![ ISSN: 2088-8708
Int J Elec & Comp Eng, Vol. 13, No. 2, April 2023: 1817-1826
1818
categorization of selection criteria into three dimensions viz. Development and architecture, code security
and maintenance and support. To get a crisp clear understanding of these criteria we have focused on one
dimension of selection criteria. These evaluation or selection criteria are compared for three representative
SCM automation tools-BitBucket cloud, GitHub Actions and GitLab, taken from alternative sets of
automation tools based on current market demand and their utility scenario. These representatives are studied
in-depth to find out their weightage and evaluation score as per their requirement in the real world. For the
fulfillment of which, case study of TekMentors Pvt Ltd as software development industry has also been taken
into consideration. These computed figures are considered to evaluate their performance on the basis of final
outcome after analysis. As the presence of these criteria in the tool, escalate the software development along
with its delivery process up to much greater extent, so the research here involves the quantification of SCM
tools in terms of overall or final score based on these several selection criteria. The score computed can be
used in the selection of the best performer tool. The structure of the research work in the form of schematic
diagram is described in Figure 1.
Figure 1 clearly lists the steps that are to be followed for the completion of underlying research
work. The analysis performed can be used by researchers/students to realize best automation tool in source
code management phase of DevOps culture. Research can be further extended to cover either remaining tools
or other dimensions of selection criteria for source code management phase in DevOps culture.
Figure 1. Schematic flow of tasks performed in comparing and evaluating SCM automation tools
2. RELATED WORK
The selection/evaluation criteria, tasks, activities or practices considered by DevOps for the
selection of appropriate source code automation tools have been reviewed with respect to different existing
research papers. As Erich et al. [1], in their paper “A qualitative study of DevOps usage in practice,”
highlighted that software industries are focusing on speeding up the development process and also improving
the quality of their software by introducing the term DevOps as a portmanteau term to describe their efforts.
The report details the DevOps implementation practices used by enterprises and the results they see as a
result. In their article, Perera et al. [2] and Erich et al. [3] conducted research to determine how DevOps
practices enhance software quality. They show through the use of multiple regression analysis that culture,
automation, measurement, and sharing are crucial elements in raising the caliber of software. DevOps based
software development was recommended to achieve high quality software.
DevOps has been recognized by Lwakatare et al. [4] as a crucial component of the continuous
deployment paradigm in academic research circles and practitioner communities. It was also characterized by
collaboration, automation, and measurement and monitoring as four major contributors to the DevOps
movement. Riungu-Kalliosaari et al. [5] collectively in their paper on DevOps confirms benefits of adopting
DevOps by conducting qualitative multiple-case study. This study acknowledges that DevOps practitioners
foster stronger departmental collaboration, which improves communication and worker well-being. Dwivedi
et al. [6] and taking into account additional DevOps research that addresses software quality concurrently and
examines the effects of DevOps features on software quality. This study provided a methodical mapping of
the effect of DevOps on software quality. The primary research areas were DevOps automation, culture,
continuous delivery, and quick feedback. In their works on DevOps, Olguin [7] and Jabbari et al. [8]](https://image.slidesharecdn.com/v5927920emr29sep2226mar22k-230124065605-1eb420c9/85/Performance-assessment-and-analysis-of-development-and-operations-based-automation-tools-for-source-code-management-2-320.jpg)
![Int J Elec & Comp Eng ISSN: 2088-8708
Performance assessment and analysis of development and operations based … (Pooja Mittal)
1819
emphasize that the goal of the DevOps movement is to automate the processes of continuous delivery of new
software updates while simultaneously ensuring their accuracy and dependability. The definition of DevOps
has also been thoroughly reviewed in the literature, and Jabbari et al. [8] concur that DevOps expands the
agility element in the software development paradigm.
The literature also shows the development of several models for the guidance of industry or
software practitioners for fruitful and successful implementation of DevOps in practice. Many models have
been proposed in this context like the unicorn framework proposed by Trihinas and others [9], to overcome
different challenges of DevOps through continuous releases. In a similar vein, the DevOps Research and
Assessment model proposed by Forsgren et al. [10] discusses successful product delivery, and Hussaini [11]
accepts the emerging DevOps paradigm as a response to the growing understanding of the gap between the
development and operation teams functions of an organization, which he refers to as the “4 Cs”
(communications, cooperation, culture, and collaboration). The “Wall of confusion” between these teams is
also accepted by authors. Conflicting motives among people, processes, and technology/tools all contribute to
this “Wall”. Thus, there is a need to improve the coordination between the Dev and Ops teams. For
improving the effectiveness and efficiency of DevOps stakeholder’s interest, the concept was also described
in [11]. As many renowned researchers restrict the adoption of DevOps in practice though there are multiples
of theories that are against DevOps application and talk about its challenges and lack of performance measure
[12]. For example, Leite et al. conducted a survey in their paper [13] and discussed different challenges in
DevOps adoption. Other researches also force the compulsion of DevOps practices for the organization to
move towards delivering higher performance and quality software [14].
The individual automation tool discussion that is conducted as the core component of this study
work was missing from the literature review studied, mentioned above, or discussed thus far. At the source
code management stage of software development, we also conducted a comparative analysis of automation
tools to get the best out of many alternatives present in current market scenario. Underlying research has
selected BitBucket, GitHub Actions; GitLab has been taken into consideration as per the tool market demand
and also based on most commonly followed tools.
3. ALTERNATIVE AUTOMATION TOOLS EXISTENCE FOR DIFFERENT PHASES OF
SOFTWRE DEVELOPMENT IN DevOps CULTURE
Different set of tools exists to render automation in all software development phases of DevOps
culture. These automation tools help development and operations teams to speed up the tasks of software
release. Existence of different automation tools in DevOps is shown in Table 1. Table 1 includes alternative
and most commonly followed automation tools at different stages of software development. These
automation tools ease the tasks of development and operations teams but challenges exist to choose the best
tool from the corresponding list of alternative tools. Tools chosen prior to development not only save
development time but also reduce the cost of rework at the same time. Current research works around the
coding phase of development to comparatively analyze the best SCM automation tool for the underlying
phase. To conduct this work, three most commonly followed SCM tools have been taken are BitBucket,
GitHub Actions, GitLab. These tools are compared to one another based on their needful characteristics and
requirements of the business market. Results of the comparison helps to choose the best out of many
available alternatives.
Table 1. Alternative automation tools for different phases of software development in DevOps culture
DevOps Phase Automation Tools
Requirement management Confluence, BookStack, Notion
Project planning
Coding
AgileCraft, Trello, Jira
BitBucket, GitHub Actions, GitLab
Build Teamcity, Jenkins, Bamboo
Testing UFT, TestComplete, Selenium
4. SCM TOOLS FOR CODING PHASE OF DEVOPS SOFTWARE DEVELOPMENT CULTURE
Primary aim of DevOps is to optimize the end-to-end delivery process of the software. For this
purpose, DevOps employs different automation tools at each and every phase of development life cycle with
the objective to cut down the manual efforts and pace the cycle time from requirement to software release.
