IRJET-Comparative Analysis of Disaster Recovery Solutions in Cloud Computing

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 05 Issue: 06 | June 2018 www.irjet.net p-ISSN: 2395-0072
© 2018, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 2851
COMPARATIVE ANALYSIS OF DISASTER RECOVERY SOLUTIONS IN
CLOUD COMPUTING
Mr Tejas Abhimanyu Patil1 & Ms. Vaidehi K. Gadewar2
1,2,Dept. of MCA, Sterling Institute of Management Studies, Nerul, Navi Mumbai, Maharashtra, India
----------------------------------------------------------------------------***-------------------------------------------------------------------------------
Abstract: Cloud computing provides accessing of any kind of
services dynamically over Internet on demandbasis.Oneofthe
most significant service that is being provided is storage as a
service. Nowadays cloud computing has easily reached on
many top lists because of its far-reaching implicationsinmany
areas in computing, especially big data which without cloud
computing is at best a concept. Cloud customer can store any
amount of data into cloud storage results to huge amount of
data at the data center. Nowadays, data has beengeneratedin
large quantity that requires the data recovery services or
techniques. Therefore, there is a requirement for designingan
efficient data recovery technique to recover the lost data.
Keywords: Cloud Computing, Data Recovery,
Replication, Backup.
INTRODUCTION:
Cloud Computing is a buzzword that has many different
meanings to many different people. Cloud computing is
based on virtualization that has another perspective for
disaster recovery. With virtualization, the entire server,
including the operating system, applications, patches and
data is encapsulated into a single software bundle or virtual
server. One can easily take the entire backup or a copy of a
data and host it on a server. Nowadays, data has been
generated in large amount that required the data recovery
services. This type of computing will generate a large
amount of private data on main cloud. Therefore ,the
necessity of data recovery services is growing day-by-day
and it requires a development of an efficient and effective
data recovery technique. The main purpose of recovery
technique is to help user to collect information from any
backup server when the data is lost and unable to provide
data to the user.
Today many different techniqueshave been proposed inthe
market till date. The main objective of this research paper is
to summarize the powerful data recovery techniques that
are used in cloud computing domain.
Data Recovery in Cloud Computing: Today data storagein
data centre isincreasing rapidly day-by-day, henceresulting
into large amount of data loss, copying of same data, data
breach, etc. To avoid these problems there is immense need
of finding of some new solution to avoid this kindofdataloss
in near future. We need to find out the proper technique to
recover the data if the gets deleted or destroyed. Lossofdata
is very harmful to the organizational level, as it can cause
huge financial lossto the company’sbusiness. Byusingsome
of the data recovery techniques the original data can be
recovered. Hence, to recover the loss data, efficient
technique is needed to get the loss data back which should
be reliable and efficient enough.
PROBLEM DEFINITION
Causes of Data Loss:
A. Natural Disasters
Nowadays many businesses are totally dependent on
electronic devices or electronic data there after suffers
huge temporary data loss for the company.
This natural disasters like earthquake, tsunami, etc causes
the company a major setback and make them responsibleor
realize the importance of data backup and security.
According to the survey, when natural disasters occursthen
2% of the data is lost. The main reasons of occurrence
natural disasters are global warming. Due to mundane and
nefarious effects one cannot recognize the data loss when
disaster occurs.
B. Critical application failure
It is natural, when an application is left unusable for many
days then it causes a great damage failure to the
organizations. This can cause some critical application
failure leading to some important data loss. By using all
applications that are stored in cloud may reduce the sudden
great damage.

