arbitrarily doling out interior IP delivers to
propelled VMs. In order to diminish the achievement rate of
situation, cloud suppliers may give the clients a chance to
choose where to put their VMs; notwithstanding, this strategy
does not avoid a savage power system.
Co-residency recognition for more than two virtual
machines: a definitive arrangement of cross-VM attack is to
eliminate co-residency. The users may require physical
disengagement, which can even be composed into the Service
Level Agreements (SLAs). In any case, cloud merchant might
be hesitant to desert virtualization that is gainful to cost
sparing and asset usage. One of the left choices is to share the
foundation just with agreeable VMs, which are claimed by the
same client or other dependable clients. To guarantee physical
segregation, a client ought to be empowered to check its VMs
selective utilization of a physical machine.
V. SECURITY AND ISOLATION IN THE MULTITENANT
FRAMEWORK
Multi-tenant’s information disengagement is a basic
innovation in multi-occupant plot, the inability to confine the

information will be the safe concealed inconvenience in multi-
tenant’s condition to each occupant [3]. In Multi-tenant’s
condition, the information does not permit to be shared, so the
related defensive measures ought to be taken to help multi-
tenants information cannot be gotten to by different occupants.
Multi-tenant’s information seclusion is normally partitioned
into three dimensions, including: i) An autonomous database,
ii) A common database, separated information structure and
finally iii) Shared database, shared information design.
The sorts of multi-tenure security utilized to a great
extent rely upon the cloud-based administration and how it has
been actualized. Most cloud specialist organizations will
arrangement security at all layers—for the most part since they
will have a wide range of multitenancy in their surroundings.
IaaS clients need to comprehend whether their VMs are being
housed in a similar host nearby those of different clients and
what, assuming any, arrangements the cloud specialist
organization has made to detach them.
For the situation where the onus is on the tenants to
arrange the division, care must be taken to utilize master
guidance in characterizing and keeping up access control
approaches that empower justified access yet limit chance
VI. CONCLUSION
In cloud-based multi-tenant environment in which
organizations, customers as well as individual consumers
share infrastructure or databases so that they leverage the
advantage in terms of the cost, performance at economic scale.
The tenants in a shared architecture may lead to the sharing
the hardware/infrastructure to use the virtual machines/servers.
Other in addition to that, there are scenarios in which they
share databases where the data of one customer resides in a
row and that of another consumer resides in another.
In the cloud services the consumers comprise of both
kind of tenants. In either scenario it is important to have
security measures in place so that it can be ensured that any
tenant does not pose any kind of risk to another consumer in
terms misuse of the applications, privacy violation or data
loss. Therefore, it is important for the service providers to
ensure the protection in the Multi-tenancy environment at all
the layers of their provided services.
VII. REFERENCES
[1] M. Ficco and F. Palmieri, “Introducing fraudulent
energy consumption in cloud infrastructures: A new
generation of denial-of-service attacks,” IEEE Syst. J.,
2017.
[2] Y. Zhang, F. Patwa, and R. Sandhu, “Community-
Based Secure Information and Resource Sharing in
AWS Public Cloud,” in Proceedings - 2015 IEEE
Conference on Collaboration and Internet Computing,
CIC 2015, 2016.
[3] M. Gupta, D. K. Srivastava, and D. S. Chauhan,
“Security Challenges of Virtualization in Cloud
Computing,” 2016.
[4] J. Yin, X. Lu, C. Pu, Z. Wu, and H. Chen,
“JTangCSB: A cloud service bus for cloud and
enterprise application integration,” IEEE Internet
Comput., 2015.
[5] L. C. Ochei, A. Petrovski, and J. M. Bass,
“Implementing the required degree of multitenancy
isolation: A case study of cloud-hosted bug tracking
system,” in Proceedings - 2016 IEEE International
Conference on Services Computing, SCC 2016, 2016.

[6] R. K. Aluvalu and L. Muddana, “A survey on access
control models in cloud computing,” in Advances in
Intelligent Systems and Computing, 2015.
[7] L. Schubert and K. Jeffery, “New software
engineering requirements in clouds and large-scale
systems,” IEEE Cloud Comput., 2015.
[8] M. Anwar and A. Imran, “Access Control for Multi-
tenancy in Cloud-Based Health Information Systems,”
in Proceedings - 2nd IEEE International Conference
on Cyber Security and Cloud Computing, CSCloud
2015 - IEEE International Symposium of Smart Cloud,
IEEE SSC 2015, 2016.
[9] A. Furda, C. Fidge, A. Barros, and O. Zimmermann,
“Reengineering Data-Centric Information Systems for
the Cloud – A Method and Architectural Patterns
Promoting Multitenancy,” in Software Architecture for
Big Data and the Cloud, 2017.
[10] M. Kim, M. Asthana, S. Bhargava, K. K. Iyyer, R.
Tangadpalliwar, and J. Gao, “Developing an On-
Demand Cloud-Based Sensing-as-a-Service System
for Internet of Things,” J. Comput. Networks
Commun., 2016.
[11] L. C. Ochei, A. Petrovski, and J. M. Bass, “Evaluating
degrees of tenant isolation in multitenancy patterns: A
case study of cloud-hosted Version Control System
(VCS),” in International Conference on Information
Society, i-Society 2015, 2015.
[12] V. Del Piccolo, A. Amamou, K. Haddadou, and G.
Pujolle, “A Survey of Network Isolation Solutions for
Multi-Tenant Data Centers,” IEEE Communications
Surveys and Tutorials. 2016.
[13] M. Saraswathi and T. Bhuvaneswari, “Multitenant
SaaS model of cloud computing: Issues and
solutions,” in 2014 International Conference on
Communication and Network Technologies, ICCNT
2014, 2015.
[14] A. Almutairi, M. I. Sarfraz, and A. Ghafoor, “Risk-
Aware Management of Virtual Resources in Access
Controlled Service-Oriented Cloud Datacenters,” IEEE Trans.
Cloud Comput., 2018.
[15] M. Alam and K. A. Shakil, “Recent Developments in
Cloud Based Systems: State of Art,” CoRR, 2015.