This optimization process starts from the requirement gathering/management phase where it is recommended
to use agile practices and continuous prioritization of requirements. Once the development team has
prioritized a subset of requirements in the form of stories, the development team picks the stories in their
priority order and starts working on the software development.](https://image.slidesharecdn.com/v5927920emr29sep2226mar22k-230124065605-1eb420c9/85/Performance-assessment-and-analysis-of-development-and-operations-based-automation-tools-for-source-code-management-3-320.jpg)
![Int J Elec & Comp Eng ISSN: 2088-8708
Performance assessment and analysis of development and operations based … (Pooja Mittal)
1821
There exist many selection criteria for tool excerption and for their crisp understanding we have
categorized them into three pillars or dimensions as shown in Table 2. For this current research, we are
targeting only development and architecture dimension of selection criteria that covers multiple of
parameters like support for different repositories, access management, and issue board. All these parameters
for selected criteria are explained beneath.
5.1. Support for multiple repositories in project
Source code of any application can be organized in many different ways. Legacy way of managing
the code is to logically divide the code based on application functionality, where all the code is packaged
together in the form of artifacts and deployed at once. In this form of delivery, complete code will be
deployed at once even if there is a change expected in any single file or feature. After the deployment of
code, need arises for stringent management of deployed code to maintain the integrity, consistency,
uniformity along with unambiguous features. This need advances towards the SCM Automation tool, a much
more optimized way of managing the code, employed by DevOps culture in which application is divided into
multiple components and each component is independent of each other and can even be deployed to
production individually. Architecturally each component can be developed as a micro service where all micro
service code is separate from each other and each micro service can be deployed to production independently
without any impact on the availability of other services.
In this type of model, during the deployment or release process, all the application would be
available to users except only one feature of the application. In order to support this type of flexible and
scalable architecture, each micro service code should be stored separately in their respective repository and
each repo has its own CI/CD pipeline to support deployment of individual micro service as assigned in the
Figure 3 that explains the steps to create a new repository in GitHub. Hence it becomes most important that
SCM tools support multiple repositories within a single project where each repository can store and manage
the code of each application component i.e., micro service.
Figure 3. Creating new repository in GitHub source code automation tool [15]
5.2. Access management-project level/repository level/both
With the increase in microservices culture, there is always a need for a separate code repository for
each micro service. Each micro service has a predefined set of objectives and functionalities for delivery to
different stakeholders out of which only a very small number of stakeholders need access to all microservices
e.g. the solution architect of the complete software may require all Microservices. In order to manage these
selection criteria well, there is a need for provision of both project level code and repository level access with
different sets of roles viz. admin, read-only, read-write, and reviewer.
5.3. In-built CI/CD support
While tools like BitBucket, GitLab and GitHub Actions are primarily used for source code
management; there is a huge demand for additional features that can ease out many tasks of developers along
with less maintenance overhead and requirement of low learning curve. One of these most demanding
features in the list is in-built support for build and deployment, which can optimize the end-to-end delivery
process up to much greater extent. After the code commit statement by the developer, this in-built support
feature of the tool automatically kicks the build pipeline integrated with required repository to bundle the](https://image.slidesharecdn.com/v5927920emr29sep2226mar22k-230124065605-1eb420c9/85/Performance-assessment-and-analysis-of-development-and-operations-based-automation-tools-for-source-code-management-5-320.jpg)
![Int J Elec & Comp Eng ISSN: 2088-8708
Performance assessment and analysis of development and operations based … (Pooja Mittal)
1823
flexibility to assign a review to multiple reviewers like senior developers or the architect, so that anyone who
has bandwidth to do the code review can pick that review request and give the feedback with no dependency
on a single reviewer.
Figure 5. Adding of multiple reviewers for different branches of source code in management tools [16]
5.8. Code snippet
Code snippet is a small piece of code that can easily be shared with anyone in the team. It is a
reusable piece of code that we can make public so that other people can copy it or refer to it for common
functionality. Generally, this code snippet feature is more important for companies who are doing open
source projects or for a community that is supporting any open source project. This snippet of code can be
stored in separate files from where it can be referred to. It also has the advantage of adding other people to
put comments to allow them make any changes in the code to make it more improved and optimized. These
Snippets could be a single file or a collection of files.
5.9. Hooks (pre-commit/post-commit/WebHooks)
Hooks are the mechanism to create and validate information from other tools that are integrated with
source code management tools. Generally, event-based, specific event triggers hooks do this work of
validation for other tools. Hooks are of different types viz. pre-commit hook, post-commit or WebHooks.
Consider a good example of a pre-commit hook that validates ticket ID in the commit message. Now if a
developer is working on a story or defect, it is mandatory for the developer to mention the story/defect ID
from the task management tool like Jira or Trello but if that issue ID is not mentioned in the commit
message, then the source code management tool can block that commit. This block will not only help to
maintain the automatic traceability from the requirement to source code but also help the developers to
maintain the code of the live application. These hook features prevent the developers from introducing the
regression defect or breaking the existing functionality of the application.
In a similar way, post-commit hook helps to trigger any action in other tools after the commit.
Examples could be resolving tickets in task management systems like Jira, and Trello. after the developer
commits the code for that particular ticket or triggers the build in tools like Jenkins, Bamboo, and Teamcity.
Tools from the same vendor are integrated implicitly by application links like BitBucket, JIRA, and Bamboo.
Hence implementation of pre/post commit hooks is through native integration but for cross vendor tool
integration like BitBucket-Teamcity, BitBucket-Jenkins needs WebHooks.
5.10. Rest application programming interface (APIs)
Most of the DevOps tools expose both graphical user interface and APIs for interaction. Rest API is
important to communicate the tool programmatically. Organizations create their own tool to extract data for
analytical purposes. Not all the reports required by the organizations are available off the shelf by the tool.
Thus, they need to write custom solutions to extract the data using rest APIs. This section clearly describes](https://image.slidesharecdn.com/v5927920emr29sep2226mar22k-230124065605-1eb420c9/85/Performance-assessment-and-analysis-of-development-and-operations-based-automation-tools-for-source-code-management-7-320.jpg)
![ ISSN: 2088-8708
Int J Elec & Comp Eng, Vol. 13, No. 2, April 2023: 1817-1826
1824
the need or usage of different selection criteria for development and architecture dimension. These criteria
are taken into consideration before the actual selection of the automation tools. Other criteria given in Table 2
are also checked for select the best automation tool, but currently this paper works on parameters under
development criteria.
6. PERFORMANCE ANALYSIS OF SOURCE CODE AUTOMATION TOOLS-A TABULAR
COMPARISON BASED ON SELECTION CRITERIA
Many alternative source code automation tools exist for managing code in DevOps culture. This
research has taken BitBucket Cloud, GitHub Actions, and GitLab as representative tools for source code
management based on their market analysis and common usage. These tools are assigned scores in 3-Steps:
Firstly, a numerical value called weightage/weight is assigned to each selection criteria. Weights are assigned
as per the merits or requirements of criteria to their target customers. Next step includes assigning an
evaluation score to the SCM automation tool as per the selection criteria or corresponding feature availability
in the tool and the last step multiplies the weightage with the corresponding evaluation score and finally adds
them for each tool to get an overall score for that tool. These assigned and evaluated scores for all the
representative tools are mentioned in a comparative tabular format as shown in Table 3.
Table 3. Performance analysis table containing weightage assigned along with evaluation and overall score
assigned to the representative SCM automation tool [17]–[25]
Selection Criteria Description
Evaluation Score (out of 10)
Overall Score=Weightage *
Evaluation Score
Weigh-
tage
Bit-Bucket
Cloud
Git-Hub
Actions
Git-
Lab
Bit-Bucket
Cloud
Git-Hub
Actions
Git-
Lab
Dimension-Development and Architecture
Support for multiple
Repo in Project
Maximum number of repositories which
can be created in a single project/group
10 9 9 9 90 90 90
Access management-
project level/
Repository Level/Both
Managing permissions at project level as
well as at repository level 10 7.5 7 [18] 8.5 75 70 85
In built CI/CD Support In built feature to build test, deploy code 8 6.5 [20] 8 8 52 64 64
Feature Request Ability to build features requests within
SCM tools
8 7 8 8 56 64 64
Issue Board In built features to manage tasks with
help of Kanban and scrum boards
6 5 8 7.5 30 48 45
In-Built Code Review Ability to perform code review within
the tool
9 8.5 8.5 9 76.5 76.5 81
Multiple Reviewers Ability to add multiple reviewers for a
single review/pull request
9 8 8.5 9 72 76.5 81
Code Snippet Snippets allow to share files with
yourself, members of your workplace
7 7.8 9 7.5 54.6 63 52.5
Hooks [Pre-
Commit/Post-
Commit/WebHooks]
Ability to enforce hooks/merge check
functionalities 10 7.5 7.5 8.5 75 75 85
Rest API usage matrix and other data available as
Rest API call
9 7.5 8.5 9 [18] 67.5 76.5 81
Best In-Class
reporting and
analytics
In built report for all below mentioned
items 0 0 0
Most commit by
persons
For a specific repository who is the
active contributor
7 7 8 7.5 49 56 52.5
Most active
Project/Repo
Most active repository of a project based
on commits
7 0 0 0 0 0 0
Inactive Repo/Branch Inactive Repositories and Branch 7 0 0 0 0 0 0
Commit Frequency Frequency of commits for a repository 7 7 7.5 8.5 49 52.5 59.5
Merge/Pull request
Pending/Overdue
Pull request dashboard for
repository/Project
9 7.5 9 8.5 67.5 81 76.5
814.1 893 917
Analysis Table 3 is the result of implementation and evaluation with respect to all mentioned
features with SCM tools-BitBucket, GitLab, and GitHub for TekMentors Consulting Pvt Ltd. Software
company and renowned researchers already referred to in the paper. Tool recommended is also being used in
companies for the efficient management of source code. Based on the overall score of all the representative
SCM tools, GitLab can be concluded as the best performer tool among others.](https://image.slidesharecdn.com/v5927920emr29sep2226mar22k-230124065605-1eb420c9/85/Performance-assessment-and-analysis-of-development-and-operations-based-automation-tools-for-source-code-management-8-320.jpg)
![Int J Elec & Comp Eng ISSN: 2088-8708
Performance assessment and analysis of development and operations based … (Pooja Mittal)
1825
Several different functionalities or features available in GitLab, as seen in the performance analysis
table, makes it the most commonly observed automation tool for source code management. This can further
be analyzed from interest over time report for previous 6-7 years according to Stack Overflow report. Among
these representative SCM tools, again GitLab shows the highest usage trend in last years based on stack
overflow report as depicted in the Figure 6.
Figure 6. SCM Automation Tools interest over time depicted by stack overflow latest trends report over past
7-8 years [26]
7. CONCLUSION AND DISCUSSION
This paper discussed many tasks or activities as the selection criteria to choose the best performing
source code automation tool for the coding phase of DevOps software development culture. As there are a
number of selection criteria in the market and these were not narrowed till, so this was our idea to first
categorize them into three dimensions in order to smoothen out the selection process of automation tools. For
this work, we have only focused on the development and architecture dimension of selection criteria. Opting
on these criteria not only helps to manage the source code with the selection of the right set of automation
tools but at the same time it also escalates the development process up to a much greater extent. This research
computes the performance of the representative tools on the basis of weightage assigned, evaluation and
overall scores along with the case study of TekMentors Pvt Ltd. The analysis performed reveals GitLab as
the best source code management tool depending upon the outcome of overall score. GitLab is selected as the
best performer tool on the basis of several selection criteria that are generic in nature. Other alternative
automation tools for source code management can also be chosen as per the requirements of project-specific
criteria. This performance or comparative analysis of source code automation tools will be useful for
students/researchers to learn in-depth about DevOps culture and usage of alternative automation tools at
different stages of software development. A performance evaluation or comparative analysis of other
automation tools may also be carried as a part of further research. For future scope, our work can also be
extended to carry out research on other two dimensions of selection criteria to complete the process of code
management automation tool selection in the software development process.
REFERENCES
[1] F. M. A. Erich, C. Amrit, and M. Daneva, “A qualitative study of DevOps usage in practice,” Journal of Software: Evolution and
Process, vol. 29, no. 6, Jun. 2017, doi: 10.1002/smr.1885.
[2] P. Perera, R. Silva, and I. Perera, “Improve software quality through practicing DevOps,” in 2017 Seventeenth International
Conference on Advances in ICT for Emerging Regions (ICTer), Sep. 2017, pp. 1–6, doi: 10.1109/ICTER.2017.8257807.
[3] F. Erich, C. Amrit, and M. Daneva, “A mapping study on cooperation between information system development and operations,”
in Product-Focused Software Process Improvement, Springer International Publishing, 2014, pp. 277–280.
[4] L. E. Lwakatare, P. Kuvaja, and M. Oivo, “Dimensions of DevOps,” in Lecture Notes in Business Information Processing,
Springer International Publishing, 2015, pp. 212–217.
[5] L. Riungu-Kalliosaari, S. Mäkinen, L. E. Lwakatare, J. Tiihonen, and T. Männistö, “DevOps adoption benefits and challenges in
practice: a case study,” in Product-Focused Software Process Improvement, Springer International Publishing, 2016,
pp. 590–597.
Year](https://image.slidesharecdn.com/v5927920emr29sep2226mar22k-230124065605-1eb420c9/85/Performance-assessment-and-analysis-of-development-and-operations-based-automation-tools-for-source-code-management-9-320.jpg)
![ ISSN: 2088-8708
Int J Elec & Comp Eng, Vol. 13, No. 2, April 2023: 1817-1826
1826
[6] A. Dwivedi, G. Srivastava, S. Dhar, and R. Singh, “A decentralized privacy-preserving healthcare blockchain for IoT,” Sensors,
vol. 19, no. 2, Jan. 2019, doi: 10.3390/s19020326.
[7] R. J. Olguín, “DevOps challenges and implications,” University of Murcia, Murcia, Spain, pp. 1–5, 2019.
[8] R. Jabbari, N. bin Ali, K. Petersen, and B. Tanveer, “What is DevOps?,” in Proceedings of the Scientific Workshop Proceedings
of XP2016, May 2016, pp. 1–11, doi: 10.1145/2962695.2962707.
[9] D. Trihinas, A. Tryfonos, M. D. Dikaiakos, and G. Pallis, “DevOps as a service: pushing the boundaries of microservice
adoption,” IEEE Internet Computing, vol. 22, no. 3, pp. 65–71, May 2018, doi: 10.1109/MIC.2018.032501519.
[10] N. Forsgren, M. C. Tremblay, D. VanderMeer, and J. Humble, “DORA platform: DevOps assessment and benchmarking,” in
Lecture Notes in Computer Science, Springer International Publishing, 2017, pp. 436–440.
[11] S. W. Hussaini, “Strengthening harmonization of development (Dev) and operations (Ops) silos in IT environment through
systems approach,” in 17th International IEEE Conference on Intelligent Transportation Systems (ITSC), Oct. 2014, pp. 178–183,
doi: 10.1109/ITSC.2014.6957687.
[12] A. A. Khan and M. Shameem, “Multicriteria decision‐making taxonomy for DevOps challenging factors using analytical
hierarchy process,” Journal of Software: Evolution and Process, vol. 32, no. 10, Oct. 2020, doi: 10.1002/smr.2263.
[13] L. Leite, C. Rocha, F. Kon, D. Milojicic, and P. Meirelles, “A survey of DevOps concepts and challenges,” ACM Computing
Surveys, vol. 52, no. 6, pp. 1–35, Nov. 2020, doi: 10.1145/3359981.
[14] R. Bolscher and M. Daneva, “Designing software architecture to support continuous delivery and DevOps: a systematic literature
review,” in Proceedings of the 14th International Conference on Software Technologies, 2019, pp. 27–39, doi:
10.5220/0007837000270039.
[15] “Documentation for BitBucket server 7.5,” Atlassian. 2020, Accessed: Sep. 15, 2022. [Online]. Available:
https://confluence.atlassian.com/alldoc/files/278071997/1019389387/1/1597212077409/BitbucketServer-075.pdf.
[16] R. Banerjee, “Backup blog, using the BitBucket API,” Rewind. 2021, Accessed: Sep. 15, 2022. [Online]. Available:
https://www.backhub.co/blog/using-bitbucket-api.
[17] A. Q. Gill, A. Loumish, I. Riyat, and S. Han, “DevOps for information management systems,” VINE Journal of Information and
Knowledge Management Systems, vol. 48, no. 1, pp. 122–139, Feb. 2018, doi: 10.1108/VJIKMS-02-2017-0007.
[18] S. Uzunbayir and K. Kurtel, “A review of source code management tools for continuous software development,” in 2018 3rd
International Conference on Computer Science and Engineering (UBMK), Sep. 2018, pp. 414–419, doi:
10.1109/UBMK.2018.8566644.
[19] S. Bokefode, R. N. Tendulkar, and R. R. Nakade, “Project management system like BitBucket (App),” International Research
Journal of Engineering and Technology (IRJET), vol. 7, no. 6, pp. 5240–5243, 2020.
[20] S. M. Mohammad, “Continuous integration and automation,” International Journal of Creative Research Thoughts (IJCRT),
pp. 2320–2882, 2016.
[21] B. Vasilescu, S. van Schuylenburg, J. Wulms, A. Serebrenik, and M. G. J. van den Brand, “Continuous integration in a social-
coding world: empirical evidence from GitHub,” in 2014 IEEE International Conference on Software Maintenance and
Evolution, Sep. 2014, pp. 401–405, doi: 10.1109/ICSME.2014.62.
[22] L. E. Lwakatare, P. Kuvaja, and M. Oivo, “An exploratory study of DevOps extending the dimensions of DevOps with practices,”
ICSEA, vol. 104, 2016.
[23] H. L. Akshaya, J. Vidya, and K. Veena, “A basic introduction to DevOps tools,” International Journal of Computer Science and
Information Technologies, vol. 6, no. 3, pp. 5–6, 2015.
[24] M. Shahin, M. Ali Babar, and L. Zhu, “Continuous integration, delivery and deployment: a systematic review on approaches,
tools, challenges and practices,” IEEE Access, vol. 5, pp. 3909–3943, 2017, doi: 10.1109/ACCESS.2017.2685629.
[25] I. Sommerville, Software engineering, 9th Editio. Pearson, 2010.
[26] “Stack overflow trends,” Stack Overflow. https://insights.stackoverflow.com/trends?tags=bitbucket%2Cgithub-actions%2Cgitlab
(accessed Sep. 17, 2022).
BIOGRAPHIES OF AUTHORS
Pooja Mittal obtained her Ph.D. degree from Maharshi Dayanand University. Her
area of research and specialization include data mining, data warehousing, and computer
science. She had published more than 50 research papers in renowned International and
National Journals and attended more than 30 Conferences. Currently she is working as
Assistant Professor in the Department of Computer Science and Applications, Maharshi
Dayanand University, Rohtak (Haryana). She can be contacted at mpoojamdu@gmail.com.
Poonam Narang Research Scholar, pursuing Ph.D. from the Department of
Computer Science and Applications, Maharshi Dayanand University, Rohtak, Haryana under
the supervision of Respected Dr. Pooja Mittal (Research Guide and Second Author). Author’s
Qualification is M.Phil. (CS), MCA. She had attended many National and International
Conferences including Springer and IEEE and also published many research papers. She can
be contacted at email: poonam.mehta20@gmail.com.](https://image.slidesharecdn.com/v5927920emr29sep2226mar22k-230124065605-1eb420c9/85/Performance-assessment-and-analysis-of-development-and-operations-based-automation-tools-for-source-code-management-10-320.jpg)
Performance assessment and analysis of development and operations based automation tools for source code management
- 1. International Journal of Electrical and Computer Engineering (IJECE) Vol. 13, No. 2, April 2023, pp. 1817~1826 ISSN: 2088-8708, DOI: 10.11591/ijece.v13i2.pp1817-1826 1817 Journal homepage: http://ijece.iaescore.com Performance assessment and analysis of development and operations based automation tools for source code management Pooja Mittal, Poonam Narang Department of Computer Science and Applications, Maharshi Dayanand University, Rohtak, India Article Info ABSTRACT Article history: Received Mar 26, 2022 Revised Sep 29, 2022 Accepted Oct 24, 2022 Development and operations (DevOps), an accretion of automation tools, efficiently reaches the goals of software development, test, release, and delivery in terms of optimization, speed and quality. Diverse set of alternative automation tools exist for different phases of software development, for which DevOps adopts several selection criteria to choose the best tool. This research paper represents the performance evaluation and analysis of automation tools employed in the coding phase of DevOps culture. We have taken most commonly followed source code management tools-BitBucket, GitHub actions, and GitLab into consideration. Current work assesses and analyzes their performance based on DevOps evaluation criteria that too are categorized into different dimensions. For the purpose of performance evaluation, weightage and overall score is assigned to these criteria based on existing renowned literature and industrial case study of TekMentors Pvt Ltd. On the ground of performance outcome, the tool with the highest overall score is realized as the best source code automation tool. This performance analysis or measure will be a great benefit to our young researchers/students to gain an understanding of the modus operandi of DevOps culture, particularly source code automation tools. As a part of future research, other dimensions of selection criteria can also be considered for evaluation purposes. Keywords: Automation Automation tools DevOps Software development Source code management This is an open access article under the CC BY-SA license. Corresponding Author: Poonam Narang Department of Computer Science and Applications, Maharshi Dayanand University (MDU) Rohtak, Haryana-124001, India Email: poonam.mehta20@gmail.com 1. INTRODUCTION Development and operations (DevOps), a fine fusion of development and operations teams together for speedy, reliable and quality delivery of software. For this purpose, DevOps employs a set of different automation tools at each stage of software development. These tools help in automating different activities or tasks of teams from the development of software till production. This research work revolves distinctively around the coding phase of DevOps culture. Different automation tools that exist for the purpose of source code management (SCM) include Gerrit, BitBucket Cloud, GitLab, Mercurial, GitHub Actions, and Team Foundation Server (TFS). Management goes around with several selection criteria or accustomed activities viz. code security, vulnerability management, internet protocol (IP) whitelisting, in-built continuous integration and continuous delivery (CI/CD), and commit hooks (pre and post). to ease out the critical task of best performer tool selection out of alternative automation tools list. To peculiarly focus on the challenges of Information technology (IT) management or software development teams, present analysis considers
- 2. ISSN: 2088-8708 Int J Elec & Comp Eng, Vol. 13, No. 2, April 2023: 1817-1826 1818 categorization of selection criteria into three dimensions viz. Development and architecture, code security and maintenance and support. To get a crisp clear understanding of these criteria we have focused on one dimension of selection criteria. These evaluation or selection criteria are compared for three representative SCM automation tools-BitBucket cloud, GitHub Actions and GitLab, taken from alternative sets of automation tools based on current market demand and their utility scenario. These representatives are studied in-depth to find out their weightage and evaluation score as per their requirement in the real world. For the fulfillment of which, case study of TekMentors Pvt Ltd as software development industry has also been taken into consideration. These computed figures are considered to evaluate their performance on the basis of final outcome after analysis. As the presence of these criteria in the tool, escalate the software development along with its delivery process up to much greater extent, so the research here involves the quantification of SCM tools in terms of overall or final score based on these several selection criteria. The score computed can be used in the selection of the best performer tool. The structure of the research work in the form of schematic diagram is described in Figure 1. Figure 1 clearly lists the steps that are to be followed for the completion of underlying research work. The analysis performed can be used by researchers/students to realize best automation tool in source code management phase of DevOps culture. Research can be further extended to cover either remaining tools or other dimensions of selection criteria for source code management phase in DevOps culture. Figure 1. Schematic flow of tasks performed in comparing and evaluating SCM automation tools 2. RELATED WORK The selection/evaluation criteria, tasks, activities or practices considered by DevOps for the selection of appropriate source code automation tools have been reviewed with respect to different existing research papers. As Erich et al. [1], in their paper “A qualitative study of DevOps usage in practice,” highlighted that software industries are focusing on speeding up the development process and also improving the quality of their software by introducing the term DevOps as a portmanteau term to describe their efforts. The report details the DevOps implementation practices used by enterprises and the results they see as a result. In their article, Perera et al. [2] and Erich et al. [3] conducted research to determine how DevOps practices enhance software quality. They show through the use of multiple regression analysis that culture, automation, measurement, and sharing are crucial elements in raising the caliber of software. DevOps based software development was recommended to achieve high quality software. DevOps has been recognized by Lwakatare et al. [4] as a crucial component of the continuous deployment paradigm in academic research circles and practitioner communities. It was also characterized by collaboration, automation, and measurement and monitoring as four major contributors to the DevOps movement. Riungu-Kalliosaari et al. [5] collectively in their paper on DevOps confirms benefits of adopting DevOps by conducting qualitative multiple-case study. This study acknowledges that DevOps practitioners foster stronger departmental collaboration, which improves communication and worker well-being. Dwivedi et al. [6] and taking into account additional DevOps research that addresses software quality concurrently and examines the effects of DevOps features on software quality. This study provided a methodical mapping of the effect of DevOps on software quality. The primary research areas were DevOps automation, culture, continuous delivery, and quick feedback. In their works on DevOps, Olguin [7] and Jabbari et al. [8]
- 3. Int J Elec & Comp Eng ISSN: 2088-8708 Performance assessment and analysis of development and operations based … (Pooja Mittal) 1819 emphasize that the goal of the DevOps movement is to automate the processes of continuous delivery of new software updates while simultaneously ensuring their accuracy and dependability. The definition of DevOps has also been thoroughly reviewed in the literature, and Jabbari et al. [8] concur that DevOps expands the agility element in the software development paradigm. The literature also shows the development of several models for the guidance of industry or software practitioners for fruitful and successful implementation of DevOps in practice. Many models have been proposed in this context like the unicorn framework proposed by Trihinas and others [9], to overcome different challenges of DevOps through continuous releases. In a similar vein, the DevOps Research and Assessment model proposed by Forsgren et al. [10] discusses successful product delivery, and Hussaini [11] accepts the emerging DevOps paradigm as a response to the growing understanding of the gap between the development and operation teams functions of an organization, which he refers to as the “4 Cs” (communications, cooperation, culture, and collaboration). The “Wall of confusion” between these teams is also accepted by authors. Conflicting motives among people, processes, and technology/tools all contribute to this “Wall”. Thus, there is a need to improve the coordination between the Dev and Ops teams. For improving the effectiveness and efficiency of DevOps stakeholder’s interest, the concept was also described in [11]. As many renowned researchers restrict the adoption of DevOps in practice though there are multiples of theories that are against DevOps application and talk about its challenges and lack of performance measure [12]. For example, Leite et al. conducted a survey in their paper [13] and discussed different challenges in DevOps adoption. Other researches also force the compulsion of DevOps practices for the organization to move towards delivering higher performance and quality software [14]. The individual automation tool discussion that is conducted as the core component of this study work was missing from the literature review studied, mentioned above, or discussed thus far. At the source code management stage of software development, we also conducted a comparative analysis of automation tools to get the best out of many alternatives present in current market scenario. Underlying research has selected BitBucket, GitHub Actions; GitLab has been taken into consideration as per the tool market demand and also based on most commonly followed tools. 3. ALTERNATIVE AUTOMATION TOOLS EXISTENCE FOR DIFFERENT PHASES OF SOFTWRE DEVELOPMENT IN DevOps CULTURE Different set of tools exists to render automation in all software development phases of DevOps culture. These automation tools help development and operations teams to speed up the tasks of software release. Existence of different automation tools in DevOps is shown in Table 1. Table 1 includes alternative and most commonly followed automation tools at different stages of software development. These automation tools ease the tasks of development and operations teams but challenges exist to choose the best tool from the corresponding list of alternative tools. Tools chosen prior to development not only save development time but also reduce the cost of rework at the same time. Current research works around the coding phase of development to comparatively analyze the best SCM automation tool for the underlying phase. To conduct this work, three most commonly followed SCM tools have been taken are BitBucket, GitHub Actions, GitLab. These tools are compared to one another based on their needful characteristics and requirements of the business market. Results of the comparison helps to choose the best out of many available alternatives. Table 1. Alternative automation tools for different phases of software development in DevOps culture DevOps Phase Automation Tools Requirement management Confluence, BookStack, Notion Project planning Coding AgileCraft, Trello, Jira BitBucket, GitHub Actions, GitLab Build Teamcity, Jenkins, Bamboo Testing UFT, TestComplete, Selenium 4. SCM TOOLS FOR CODING PHASE OF DEVOPS SOFTWARE DEVELOPMENT CULTURE Primary aim of DevOps is to optimize the end-to-end delivery process of the software. For this purpose, DevOps employs different automation tools at each and every phase of development life cycle with the objective to cut down the manual efforts and pace the cycle time from requirement to software release. This optimization process starts from the requirement gathering/management phase where it is recommended to use agile practices and continuous prioritization of requirements. Once the development team has prioritized a subset of requirements in the form of stories, the development team picks the stories in their priority order and starts working on the software development.
- 4. ISSN: 2088-8708 Int J Elec & Comp Eng, Vol. 13, No. 2, April 2023: 1817-1826 1820 After the completion of the continuous prioritization phase, the next critical phase is coding. There are N numbers of developers involved in the coding phase where they all work on different requirements viz. new feature development, and enhancement of existing features or bug fixes. All these works go in parallel and code of all these developers bundled together to create artifacts like .jar, .war, .ear, .egg that can be deployed to the production environment. In coding or development phase, the most critical aspect is to resolve the issue of SCM. One needs to have an efficient SCM tool that can not only handle versioning of source code but is also enriched in features that make developers life easy along with handling complex code merge issues, well structuring of code, security, reporting and analytics capability, code scalability, and availability. For this purpose of best performer SCM tool selection step by step procedure followed by coding team is shown diagrammatically in Figure 2. As depicted in Figure 2, automation tool selection process for a project includes requirement identification of tool for the project, selection of the vendor for tool followed by computing weightage or evaluation scores for all alternative tools and finally choosing highest scorer or performer automation tool as the best suitable tool for the project. This paper works on the third step of the selection process that includes assigning weightage and computation of evaluation and overall scores afterwards selection of best performer tool based on several selection criteria. For this analysis purpose, we have taken three market leader tools as the representatives of SCM tools-BitBucket, GitHub Actions and GitLab. A tabular comparative analysis is performed for these source code management tools based on several selection criteria covered in the following section. Figure 2. Steps for project tool selection process from the set of available automation tools 5. SELECTION CRITERIA FOR THE ACQUISITION OF OPTIMAL SCM TOOLS The acquisition of the best SCM automation tool in coding phase of DevOps depends on several selection criteria that are taken into consideration. Based on these criteria or evaluation parameters, automation tools are compared to one another and the tool getting the highest score for those generic features/criteria is chosen for the particular project. For the better understanding of these criteria and comparative analysis, these selection criteria have been categorized into three different dimensions- development and architecture, code security and maintenance and support. These different selection criteria that are considered under these dimensions are also been tabulated in Table 2. Table 2. Categorization and description of different dimensions of selection criteria for automation tools Selection Criteria Description Development and Architecture Support for multiple Repo in Project Maximum number of repositories Access management Managing permissions at project level In built CI/CD support In built feature to build test, deploy code Feature request Addition of new features in SCM tool itself Issue board Single tool option for all manageable In-built code review Allow to contain multiple observable Multiple reviewers Authorization to multiple reviewers for same repository Code snippet Shareable and reusable piece of code Hooks Mechanism to get integration information Rest APIs GUI interface for interaction purpose Code Security Code encryption (In-Transit/At Rest) Code migrated to vendor cloud is encrypted Scanning repo for Password/Keys Scanning of repositories for password/keys Vulnerability scanning Ability to highlight vulnerability in the code Maintenance and Support Import code from other SCM codes Import code from other hosted repositories Multiple LDAP integration Integration with active directory In-premise tool integration Integration with other tools
- 5. Int J Elec & Comp Eng ISSN: 2088-8708 Performance assessment and analysis of development and operations based … (Pooja Mittal) 1821 There exist many selection criteria for tool excerption and for their crisp understanding we have categorized them into three pillars or dimensions as shown in Table 2. For this current research, we are targeting only development and architecture dimension of selection criteria that covers multiple of parameters like support for different repositories, access management, and issue board. All these parameters for selected criteria are explained beneath. 5.1. Support for multiple repositories in project Source code of any application can be organized in many different ways. Legacy way of managing the code is to logically divide the code based on application functionality, where all the code is packaged together in the form of artifacts and deployed at once. In this form of delivery, complete code will be deployed at once even if there is a change expected in any single file or feature. After the deployment of code, need arises for stringent management of deployed code to maintain the integrity, consistency, uniformity along with unambiguous features. This need advances towards the SCM Automation tool, a much more optimized way of managing the code, employed by DevOps culture in which application is divided into multiple components and each component is independent of each other and can even be deployed to production individually. Architecturally each component can be developed as a micro service where all micro service code is separate from each other and each micro service can be deployed to production independently without any impact on the availability of other services. In this type of model, during the deployment or release process, all the application would be available to users except only one feature of the application. In order to support this type of flexible and scalable architecture, each micro service code should be stored separately in their respective repository and each repo has its own CI/CD pipeline to support deployment of individual micro service as assigned in the Figure 3 that explains the steps to create a new repository in GitHub. Hence it becomes most important that SCM tools support multiple repositories within a single project where each repository can store and manage the code of each application component i.e., micro service. Figure 3. Creating new repository in GitHub source code automation tool [15] 5.2. Access management-project level/repository level/both With the increase in microservices culture, there is always a need for a separate code repository for each micro service. Each micro service has a predefined set of objectives and functionalities for delivery to different stakeholders out of which only a very small number of stakeholders need access to all microservices e.g. the solution architect of the complete software may require all Microservices. In order to manage these selection criteria well, there is a need for provision of both project level code and repository level access with different sets of roles viz. admin, read-only, read-write, and reviewer. 5.3. In-built CI/CD support While tools like BitBucket, GitLab and GitHub Actions are primarily used for source code management; there is a huge demand for additional features that can ease out many tasks of developers along with less maintenance overhead and requirement of low learning curve. One of these most demanding features in the list is in-built support for build and deployment, which can optimize the end-to-end delivery process up to much greater extent. After the code commit statement by the developer, this in-built support feature of the tool automatically kicks the build pipeline integrated with required repository to bundle the
- 6. ISSN: 2088-8708 Int J Elec & Comp Eng, Vol. 13, No. 2, April 2023: 1817-1826 1822 code into deployable artifacts, runs the unit test, followed by the execution of quality checks, automated functional test, performance test, and security test. Another CI/CD feature is also made available in the SCM tool itself that makes the overall DevOps tool stack simpler to use, maintain and scale. BitBucket pipeline, GitHub actions and GitLab CI are examples of SCM Tools having such features. 5.4. Feature request Few organizations take feature requests attributes available in the source code management tool itself to post the release date, backlog management, and prioritized list of features on the tool to make their product open source. These features also help in source code management. Figure 4 mention different steps to add backlog feature in automation tools as creation of issue, development and quality review. Most of organizations also use likes, votes, watchers attribute of issue to analyze which feature is awaited by most of the users. Usually, the community picks the most sought-after feature and fixes that for all. Thus, the organization looks for in-built support for feature request in the source code management tool. Figure 4. Steps for adding or managing backlog feature with ease in automation tools 5.5. Issue board Issue board is one of the critical features that companies are looking for in the source code management tool. Issues could be anything like a new feature request, and enhancement of an existing feature or a defect in the application. This feature becomes more important for smaller companies or startups where they can use a single tool for source code management as well as for taking issues regarding their requirement backlog management. Having an issue board within the source code management tool has multiple advantages like cost-saving, automatic traceability between source code and requirements including less training needs for the whole team. 5.6. In-built code review Another critical aspect is the code review that contains multiple observables to consider regarding product design. Examples of code review could be checking of code similarity with the design, correspondence of design with issued guidelines, confirming of the code optimization with application demands of performance, and response time. Modern SCM tools like BitBucket, GitHub Actions and GitLab provide code review features within their tool itself. In most of the systems it is called a pull request. Instead of directly committing the code to the main branch, developers raise a pull request which goes to one of the senior developers or architects of the team who reviews the code and approves the merge request of the code from feature or story branch to the main branch from where CI/CD pipeline will be kicked off. Having this feature already available with the source code management system saves developer time as they can see the review comments in the same tool. They can also fix the code directly and again raise the pull request if required. Now no need to run two different tools, no switching between tools required that again optimize the hand holding process. 5.7. Multiple reviewers In organizations there is usually more than one person who is authorized to review the code so that if one person is not available it should not delay the commit or the CI/CD process. Snapshot of authorizing multiple reviewers in one repository is represented in Figure 5. Ideally code review process should have the
- 7. Int J Elec & Comp Eng ISSN: 2088-8708 Performance assessment and analysis of development and operations based … (Pooja Mittal) 1823 flexibility to assign a review to multiple reviewers like senior developers or the architect, so that anyone who has bandwidth to do the code review can pick that review request and give the feedback with no dependency on a single reviewer. Figure 5. Adding of multiple reviewers for different branches of source code in management tools [16] 5.8. Code snippet Code snippet is a small piece of code that can easily be shared with anyone in the team. It is a reusable piece of code that we can make public so that other people can copy it or refer to it for common functionality. Generally, this code snippet feature is more important for companies who are doing open source projects or for a community that is supporting any open source project. This snippet of code can be stored in separate files from where it can be referred to. It also has the advantage of adding other people to put comments to allow them make any changes in the code to make it more improved and optimized. These Snippets could be a single file or a collection of files. 5.9. Hooks (pre-commit/post-commit/WebHooks) Hooks are the mechanism to create and validate information from other tools that are integrated with source code management tools. Generally, event-based, specific event triggers hooks do this work of validation for other tools. Hooks are of different types viz. pre-commit hook, post-commit or WebHooks. Consider a good example of a pre-commit hook that validates ticket ID in the commit message. Now if a developer is working on a story or defect, it is mandatory for the developer to mention the story/defect ID from the task management tool like Jira or Trello but if that issue ID is not mentioned in the commit message, then the source code management tool can block that commit. This block will not only help to maintain the automatic traceability from the requirement to source code but also help the developers to maintain the code of the live application. These hook features prevent the developers from introducing the regression defect or breaking the existing functionality of the application. In a similar way, post-commit hook helps to trigger any action in other tools after the commit. Examples could be resolving tickets in task management systems like Jira, and Trello. after the developer commits the code for that particular ticket or triggers the build in tools like Jenkins, Bamboo, and Teamcity. Tools from the same vendor are integrated implicitly by application links like BitBucket, JIRA, and Bamboo. Hence implementation of pre/post commit hooks is through native integration but for cross vendor tool integration like BitBucket-Teamcity, BitBucket-Jenkins needs WebHooks. 5.10. Rest application programming interface (APIs) Most of the DevOps tools expose both graphical user interface and APIs for interaction. Rest API is important to communicate the tool programmatically. Organizations create their own tool to extract data for analytical purposes. Not all the reports required by the organizations are available off the shelf by the tool. Thus, they need to write custom solutions to extract the data using rest APIs. This section clearly describes
- 8. ISSN: 2088-8708 Int J Elec & Comp Eng, Vol. 13, No. 2, April 2023: 1817-1826 1824 the need or usage of different selection criteria for development and architecture dimension. These criteria are taken into consideration before the actual selection of the automation tools. Other criteria given in Table 2 are also checked for select the best automation tool, but currently this paper works on parameters under development criteria. 6. PERFORMANCE ANALYSIS OF SOURCE CODE AUTOMATION TOOLS-A TABULAR COMPARISON BASED ON SELECTION CRITERIA Many alternative source code automation tools exist for managing code in DevOps culture. This research has taken BitBucket Cloud, GitHub Actions, and GitLab as representative tools for source code management based on their market analysis and common usage. These tools are assigned scores in 3-Steps: Firstly, a numerical value called weightage/weight is assigned to each selection criteria. Weights are assigned as per the merits or requirements of criteria to their target customers. Next step includes assigning an evaluation score to the SCM automation tool as per the selection criteria or corresponding feature availability in the tool and the last step multiplies the weightage with the corresponding evaluation score and finally adds them for each tool to get an overall score for that tool. These assigned and evaluated scores for all the representative tools are mentioned in a comparative tabular format as shown in Table 3. Table 3. Performance analysis table containing weightage assigned along with evaluation and overall score assigned to the representative SCM automation tool [17]–[25] Selection Criteria Description Evaluation Score (out of 10) Overall Score=Weightage * Evaluation Score Weigh- tage Bit-Bucket Cloud Git-Hub Actions Git- Lab Bit-Bucket Cloud Git-Hub Actions Git- Lab Dimension-Development and Architecture Support for multiple Repo in Project Maximum number of repositories which can be created in a single project/group 10 9 9 9 90 90 90 Access management- project level/ Repository Level/Both Managing permissions at project level as well as at repository level 10 7.5 7 [18] 8.5 75 70 85 In built CI/CD Support In built feature to build test, deploy code 8 6.5 [20] 8 8 52 64 64 Feature Request Ability to build features requests within SCM tools 8 7 8 8 56 64 64 Issue Board In built features to manage tasks with help of Kanban and scrum boards 6 5 8 7.5 30 48 45 In-Built Code Review Ability to perform code review within the tool 9 8.5 8.5 9 76.5 76.5 81 Multiple Reviewers Ability to add multiple reviewers for a single review/pull request 9 8 8.5 9 72 76.5 81 Code Snippet Snippets allow to share files with yourself, members of your workplace 7 7.8 9 7.5 54.6 63 52.5 Hooks [Pre- Commit/Post- Commit/WebHooks] Ability to enforce hooks/merge check functionalities 10 7.5 7.5 8.5 75 75 85 Rest API usage matrix and other data available as Rest API call 9 7.5 8.5 9 [18] 67.5 76.5 81 Best In-Class reporting and analytics In built report for all below mentioned items 0 0 0 Most commit by persons For a specific repository who is the active contributor 7 7 8 7.5 49 56 52.5 Most active Project/Repo Most active repository of a project based on commits 7 0 0 0 0 0 0 Inactive Repo/Branch Inactive Repositories and Branch 7 0 0 0 0 0 0 Commit Frequency Frequency of commits for a repository 7 7 7.5 8.5 49 52.5 59.5 Merge/Pull request Pending/Overdue Pull request dashboard for repository/Project 9 7.5 9 8.5 67.5 81 76.5 814.1 893 917 Analysis Table 3 is the result of implementation and evaluation with respect to all mentioned features with SCM tools-BitBucket, GitLab, and GitHub for TekMentors Consulting Pvt Ltd. Software company and renowned researchers already referred to in the paper. Tool recommended is also being used in companies for the efficient management of source code. Based on the overall score of all the representative SCM tools, GitLab can be concluded as the best performer tool among others.
- 9. Int J Elec & Comp Eng ISSN: 2088-8708 Performance assessment and analysis of development and operations based … (Pooja Mittal) 1825 Several different functionalities or features available in GitLab, as seen in the performance analysis table, makes it the most commonly observed automation tool for source code management. This can further be analyzed from interest over time report for previous 6-7 years according to Stack Overflow report. Among these representative SCM tools, again GitLab shows the highest usage trend in last years based on stack overflow report as depicted in the Figure 6. Figure 6. SCM Automation Tools interest over time depicted by stack overflow latest trends report over past 7-8 years [26] 7. CONCLUSION AND DISCUSSION This paper discussed many tasks or activities as the selection criteria to choose the best performing source code automation tool for the coding phase of DevOps software development culture. As there are a number of selection criteria in the market and these were not narrowed till, so this was our idea to first categorize them into three dimensions in order to smoothen out the selection process of automation tools. For this work, we have only focused on the development and architecture dimension of selection criteria. Opting on these criteria not only helps to manage the source code with the selection of the right set of automation tools but at the same time it also escalates the development process up to a much greater extent. This research computes the performance of the representative tools on the basis of weightage assigned, evaluation and overall scores along with the case study of TekMentors Pvt Ltd. The analysis performed reveals GitLab as the best source code management tool depending upon the outcome of overall score. GitLab is selected as the best performer tool on the basis of several selection criteria that are generic in nature. Other alternative automation tools for source code management can also be chosen as per the requirements of project-specific criteria. This performance or comparative analysis of source code automation tools will be useful for students/researchers to learn in-depth about DevOps culture and usage of alternative automation tools at different stages of software development. A performance evaluation or comparative analysis of other automation tools may also be carried as a part of further research. For future scope, our work can also be extended to carry out research on other two dimensions of selection criteria to complete the process of code management automation tool selection in the software development process. REFERENCES [1] F. M. A. Erich, C. Amrit, and M. Daneva, “A qualitative study of DevOps usage in practice,” Journal of Software: Evolution and Process, vol. 29, no. 6, Jun. 2017, doi: 10.1002/smr.1885. [2] P. Perera, R. Silva, and I. Perera, “Improve software quality through practicing DevOps,” in 2017 Seventeenth International Conference on Advances in ICT for Emerging Regions (ICTer), Sep. 2017, pp. 1–6, doi: 10.1109/ICTER.2017.8257807. [3] F. Erich, C. Amrit, and M. Daneva, “A mapping study on cooperation between information system development and operations,” in Product-Focused Software Process Improvement, Springer International Publishing, 2014, pp. 277–280. [4] L. E. Lwakatare, P. Kuvaja, and M. Oivo, “Dimensions of DevOps,” in Lecture Notes in Business Information Processing, Springer International Publishing, 2015, pp. 212–217. [5] L. Riungu-Kalliosaari, S. Mäkinen, L. E. Lwakatare, J. Tiihonen, and T. Männistö, “DevOps adoption benefits and challenges in practice: a case study,” in Product-Focused Software Process Improvement, Springer International Publishing, 2016, pp. 590–597. Year
- 10. ISSN: 2088-8708 Int J Elec & Comp Eng, Vol. 13, No. 2, April 2023: 1817-1826 1826 [6] A. Dwivedi, G. Srivastava, S. Dhar, and R. Singh, “A decentralized privacy-preserving healthcare blockchain for IoT,” Sensors, vol. 19, no. 2, Jan. 2019, doi: 10.3390/s19020326. [7] R. J. Olguín, “DevOps challenges and implications,” University of Murcia, Murcia, Spain, pp. 1–5, 2019. [8] R. Jabbari, N. bin Ali, K. Petersen, and B. Tanveer, “What is DevOps?,” in Proceedings of the Scientific Workshop Proceedings of XP2016, May 2016, pp. 1–11, doi: 10.1145/2962695.2962707. [9] D. Trihinas, A. Tryfonos, M. D. Dikaiakos, and G. Pallis, “DevOps as a service: pushing the boundaries of microservice adoption,” IEEE Internet Computing, vol. 22, no. 3, pp. 65–71, May 2018, doi: 10.1109/MIC.2018.032501519. [10] N. Forsgren, M. C. Tremblay, D. VanderMeer, and J. Humble, “DORA platform: DevOps assessment and benchmarking,” in Lecture Notes in Computer Science, Springer International Publishing, 2017, pp. 436–440. [11] S. W. Hussaini, “Strengthening harmonization of development (Dev) and operations (Ops) silos in IT environment through systems approach,” in 17th International IEEE Conference on Intelligent Transportation Systems (ITSC), Oct. 2014, pp. 178–183, doi: 10.1109/ITSC.2014.6957687. [12] A. A. Khan and M. Shameem, “Multicriteria decision‐making taxonomy for DevOps challenging factors using analytical hierarchy process,” Journal of Software: Evolution and Process, vol. 32, no. 10, Oct. 2020, doi: 10.1002/smr.2263. [13] L. Leite, C. Rocha, F. Kon, D. Milojicic, and P. Meirelles, “A survey of DevOps concepts and challenges,” ACM Computing Surveys, vol. 52, no. 6, pp. 1–35, Nov. 2020, doi: 10.1145/3359981. [14] R. Bolscher and M. Daneva, “Designing software architecture to support continuous delivery and DevOps: a systematic literature review,” in Proceedings of the 14th International Conference on Software Technologies, 2019, pp. 27–39, doi: 10.5220/0007837000270039. [15] “Documentation for BitBucket server 7.5,” Atlassian. 2020, Accessed: Sep. 15, 2022. [Online]. Available: https://confluence.atlassian.com/alldoc/files/278071997/1019389387/1/1597212077409/BitbucketServer-075.pdf. [16] R. Banerjee, “Backup blog, using the BitBucket API,” Rewind. 2021, Accessed: Sep. 15, 2022. [Online]. Available: https://www.backhub.co/blog/using-bitbucket-api. [17] A. Q. Gill, A. Loumish, I. Riyat, and S. Han, “DevOps for information management systems,” VINE Journal of Information and Knowledge Management Systems, vol. 48, no. 1, pp. 122–139, Feb. 2018, doi: 10.1108/VJIKMS-02-2017-0007. [18] S. Uzunbayir and K. Kurtel, “A review of source code management tools for continuous software development,” in 2018 3rd International Conference on Computer Science and Engineering (UBMK), Sep. 2018, pp. 414–419, doi: 10.1109/UBMK.2018.8566644. [19] S. Bokefode, R. N. Tendulkar, and R. R. Nakade, “Project management system like BitBucket (App),” International Research Journal of Engineering and Technology (IRJET), vol. 7, no. 6, pp. 5240–5243, 2020. [20] S. M. Mohammad, “Continuous integration and automation,” International Journal of Creative Research Thoughts (IJCRT), pp. 2320–2882, 2016. [21] B. Vasilescu, S. van Schuylenburg, J. Wulms, A. Serebrenik, and M. G. J. van den Brand, “Continuous integration in a social- coding world: empirical evidence from GitHub,” in 2014 IEEE International Conference on Software Maintenance and Evolution, Sep. 2014, pp. 401–405, doi: 10.1109/ICSME.2014.62. [22] L. E. Lwakatare, P. Kuvaja, and M. Oivo, “An exploratory study of DevOps extending the dimensions of DevOps with practices,” ICSEA, vol. 104, 2016. [23] H. L. Akshaya, J. Vidya, and K. Veena, “A basic introduction to DevOps tools,” International Journal of Computer Science and Information Technologies, vol. 6, no. 3, pp. 5–6, 2015. [24] M. