Content uploaded by Gopala Krishna Sriram
Author content
All content in this area was uploaded by Gopala Krishna Sriram on Jan 30, 2022
Content may be subject to copyright.
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology
and
Science
Volume:04/Issue:01/ January-2022 Impact Factor- 6.752 www.irjmets.com
[1]
Resolving Security and Data Concerns in Cloud Computing by Utilizing a
Decentralized Cloud Computing Option
Mr. Gopala Krishna Sriram*1
*1 Software Architect, EdgeSoft Corp, McKinney, TX USA
ABSTRACT
There are a variety of security concerns around cloud computing infrastructure technology. Some of these
include infrastructure security against threats, data privacy, integrity, and infrastructure stability. In modern
cloud computing, there are two models that cloud computing infrastructures follow: centralized cloud
computing and decentralized cloud computing. Centralized cloud computing is susceptible to outages, data
breaches, and other security threats. Decentralized cloud computing is more resilient to outages due to geo-
redundancy technology, and data is better protected by encryption through Reid Solomon erasure coding.
Keywords: Security practices; Cybersecurity; Data integrity; Cloud computing; Decentralized cloud computing;
Blockchain; Geo-redundancy; Reed Solomon erasure coding, Etc.
I. Introduction
Cloud computing refers to a computing infrastructure solution where resources such as data storage are
delivered as a service that is referenceable from anywhere in the world.¹ The most commonly referenced and
accepted definition of cloud computing comes from The National Institute of Standards and Technology's
(NIST), which defines cloud computing as, “A template for providing the suitable and when needed access to
the internet, to a collective pool of programmable grids, storage, servers, software, and amenities that can be
rapidly emancipated, with little communication and supervision from the provider”.² The idea of cloud
computing has revolutionized the way that computing infrastructures are developed, deployed, and utilized by
users, enterprises, and developers. Cloud computing has evolved and gained popularity due to its desirable
attributes such as scalability and elasticity in both infrastructure itself and the cost, with many providers
offering pay-as-you-go methods instead of all-encompassing prices, and the increased security and data
integrity that comes with cloud infrastructures.
Cloud computing allows services and resources to be consumed using an on-demand method. Resources
such as storage or virtualization resources can be accessed from anywhere in the world at a moment’s notice.
This is different from traditional resource accessibility, where one would have to install hardware to a local
workstation or server before beginning to utilize it, and the hardware was limited in its capacity. Cloud
resources can be added easily and seamlessly without any manual intervention on local hardware.
By storing data or utilizing resources that are part of cloud infrastructure, resources and data are
physically stored in either one geographical location such as a data center, or they are stored in a variety of
geographically diverse locations, such as a variety of data centers. Cloud computing infrastructures that have
the entirety of data and resources stored in one geographical location are referred to as centralized cloud
computing infrastructures, whereas cloud computing infrastructures that have data and resources stored in a
variety of different geographical locations are referred to as decentralized cloud computing infrastructures.³
Figure 1 depicts a visual representation of a centralized cloud computing infrastructure versus a decentralized
cloud computing infrastructure.
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology
and
Science
Volume:04/Issue:01/ January-2022 Impact Factor- 6.752 www.irjmets.com
[2]
A blockchain network is a technology infrastructure that is distributed and uses digital ledger
technology to encrypt, track, and secure all transactions on the network. Blockchain networks are immutable,
meaning every transaction and record that is transmitted over a blockchain network is unable to be changed or
edited.³ This is a layer of security that decentralized cloud computing infrastructures utilize since most
decentralized cloud providers build their infrastructures off of blockchain networks. Some of the most common
of these networks are the IPFS, Sia, or Storj networks.
Blockchain networks are inherently more secure than traditional networks, which are what most
centralized cloud computing infrastructures utilize.
Though cloud computing infrastructures can be used for resources to run virtual machines,
containerized systems, or other virtual systems, this research paper will cover the aspect of storing data within
a cloud computing infrastructure and the security concerns relating to data storage.
Since data stored in cloud computing infrastructures can be accessed from anywhere in the world, there
are several security concerns and problems associated with using cloud computing, despite the numerous
benefits associated with it. The main points of concern regarding cloud computing security include data
security, with the first and foremost being user data privacy and protection, data integrity when stored in data
center locations, (as opposed to users or enterprises storing their data locally in their workstation
environment), cloud computing infrastructure stability, and cloud computing infrastructure administration.
Fig 1: Centralized cloud computing infrastructure (left), decentralized cloud computing infrastructure (right).
II. Security
Regardless of the type of cloud computing infrastructure, users and enterprises alike are heavily
concerned with a variety of security concerns, both regarding cybersecurity attacks, data privacy and integrity,
and cloud computing infrastructure stability.
Types of Cybersecurity Threats
There are new types of cybersecurity threats emerging every day as new technology develops and
evolves. Cloud computing is not immune to traditional cybersecurity threats, and in fact, is more susceptible to
certain types of threats.
Phishing: Phishing refers to the practice of sending fraudulent information or messages that appear to be from
genuine, trusted sources in an attempt to elicit sensitive information from the target.
Ransomware: Ransomware refers to malicious computer programs or software that prevent the user from
using the computer or workstation until a sum of money or another demand is relinquished.
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology
and
Science
Volume:04/Issue:01/ January-2022 Impact Factor- 6.752 www.irjmets.com
[3]
Trojan: In Cybersecurity, Trojan refers to a malicious computer program or software that is packaged to
appear as if it is a useful, legitimate piece of software but in the background runs malicious processes meant to
record sensitive information and relay it back to the distributor of the Trojan software.
Botnet: A botnet refers to a private network of computers that have all been infected with a harmful or
malicious piece of software and controlled in unison for unwanted activity such as mass distribution of spam
messages.
Distributed Denial of Service: A distributed denial of service attack refers to a malicious attack meant to
disturb a network service or resource by flooding the resource with inbound requests to overload its resources
and make it unavailable to legitimate requests.
Adware: Adware refers to malicious programs that display advertisements with the intent to sell products or
services.
Crypto-mining: Crypto-mining refers to the practice of using computers for the mining of cryptocurrency
without the user’s knowledge. This results in a monetary gain for the party who deployed the crypto mining
malicious software.
According to the CISCO 2021 Cyber Security Threat Trends report, the top cybersecurity threat of 2021
was crypto-mining attacks. Table 1 displays the results of the CISCO Cyber Security Threat Trends report,!
showing the different types of cybersecurity threats and the percentage that each threat type has gone up since
2020, with different percentages based on industry. This table uses the financial, healthcare, and manufacturing
industries for comparison.
Table 1. 2021 Percentage of Increase for Different Cybersecurity Threats Since 2020. (Percentages broken down by threats
targeted to different industries).
Cybersecurity
Threat
Target Industry
Manufacturing Healthcare Financial
Phishing
Ransomware
Trojan
Botnet
Cryptomining
All Others
13%
20%
6%
4%
48%
9%
29%
8%
46%
4%
13%
46%
5%
31%
2%
5%
11%
Table 1: Percentage of Increase for Different Cybersecurity Threats Since 2020
Industries that store their data on cloud computing infrastructure are included in this data, though they
are not represented individually. Through the data, the threat that had the largest increase overall (all industry
percentages together) was the cybersecurity threat of phishing, which was up 88% total from 2020.
Phishing can take many forms and cloud computing is especially susceptible to phishing attacks. Many
cloud computing infrastructures include file-sharing options, often in the form of an email link sent to the
individual the file is being shared with. This email can be replicated and made fraudulent, leading to successful
phishing attacks in which the target believes a colleague has sent them a file through the cloud computing
infrastructure file share, only to be led to a false document that records private information.
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology
and
Science
Volume:04/Issue:01/ January-2022 Impact Factor- 6.752 www.irjmets.com
[4]
Data stored in traditional centralized cloud computing infrastructures are susceptible to all of the
previously listed cybersecurity threats. The traditional cloud computing infrastructure model does not take into
account methods to counteract these threats, and each cloud provider has its own methods for securing its
cloud against as many of these threats as possible. The decentralized cloud computing infrastructure model,
however, has many innate security measures, with most being measures that are inherited from the blockchain
networks that decentralized cloud computing infrastructures are built on top of.
Data Privacy
Since cloud computing resources can be accessed from around the world, data privacy is often the
foremost concern of users and enterprises. Data privacy concerns include data security in regards to
cybersecurity threats as mentioned above, but also data privacy from ‘bad actors’ or individuals who act
independently to access and exploit personal data.
Data privacy can be increased through data encryption methods. When storing data in cloud infrastructure,
data is not always encrypted by default. In many cases, when storing data to a centralized cloud, the user or
enterprise uploading the data must first encrypt it before uploading it for maximum security. When storing
data on a decentralized cloud, however, data is encrypted both in transit and while at rest. Since data is stored
in a variety of geographical locations, each piece of a data file is encrypted separately. One piece of a data file is
referred to as a shard of data. Each shard is unable to be decrypted or accessed without first being compiled
with the other shards. This encryption method is known as Ried Solomon erasure coding. ³ Figure 2 shows a
visual example of how data is stored through erasure coding on different storage resources referred to as nodes
in this figure. The grid that the data is spread across can be a small network of nodes, but in modern
decentralized cloud computing, this grid often refers to a blockchain network. Data stored across a variety of
nodes has increased security against cybersecurity threats, as it must be compiled before it can be accessed,
and it can only be accessed by the user who uploaded the data to the blockchain network. This technology
eliminates malicious attacks that seek to steal and retain data content since data content is inaccessible to
anyone but the data owner.
Figure 2: Visual representation of erasure coding technology.
Data Integrity
Another concern around cloud computing and data stored on it is the integrity of that data. Cloud"
computing resources are stored in a variety of locations, which may be susceptible to common events that
affect data integrity such as power outages, hardware failure, or natural disaster. Any of these factors could
affect data integrity, and if data is only stored in the cloud without any secondary backup or storage location,
data can be lost. For this reason, many users and enterprises utilize what is known as a multi-cloud solution,
where data is replicated across a variety of cloud computing providers for redundancy. This concern only
applies to data stored in a centralized cloud computing infrastructure, since decentralized cloud computing
infrastructures store data using geo-redundancy. Geo-redundancy is the practice of storing data across a
variety of locations, so if one location is susceptible to data loss, the remaining data can still be accessed.³
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology
and
Science
Volume:04/Issue:01/ January-2022 Impact Factor- 6.752 www.irjmets.com
[5]
Another attribute to data integrity is assuring that data cannot be changed or edited by other users,
intentionally in a malicious way or by mistake. In a decentralized cloud computing infrastructure, data is not
mutable by anyone except the user that uploaded the data. This assures that data holds its integrity as far as
content, rather than physical integrity.³ Since every transaction on a decentralized cloud infrastructure that is
built on a blockchain network is recorded extensively, users can see exactly when data was modified by
themselves and track changes and edits, assuring that no one else has changed or accessed their data.
Centralized Cloud Computing Infrastructure Stability
Centralized cloud computing infrastructures are easily affected by geographical outages or disasters,
resulting in the cloud infrastructure being offline for a period of time. In December of 2021, many industries
and enterprises suffered losses due to an outage of the Amazon Web Services US East 1 region. This outage
affected companies such as Netflix, Disney, and thousands of others. This outage caused many to question the#
stability of the cloud computing infrastructure and look into other options for their data storage and resource
hosting.
Decentralized Cloud Computing Infrastructure Stability
A decentralized cloud computing infrastructure does not provoke the same stability concerns that the
centralized cloud computing model does. Decentralized cloud computing infrastructures use geo-redundant
resources, which means that if one resource or region goes down, traffic is routed to another region where data
and resources are still accessible and available.³ This is because decentralized cloud computing replicates
resources and data across different locations automatically, eliminating outages unless a significant amount of
the locations are experiencing outages. Each location is often located in drastically different places than the
other locations, such as in entirely different locations rather than just different buildings. This means there is
no single point of failure for a decentralized cloud computing infrastructure, giving decentralized cloud
computing high stability in comparison to traditional centralized cloud computing infrastructures.
Cloud Computing Infrastructure Administration
With any resource or service, the administration of the service or resource is always an area for concern.
Cloud computing is often utilized for storing hundreds of thousands of petabytes of data per enterprise, which
often includes vital business records such as revenue, customer data, and tax data. Storing this kind of data on a
cloud computing infrastructure requires that the enterprise trusts the administration of the cloud provider
since they have the means to access or modify that data. Cloud infrastructure administrators would not in good
conscience do something of that nature, but if their accounts were to be compromised due to poor security
practices or cybersecurity attacks such as phishing, their accounts could be used for malicious activity. For this
reason, many cloud providers have heavy security and security training for their administrators to avoid this
scenario, though it remains a concern of users and enterprises alike.
This concern, however, is not applicable to a decentralized cloud computing infrastructure. Blockchain
networks are not centrally administrated or managed, and no individual user has access to more permissions
than another on a blockchain network. This provides peace of mind and increased data security, along with the
aforementioned data security measures such as erasure coding and data encryption.
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology
and
Science
Volume:04/Issue:01/ January-2022 Impact Factor- 6.752 www.irjmets.com
[6]
IV. Conclusion
Modern cloud computing has numerous benefits, such as scalability, ease of use, cost-saving pay-as-you-
go methods, and universal accessibility. There are two types of cloud computing infrastructures, one that is
known as the traditional and most widely used and accepted infrastructure, and another more recently
developed and less used infrastructure. These are the centralized cloud computing infrastructure and the
decentralized cloud computing infrastructure respectively. The centralized cloud computing infrastructure
model is more widely used, but has several security risks, data privacy and integrity concerns, and has a single
point of data failure. The decentralized cloud computing infrastructure model has innate security due to its
blockchain network utilization, increased data integrity and privacy through encryption and erasure coding,
and no single point of failure through geo-redundancy. The decentralized cloud computing model solves all the
flaws and concerns associated with the traditional centralized cloud computing model. Though right now the
decentralized cloud is less utilized by consumers and enterprises alike, that is likely to change given the
number of benefits that come with the decentralized cloud and its ability to protect and secure data.
VIII. References
[1] Foster, I., Zhao, Y., Raicu, I., Lu, S.. Cloud computing and grid computing 360-degree compared.
In: Grid Computing Environments Workshop, 2008. GCE ’08. 2008, p. 1–10.
doi:10.1109/GCE.2008.4738445.
[2] Mell P, Grance T. Version 15 The NIST definition of cloud computing October 7. National
Institute of Standards and Technology; 2009 http://csrc.nist.gov/ groups/SNS/cloud-computing
[3] S. Wang, Y. Zhang and Y. Zhang, "A Blockchain-Based Framework for Data Sharing With Fine-
Grained Access Control in Decentralized Storage Systems," in IEEE Access, vol. 6, pp. 38437-38450,
2018, doi: 10.1109/ACCESS.2018.2851611.
[4] W. Liu, "Research on cloud computing security problem and strategy," 2012 2nd International
Conference on Consumer Electronics, Communications and Networks (CECNet), 2012, pp. 1216-1219,
doi: 10.1109/CECNet.2012.6202020.
[5] Cisco affiliates, 2021 Cyber security threat trends- phishing, crypto top the list, 2021.
https://learn-umbrella.cisco.com/ebook-library/2021-cyber-security-threat-trends-phishing-crypto-
top-the-list
[6] Sun, Yunchuan & Zhang 张 均 胜 , Junsheng & Xiong, Yongping & Zhu, Guangyu. (2014). Data
Security and Privacy in Cloud Computing. International Journal of Distributed Sensor Networks. 2014.
1-9. 10.1155/2014/190903.
[7] Renato Losio, AWS US-EAST-1 Outage: Postmortem and Lessons Learned, 2021.
https://www.infoq.com/news/2021/12/aws-outage-postmortem/