C. Network failure
In common man words cloud server and the clients are
connected through internet connection. If the connection
fails, then the services which are connected to internet are
crashed and data gets lost and also applications which are
based on cloud also gets affected.
D. Network trespassing(intrusion)
When an virus is invaded into our system thentherearehigh
chances of data loss. By placing unusable applicationsinthat
place on a watch list we can prevent occurrence of disaster
E. Hacking or malicious code
Nowadays, hacking is the most common thing happening in
any field. Disaster is the most unpredictablethingonecannot
say anything about it. So if any kind of disastersoccursinthe
system that may be inside or outside of the company, the
organization can prevent these kind of hacking or malicious
code from modifying data then there is a loss of data.
F. System failure
Asthe name itself suggests, if any kind of infrastructureinan
organization fails then whole systems which are connected
in that organization will crash. Since in any organization, all
the systems are connected to one main server/system.Ifthis
system fails, then the entire connection connected to that
system gets crash.This will affect the operating systems.The
main reason why disaster occurs is due to human beings.
Almost 60%-70% of data centers are failed.
Disaster Recovery Solutions:
Some recovery solutions that have been proposed to
overcome the issues mention above:
A. Local Backup
The dependency issue is tried to be resolved in. Data can be
controlled and complete application can be backed up by
arranging the Linux box on customers side. This approach
may help to migrate private servicesto publicandviceversa.
It might also merge the cloud service providers. By this
method, local backup can furnish with various services that
service provider replenishes during any disasters or issues.
B. Geographical Redundancy and Backup (GRB)
In traditional model geographical redundancy can also be
used. Using this model can be costly and cannot be afforded.
Two cloud zones are similar to each other. If any of the zone
among both becomes down, then another zone will be on
and provide the services
C. Inter-Private Cloud Storage (IPCS)
The proposed approach for storage of cloud data. For
business data storage at least three backup locations are
necessary according to the Storage Networking Industry
Association (SNIA). The Data of the user should be kept or
stored in three different locations which are Servers, Local
backup server(LBS) and remote backup server (RBS). This
model helps in increasing data integration
D. Resource Management
Several hardware and software comprisingofhybridstorage
and diverse disks are abode in Heterogeneous clouds. The
cloud storage hoardscomplete businessdata in cloud-based
enterprise. Thusdata protection, security and recovery play
significant rolesin these environments.Datathatisinscribed
into primary host stays at risk unless it directed or
processed into the backup host. Hence, use of advanced
technology is important for recovering the information in
storage clouds. In 2012, has three major solutions for data
recovery have been proposed:
o During disasters, use fastest disk technology for
replicating the data in danger.
o Changing dirty page threshold: The amount of
faulty pages in RAM that have to wait to flush into
the disk might get faster (Rudolph, 1990).
o Prediction and alteration of risky devices: Few
essential elements like power intake, heat
dispersion, carbon credit utilizationandimportance
of data (hoarded on every single disk) can be
figured out in definite period of time.
E. Scale Up/Down
At times, conducting functions with greater supremacy can
lessen the money loss or even increase the earning during
disasters. Cloud service providerswere floodedwithservice
requests after a natural disaster took place in an area. In
such situations, service providers are compelled to manage
the existing users and take care of new requests as well.
Service providers are bound to appease the existent users
and provide satisfactory service to new customers. During
natural calamities(for e.g. Earthquake) the system availsDR
scenario scaling up resources for the high-priority services
(for e.g., voice communication) and scaling down allocated
resourcesto low-priority service (fore.g.,videoon-demand).

Data Recovery Technique in Cloud Computing
Architecture of remote server
The designed architecture includes three different modules
such as:
1. Remote Backup Server
2. Main Cloud Server
3. Number of Clients/users
Remote Backup server preserves the replicated copies of
main server. It is known as remote repository. The central
server, known as the central repository accommodates all
users’ data. The user uploads the document in main cloud
server which is then stored/transferred into the backup
server. If the user demandsto restore thedocumentfromthe
cloud, it is initially chased in the main cloud server. Ifitisnot
available in the main cloud server, then the user goes
through the backup server to recover lost data. Sometimes
natural calamities or manmade disasters in main servers
cause the loss of data or crashes. A recovery technique is
essential to retrieve the lost data. This objective can be
obtained by using the proposed algorithm adequately. To
grant the dependability two or four backup data cloud
storages can be used. The Figure 2 explains the system
architecture consuming four backup cloud storages.
Figure 2 – Data Recovery Technique
SYSTEM ARCHITECTURE
The system architecture illustratesfour backup servers.The
data of the replicated copies is maintained in more than one
server to recover data. When the loss of data occurs at one
location then it can be retrieved or get fromtheotherbackup
maintained in the system.
ANALYSIS AND FINDINGS
Comparison between the Recovery Solutions
A. Local Backup
 Security
It may be more appropriate to call this
advantage peace of mind. With local backup,
which is usually in the form of external hard
drives, the data is just as protected as your
network. Plus, once it’s stored and the drive is
disconnected, it’s safe from any malicious
attacks that affect your infrastructure.
 Speed
With on-site backups, speed is not limited by
connectivity. Backing up all of your data to an
external hard drive usually takes a fraction of
the time required by cloud storage. After that
initial backup, daily updating can be done in a
few minutes.
 Control
With a local backup solution, you know exactly
where your data is, and you retain control over
who can and cannot access your files. Keeping
an offline backup locally is a good way to
ensure that your information doesn’t fall into
the wrong hands.
B. Geographical Redundancy and Backup
 Security
As more businesses move critical applications and
systems to the cloud, the impact a natural disaster
or an unexpected outage can have on your business
operations is greater today. Even small outages can
have long term business impacts. It ensures high
availability of business critical systems across
multiple locations, mitigating the risk of weather
outages.
 Control
Businesses can mitigate downtime by replicating
applications and data across multiple ‘geo-diverse’
locations. Also termed as ‘geo-replication’, the data

