AI- DevOps Using in Cloud

NAVEED AKHTER
engnaveedakhtar@gmail.com

Abstract:- This research pays attention to the merging of processes that can acquire knowledge, adjust, and enhance
Development Operations with Artificial Intelligence (AI). their performance over time
It starts by realizing that AI will be an important factor in
many aspects of work and that it will be automating some The cost of software release has been reduced in recent
job functions. Consequently, AI will be presented as a tool years, mostly because of the transition from shrink-wrapped
that will enhance knowledge acquisition, provide job software to software as a service. Organizations that possess the
performance and professional development. The story ability to deploy software at an early stage and at a frequent
stresses the opportunity cost attributed to the shift from rate have a greater capacity to effectively compete in the
software licensing to Software as a Service (SaaS) and market. DevOps, a novel methodology, offers software firms
underscores the benefits gained through early and regular the ability to achieve these objectives. DevOps is an
software release by the organizations which have adopted organizational strategy focused on fostering empathy and
the practice. DevOps, as a revolutionary approach, seeks to promoting collaboration across different functions. DevOpshas
eliminate the gaps in the two central processes namely also been referred to as a "catalyst for increased collaboration
development and operations. The technology emerging, within multinational organizations".
which includes big data, cloud computing, and mobile
internet, calls for quick software deployment and The objective of DevOps is to minimize the duration
consequently, the DevOps approach is what you get. between software creation and operation while maintaining
However, DevOps is a unified approach. In the abstract, it high-quality standards. IBM has defined Collaborative DevOps
will talk about continuous integration (CI) and continuous as the act of creating mechanisms to synchronize software
delivery (CD) spotting the cost-effectiveness and role of development teams with IT operations teams. DevOps is also
automation in the production process. Through AI and defined as the application of the same principles that control
DevOps described, air is evident which is the AI role in development to infrastructure management.
automation and troubleshooting development in the
software and hardware field. In the paragraph the author Now, the development of big data, cloud computing, and
puts forth AI-optimized DevOps as a proposal that is not mobile Internet has resulted in a growth in the number of
only efficient in development and distribution process but requests for quick implementation and frequent use of business
also fast in pacing. The overall wrap-up summarizes the software [1]. Contemporary businesses, particularly those in the
AIOps and MLOps applications in conjunction with mobile Internet sector, need swift value transmission to end
DevOps workflow to eliminate disconnection between customers, get user input, and then iterate and enhance their
machine learning model development and operational goods [2]. During this process, thedevelopment and operation
deployment. The big picture actually is condensed at the teams must abandon their conventional separation and instead
end. It outlines exactly how the AI DevOps approach works adopt the holistic approach of DevOps, which treats them as a
in modern software development, with particular focus on unified entity [3]. There has been an increase in the number of
the cloud CI/ CD platform. demands for the rapid deployment and frequent use of business
software because of the development of big data, cloud
Keywords:- AI-Optimized DevOps, Continuous Integration, computing, and mobile Internet. This has led to an increase in
Continuous Delivery, Automation in Software Deployment, the level of demand for these services. [4].
Streamlined Cloud CI/CD, AIOps and MLOps Integration
DevOps is revolutionizing the conventional software
I. INTRODUCTION development paradigm. Due to the growth and standardization
of the software development industry, software development
Analysts and observers believe that a wide range of is no more a work that can be accomplished by a single software
professional duties characterized by a predictable and engineer [5]. Product managers are required to
repetitive nature will be automated. Intelligent Automation,
often known as AI, refers to the use of artificial intelligence in
participate in contact with customers and form ties with them Continuous delivery streamlines the whole of the software
across the whole of the software development, operation, and delivery process via automation. Each commit will initiate the
maintenance life cycle. When it comes to writing program automatic process of building, testing, and deploying to the
codes, development teams, which are composed of several corresponding environments. The deployment to production
software engineers, are required to work together. In addition to may be performed manually to verify the intended
this, testing teams are accountable for analyzing the codes and configuration before releasing it to the production environment.