that is created or updated in a primary location, is
asynchronously replicated to a secondary location
so that the same data existsand isreadily accessible
in both locations.
C. Inter-Private Cloud Storage
 Control
Private cloud storage works much like public cloud
storage and implements storage virtualization
across an organization, providing a centralized
storage infrastructure that can only be accessed by
the authorized nodes. Private cloud storage
operates by installing a data centre, which houses a
series of storage clusters that are integrated with a
storage virtualization application. Administrative
policies and a management console provide access
to the different storage nodes and applications
within the organization'snetwork. Theapplications
or nodes access the private storage through file
access and data retrieving protocols, while the
automated storage administrator application
allocates storage capacity to them on run time.
D. Resource Management
 Challenges
Cloud delivers computing as a utility as it is available to the
cloud consumers on demand. It is a simple pay-per-use
consumer-provider service model. It containslarge number
of shared resources. So Resource Management is always a
major issue in cloud computing like any other computing
paradigm. Due to the availability of finite resourcesit isvery
challenging for cloud providers to provide all the requested
resources. From the cloud providers perspective cloud
resources must be allocated in a fair and efficient manner.
Distinguishing between cloud storage and cloud
computing
Cloud computing requires higher processing power than
cloud storage. Cloud storage, on the other hand,needs more
storage space.
Cloud computing is essentially targeted towardsbusinesses.
Cloud storage, on the other hand, is utilized both for
professional and personal reasons.
Cloud storage is simply a data storage and sharing medium,
while cloud computing gives you the ability to remotely
work on and transform data (for example, coding an
application remotely).
CONCLUSION:
At present it is very important to all the big or small
organization to store large amount of data in the cloud. A
thorough literature survey is presented in this paper. The
easy way of recovery technique for recovering the deleted
data is discussed in detail. The four backup servers concept
is used to recover the deleted data. The Proposed method
provides the flexibility for the user to recover their lost data
from any of the four backup servers.
LIMITATIONS:
In this research we have compare 4 recoverysolutionsoutof
5 which still have to be taken into comparison.
To improve more accuracy in data recoveryfurtherworkcan
be done using different solution techniques.
REFERENCES:
I. Efficient and Reliable Data Recovery Technique in
Cloud Computing Published: August 25, 2017
Internet of Things and Cloud Computing
II. Chintureena Thingom, “A Study on Tools for Cloud
Disaster
III. Management”, International Journal of
Interdisciplinary and
IV. Multidisciplinary Studies, 2014.
V. Jian-hua, Z., & Nan, Z. (2011). Cloud Computing-
based Data Storage and Disaster Recovery. IEEE
International Conference on Future Computer
Science and Education (ICFCSE), (pp. 629 632).
http://dx.doi.org/10.1109/ICFCSE.2011.157
VI. Kashiwazaki, H. (2012). Practical uses of cloud
computing services in a Japanese university of the
arts against
VII. aftermath of the 2011 Tohoku earthquake.
VIII. Patil, S. R., Shiraguppi, R. M., Jain, B. P., & Eda, S.
(2012). Methodology for Usage of Emerging Disk to
IX. Ameliorate Hybrid Storage Clouds. IEEE
International Conference on Cloud Computing in
Emerging
X. Markets (CCEM) (pp.1-5).
XI. http://dx.doi.org/10.1109/CCEM.2012.6354615
XII. Patterson, D. A., Gibson, G., & Katz, R. H. (1988). A
Case for Redundant Arrays of Inexpensive Disks
(RAID).International Conference of Managementof
Data (SIGMOD) (pp.109-116).Pokharel,M.,Lee, S.,&

IRJET-Comparative Analysis of Disaster Recovery Solutions in Cloud Computing

More Related Content

What's hot (20)

Similar to IRJET-Comparative Analysis of Disaster Recovery Solutions in Cloud Computing (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET-Comparative Analysis of Disaster Recovery Solutions in Cloud Computing