software products that are only half finished [6]. Following the The main objective is to streamline the application lifecycle and
successful completion of the final inspection and the audit expedite value delivery by implementing an efficient
conducted by the client, they will also be required to perform production strategy in a short timeframe.
the responsibilities of software execution, procedure, and
maintenance while it is being used [7]. The whole process relies The primary objectives of Continuous Integration (CI) are
on the collaboration of many individuals and teams inside the to expedite issue detection and resolution, enhance software
IT department, and even across other organizations [8]. excellence, and minimize the duration required for validating
and deploying new software upgrades. Continuous integration
The emergence of the Internet age has led to increased emphasizes the integration of smaller pushes and code
demands for improved quality and efficiency in software modifications. A developer regularly makes code contributions,
development and operation, due to the quick changes in with a minimum frequency of once per day. The code is
consumer demand and the external environment [9]. Given this retrieved from the code repository by the developer to ensure
context, DevOps emerged specifically to provide IT personnel that the code on the local host is integrated before it is sent to
with a streamlined working environment and streamlined the build server. The build server will conduct a large number
workflow. of tests during this step, after which it will either accept or reject
the code change. The primary issues associated with
establishing Continuous Integration (CI) include increased
frequency of code changes to the shared codebase, ensuring the
management of a unified source code repository, streamlining
the process of building software and automating the testing
phase. Further issues include conducting tests in settings that
closely resemble the production environment, ensuring the team
has clear insight into the testing process, and facilitating
developers' seamless access to any version of the application.

Due to the frequent occurrence of integration, there is a

substantial reduction in the need to backtrack and identify
errors, allowing the user to allocate more time towards
developing features.

Continuous Integration is cost-effective. Failure to

Fig 1 Continuous Delivery process integrate continually incurs high costs. Failure to adhere to a
consistent strategy will result in longer intervals between
The figure above illustrates that the Continuous Delivery integrations. This greatly increases the level of difficulty in
process encompasses the many stages of the Software identifying and resolving issues. Integration issues may easily
Development Life Cycle (SDLC) with automation disrupt project timelines or lead to complete project failure [11].
implemented at each step. This allows the user to efficiently
deliver updates to clients in a dependable manner and Automation is the central activity of DevOps, with a
accomplish seamless end-to-end automation, such as one-click particular emphasis on setting up, configuring, deploying, and
deployment and rapid release. Therefore, there is no need for a supporting infrastructure and applications. Automation
fixed timetable to implement new items. The user can facilitates the quick establishment of surroundings in a
efficiently handle, troubleshoot, and implement microservices. consistent and automated manner. In the past, server setup and
The code must be resilient, ensuring it handles all edge test application deployment were mostly carried out manually,
cases to maintain a standardized system process. This which had a significant risk of mistakes, lack of stability, and
effectively resolves intricate issues encountered throughoutthe inability to support agile business practices. To tackle these
deployment phase, resulting in a smooth and streamlined challenges, organizations were using highly proficient
operation. Users can increase the frequency of releases, personnel to carry out manual setup. However, this did not
therefore expediting the feedback cycle with consumers [10]. resolve the issue, which might still affect crucial and valuable
tasks such as software deployment, machine setup, operating merge their development work consistently, which may require
system updates, problem-solving, or bug resolution inside a integrating and merging code many times each day. Through
company. Automation serves as a savior by automating the bulk Continuous Integration (CI), software companies can
of vital company processes. accomplish shorter and more frequent release cycles, improve
the quality of their software, and increase the productivity of
The DevOps environment offers a comprehensive their staff. The use of automated software development and
automated approach that eliminates the strain on enterprises testing is included in this strategy.
caused by human intervention or access to production settings.
DevOps uses technologies to automate significant portions of The objective of Continuous Delivery is to identify
the whole software development and deployment process [12]. methods for efficiently, rapidly, and dependably delivering
software that is of high quality and value. Continuous Delivery
Continuous Integration (CI) involves the regular focuses on the velocity of the market and the ability to swiftly
integration and merging of development work, such as code, by go from conceptualization to implementation, surpassing
team members frequently, sometimes several times per day. competitors in terms of speed. The high pace of this industry
Continuous Integration (CI) makes it possible to establish a results in a reduced feedback cycle and a quicker period to
software release cycle that is both shorter and more frequent, achieve value. A shorter feedback loop enables quicker failure,
which ultimately results in improved software quality and faster resolution, more rapid adjustments, and ultimately, faster
increased productivity [10]. Utilizing automated processes for success.
the creation and assessment of software is a necessary step in
this process. After the successful completion of automated tests The agility shown by firms like Amazon and Netflix gives
and quality assessments, the purpose of continuous delivery them a clear competitive edge in the business world. This is
(CD) is to ensure that an application constantly maintains a state seen in their ability to dominate their sectors, expand into
of readiness for production. This is the ideal outcome of a CD. related industries, and establish themselves as market leaders.
Continuous integration (CI) and deploymentautomation are two These organizations possess the ability to swiftly adjust to the
methods that are used by continuous delivery (CD) to evolving marketplace and outpace their competition in
automatically release software in an environment that is delivering novel products and services. The deployment
extremely similar to production. Force is the most important pipeline automates the build, deploy, test, and release process
aspect to consider when it comes to CDs. According to the for an application; it is the central tenet of continuous delivery.
findings of the research [13], this technique provides a If an application undergoes any kind of alteration, the
multitude of advantages, including a decreased risk of deployment pipeline is instantly activated.
implementation, decreased expenses, and more prompt access
to user input. B. Continuous Integration
A common development approach in the software
II. LITERATURE REVIEW development industry is Continuous Integration (CI). At least
a few times a day, team members must merge and integrate their
A. Continuous Delivery development efforts, including code. Software companies may
Continuous delivery often consists of many essential improve software quality, increase team efficiency, and shorten
stages. Developers decompose their tasks into incremental release cycles by using Continuous Integration (CI). Automated
updates or modifications, which are then implemented on the software development and testing is a part of this strategy. As
trunk or mainline in the version control system. Automated tests part of the Continuous Integration process, developers often
provide developers with prompt feedback within minutes commit all of their changes to a central repository, after which
whenever they make a change, which is another crucial they do automated builds and tests. Building or integrating the
component. Automated testing enables prompt feedback and program is often referred to as CI in the software release
learning, giving developers the chance to promptly address any process. An automated part, like continuous integration (CI) or
issues that arise when the automated tests fail. Collectively, build service, and a cultural part, like the habit of frequent
these procedures are referred to as continuous integration. integration, are both necessary.
Following the successful completion of the build and tests,
more processes are carried out, which include doing thorough C. DevOps
automated acceptance testing, human exploratory testing, and A strategy that mixes artificial intelligence and device
performance testing. Automatic program deployment to an analytics, DevOps is a business-oriented approach to software
environment generated automatically using version control- development. When DevOps teams make use of software that
maintained system and application configuration data is is driven by artificial intelligence, they are able to make use of
essential for these to occur. a large number of data points, which enables them to simplify
the testing, development, release, and monitoring of products
One of the most common development strategies used in with increased accuracy and efficiency. It also speeds up the
the software development industry is known as Continuous whole procedure, which is another benefit. [5,6]. Multiple
Integration (CI). Members of the team must integrate and automated processes need identification and resolution of
challenges, as well as collaboration among team members to may result in a higher rate of deployment [16]. DevOps
enhance efficiency in development and operation. facilitates the uninterrupted delivery of value to end-users by
automating and assuring the seamless functioning of systems.
The use of artificial intelligence results in a considerable
improvement in the efficiency of DevOps [6]. Artificial Artificial intelligence (AI) is becoming indispensable for
intelligence (AI) provides a variety of technologies that provide the development and administration of applications [16].
users in the business world the ability to rapidly map and Artificial Intelligence (AI) is user-friendly, but corporations
integrate data, therefore enhancing crucial business choices and often grasp the true potential of this technology when they
fostering a favorable experience for customers. Machines may overlook its use in enhancing organizational activities.
optimize the data management process in DevOps systems to Artificial intelligence plays a crucial role in DevOps by
enhance value creation. effectively resolving various challenges faced by the IT
industry. They are essential for the progression of the digital
Technology enables individuals to effectively manage transformation [15].
information silos, ensuring timely access to accurate messages
and facilitating optimal decision-making [6]. Users may use an IV. AI-OPTIMIZED DEVOPS
AI platform to efficiently map a substantial amount of
fragmented data. Subsequently, this information may be Despite the fact that AIOps and MLOps are relatively new
consolidated into a unified database. Their artificial intelligence terms in the field of software development, their scopes and
employs data processing techniques that enable them to make applications may be distinct from one another. The ideas of
precise judgments devoid of any flaws or contradictions. MLOps are used throughout the whole process in order to
Artificial intelligence data mapping software facilitates the maximize the economic impact of the solutions that are
precise and effortless mapping of vast quantities of data. An generated by machine learning. According to the information
integrative approach may be used to link the separate data presented in this article, MLOps is essentially a combination of
repositories inside the IT system, leading to precise consumer the concepts of DevOps with machine learning, which ends up
insights that may enhance data integrity and efficiency. producing a variety of advantages for businesses. It has been a
regular practice in the field of software development to make
AI technology will facilitate users in efficiently and use of a wide variety of tools and methods in order to fulfill the
precisely exploring diverse data sources to accomplish needs of the applications that are now in use [17].
activities with increased speed and accuracy [6]. Artificial
intelligence has the capability to combine data from many Machine learning software is not immune from this trend,
sources to provide a coherent and effective data analysis as seen by the current excitement around machine learning and
procedure. artificial intelligence. DevOps is morecomplicated in terms of
producing applications that are both quick and effective for
modern clients. It is important to take into consideration the
III. AI AND DEVOPS
ideas of Machine Learning Operations (MLOps) and Artificial
AI and DevOps are interconnected and complementary Intelligence for IT operations when contemplating the inclusion
since AI is used inside a framework to enhance the performance of machine learning or artificial intelligence into DevOps
of information technology devices. DevOps solutions, in (AIOps). Although there are a few significant differences
contrast, are a business-oriented strategy for software delivery between the two statements, the importance of each of them is
[14]. The use of artificial intelligence (AI) enables DevOps similar [17].
teams to acquire the knowledge and skills necessary to expedite
the process of software development and distribution. It is possible to describe MLOps as the use of DevOps
Moreover, AI has the capability to enhance automation, swiftly principles, techniques, and technologies within the context of
resolve issues, and effortlessly address any remaining machine learning initiatives. The concepts of DevOps, which
challenges. AI and DevOps play crucial roles in the IT industry include collaboration, source control, testing, and continuous
[15, 16]. integration and automation (CI/CD), are used by machine
learning projects in order to effectively assist the whole process,
These are business-oriented strategies that expedite beginning with the design phase and continuing through
application development by integrating software development training and deployment [17].
and operations. Enhancing the precision of their findings might
perhaps lead to greater success. The stage may have a more For example, MLOps makes it possible to integrate data
specific focus on evaluating the firm's ability to invest in an science and IT operations teams in a seamless manner, which in
application by predicting the outcome. It is capable of providing turn makes it possible to deliver machine learning projects more
products on a timeline rather than a normal schedulebecause of efficiently and dependably. It is vital to connect a technology
its ability to respond more quickly than usual. Due to its ability sector with well-defined processes that are focusedon business
to generate faster and more accurate operations, it in order to make the most of the capabilities that it has.
Otherwise, machine learning faces the risk of being
reduced to a basic experiment and liability for a firm, with few in order to evaluate the accuracy of the machine learningmodel.
or no economic gains [17]. This would be a significant loss for This is done to guarantee that the model is both practical and
the profession. To optimize the commercial benefits of machine able to be integrated without any problems.
learning, MLOps is used throughout the iterative process of
designing, constructing models, and carrying out operations. B. Tracking
In order to ensure that a machine learning model that has
During the design phase of a machine learning project, the been deployed remains accurate, it is essential to do regular
major purpose is to define the different business and data monitoring of the model. The dependencies of the model, the
requirements that are necessary for the project. During this data version, the usage of the model, and any changes to the
phase, a business model and a solution based on machine model are all carefully monitored to ensure that it functions as
learning are built in order to successfully solve the problem that intended. It is required to predefine the behaviors that are
the user is experiencing. The process of identifying the potential supposed to be shown by the model and then utilize those
users of the application is that which is being carried out here. behaviors as a benchmark for comparison in order to validate
In addition, the software's potential applications are improved, the performance of the model. If the model does not live up to
and an evaluation is carried out to establish whether or not the expectations or displays spikes that are not consistent, the
essential data that is necessary for the training of an artificial proper actions should be taken to address the problem [18].
intelligence model is available. The data that is gathered from
each of these processes is subsequently used in the architectural C. Revision
design of the application, which is driven by machine learning It is essential to make use of a version control system to
[17]. properly manage the machine learning model. This system is
responsible for monitoring and documenting any alterations
The product that was created as a consequence of the that are made to the data sets and code. As a consequence of
design stage that came before it is put through testing during the this, the version control tool makes it possible to keep track of
experimentation and development phase. The purpose of this and save many versions of the machine learning model if events
testing is to validate the practical use of the proposed machine- cause the data to be altered or produce anomalies [18]. When a
learning solution in real-world settings. After that, the machine problem occurs, it is simple to roll back to an earlier version or
learning approach is repeated until a stable model is achieved, exactly pinpoint the area where a software issue was fixed. Both
at which point it may be put into production for use. of these options are available.

DevOps strategies that have been proven successful, such D. Persistence

as MLOps in Operations, are used in order to successfully apply The continuous workflow of DevOps has been included in
the model in a production setting. This application phase is the machine learning pipeline by MLOps, which is one of the
driven by machine learning, and it makes use of testing, greatest advantages of MLOps. The lifetime of a machine
monitoring, versioning, automation, and continuous delivery, in learning model is finite, and it is sensitive to alteration as fresh
addition to other DevOps methods [17, 18]. It is required to data is constantly being added to the system. The abbreviation
apply each of the aforementioned stages as a following step to MLOps, which stands for Machine Learning Operations, was
the one that came before it in order to get a machine learning developed to ease the process of implementing machine
project ready for the customers. learning engineering strategies. These strategies include
continuous integration (CI), continuous delivery (CD),
V. THE PRINCIPLES OF MLOPS constant testing (CT), and continual monitoring (CM) [18].

The Principles of MLOps MLOps, similar to DevOps, is E. Implementation of Automatic Processes

governed by principles in both its implementation and In order to successfully include MLOps in any machine
execution. MLOps is the machine learning counterpart of learning project, it is essential to place a considerable amount
DevOps. Although there could exist more concepts, the of emphasis on the automation process. It is possible to
following ones are considered the most crucial in MLOps. determine the level of maturity of the machine-learning process
by examining the degree of automation present in your machine-
A. Testing Phase learning model. This, in turn, increases the pace at which
As is the case with DevOps, testing is an essential machine-learning models can be trained, developed, and
component of both the process of developing software and the deployed. This theory makes use of critical events as triggers
process of machine learning. A data pipeline, an ML model for each activity in the workflow, which makes it possible to
pipeline, and an application pipeline are the three components have a machine-learning workflow that is automated and does
that makeup MLOps, which is a method of testing machine not need any intervention from humans. Three main stages of
learning systems that makes use of a systematic methodology adoption correspond to the automation ofMLOps.
[18]. A detailed evaluation is carried out on each component
 DevOps's continuous integration and continuous delivery maintains the (standards of quality). Besides revealing the
pipeline automation are used during the third phase to develop, meaning behind the IBM's Collaborative DevOps, the author
test, and autonomously deploy machine learning models. In underlines the coordination of software development and IT
order to build, test, and autonomously deploy machine learning operations, which corroborates with DevOps principles for the
models, the third phase of the DevOps process involves the use purpose of managing the infrastructure.
of the CI/CD pipeline.
While established multinational corporations maintained
control of the market through the use of big data, cloud
computing and mobile internet, the need for speed in launching
and iterating software has risen sharply. Today, the contact with
DevOps, a single entity who offers collaboration between
development and operation teams, is something that is essential
for today’s industries especially in the mobile internet sector.
Comprehensive DevOps platform is pointed out as the solution
tackling the challenges. Those can be the software delivery
whereby the platform is used, the operation costs cuts, and
development risk decreases.

DevOps, the two key phrases in the process of automation

are Continuous Integration and Continuous Delivery (CI/CD)
systems, where every step of the Software Development Life
Cycle (SDLC) is taken care of automatically. The streamlined
and automated process as shown in the Figure-1 depicts the
overall efficiency gained which is due to automation of end-to-
end release, configuration and test automation, and ease of one
click deployment and release ( Figure- 1) as a result. Apart from
removing time-constraints, users can, with ability and ease
Fig 2 MLOPS Automated CI/CD Pipeline [18] through micro-services, be able to fix, correct, and apply
service features.
VI. CONCLUSION
Continuous Integration serves its purpose by fastening the
The conclusion of AI-enhanced DevOps envisioning a bug-out process, boosting the overall quality of the software and
future where Streamlined Cloud Continuous Integration and decreasing as time goes by, the D-Day moments we all so dread.
Continuous Deployment is the norm is remarkable, as it shows As the code changes are incorporated over the timely basis, the
that software development and deployment are changing entrepreneurs of the backtracking and even more the activity of
greatly. AI/DevOps amalgamation signals a major overhaul, feature development emerges. That, in turn, leads to the low
handling existing hurdles and smoothening processes leading to integration cost.
improved productivity and economic stability.
Automation is still the driving force of DevOps practices,
The irreversible development of AI, on the other hand, can allowing us to take care of anthropogenic components, such as
be explained by the now apparent fact that many professional manual installation and implementation. Traditionally, manual
tasks are very predictable and have repetitive natures. The server configuration, software deployment and update process
narrative pinpoints AI's function as a dynamic force with the brings about many risks while stability is low and agility is
capacity for learning, improving, and maintaining performance limited. Automated processes like creating servers or installing
throughout time. In software release, the economic situation and updating software, mitigate these risks, enhance stability
went through the transformation from atraditional inner packing and support agile business activities. Being efficient as a
model, to the Software as a Service (SaaS) model. This gave an DevOps is a solution that means the speeding-up of vital
opportunity not only to those organizations that could release operation processes and also allows to create infrastructure
software earlier and frequently, but also get a positive strategic well-taking into account all consistency and automation
marketing advantage. demands.
DevOps is emerging as a novel approach that seeks to AI and DevOps trend is essential, in line with AI’s tools
infuse empathy and collaboration into a unified organizational that increase the automation and fast problem solving, and in
strategy spanning technology teams, process owners and eliminating the Disruptive factor of the IT industry. AI-based
stakeholders alike. It might be called the (catalyst for DevOps as an innovative model enables organizations to
collaboration) among the multinationals and reduces the time concentrate on capitalizing and exercising appreciated
between the idea and the operation, and, definitely, it
 economic significance to enrich efficiency in offering [8]. Xue Long, Lu Gang, Zhou Qi, et al. Cloud-native- oriented
software. intelligent operation and maintenance architecture and key
technologies, no. 12, pp. 105-112, 2021.
Moreover the abstract talks in brevity about AIOps and [9]. Liu Weiguang. Application of financial grade cloud native
MLOps featuring their association with DevOps methods to distributed architecture. China Finance, no. 6, pp. 3, 2022.
enhance machine learning solutions. The primary parts of [10]. A. Phillips, M. Sens, A. de Jonge, and M. van Holsteijn,
MLOps which are testing, tracking, revision, persistence, and The IT Manager’s Guide to Continuous Delivery:
automation control depict how a strict ordering for an ML to Delivering business value in hours, not months:
become a part of the development and operation pipelines is XebiaLabs, 2015.
important. [11]. Ishaq Azhar Mohammed. (2013). Intelligent
authentication for identity and access management: a
In short, it is not the AI-fueled DevOps exploration which review paper. International Journal of Managment, IT
brings into life an AI/DevOps merger, but the latter is the and Engineering (IJMIE), 3(1), 696–705. Retrieved from
agenda which urges technical integration. It represents a win- https://www.ijmra.us/project
win situation for the software development and deployment doc/IJMIE_JANUARY2013/IJMIEJan13Ishaq.pdf
process as it not only caters to the fast changing conditions on [12]. Sudhir Allam, "Significance of developing Multiple Big
the technological front but also provides a robust and flexible Data Analytics Platforms with Rapid Response",
platform for organizations to keep up with the competition International Journal of Emerging Technologies and
which is relying more and more on innovative technologies. Innovative Research (www.jetir.org), ISSN:2349-5162,
Vol.2, Issue 3, page no.777-783, March-2015, Available
Authors Details :http://www.jetir.org/papers/JETIR1701523.pdf
Passport picture with briefly description [13]. R. Conejo, M. Urretavizcaya and J. Prez-de-la-Cruz,
Current topics in artificial intelligence. Berlin: Springer,
REFERENCES
2004.\
[1]. Chen Yajun, lily li, Xu Xiaokun, et al. Research on the [14]. V. Sugumaran, Distributed artificial intelligence, agent
application of cloud platform for safety monitoring of technology and collaborative applications. Hershey, PA:
water conservancy and hydropower projects based on Information Science Reference, 2009.
cloud computing microservice architecture and DevOps [15]. H. Salzman, "Engineering perspectives and technology
concept. Digital Technology and Application, vol. 38, no. design in the United States", AI & Society, vol. 5, no. 4,
3, pp. 5, 2020. pp. 339-356, 1991.
[2]. Qiao Hongming, Liang Huan, Yao Wensheng, et al. [16]. H. Izadkhah, "Transforming Source Code to Mathematical
Discussion on DevOps security management for 5G Relations for Performance Evaluation", Annales
network. Mobile Communications, vol. 43, no. 10, pp. 5, Universitatis Mariae Curie-Sklodowska, sectio AI –
2019. Informatica, vol. 15, no. 2, p. 7, 2015.
[3]. Liu Liyuan. Application of CMDB in DevOps automatic [17]. T. Bradley and T. Bradley, "Why DevOps means the end
operation and maintenance. Information and Computer, of the world as we know it", TechSpective, 2016.[Online].
vol. 32,no. 11, pp. 4, 2020. Available:
[4]. Tong Xiangjie, Zheng Wu, Xie Fengling, et al. Hardware https://techspective.net/2015/08/16/why-devops means-
DevOps Practice in Digital Transformation of Enterprises. the-end-of-the-world-as-we-know-it/
Value Engineering, vol. 39, no. 1, pp. 5, [18]. D. Linthicum, "What is DevOps? DevOps Explained |
2020. Microsoft Azure", Azure.microsoft.com, 2016.
[5]. Tang Songqiang, Cai Yongjian, Tang Haitao, et al. DevOps [Online]. Available: https://azure.microsoft.com/en-
Construction Research and Practice. Computer Age, no. 4, us/overview/what-is devops/
pp. 5, 2021.
[6]. Li Yang. Application of DevOps in Team Work. Digital
Design, vol. 10, no. 3, pp. 2, 2021.
[7]. Lu Gang, Chen Changyi, Huang Zelong, et al. Research on
intelligent cloud Native Architecture and Key
Technologies for Cloud Network Convergence.
Telecommunications Science, vol. 36, no. 9, pp. 8, 2020.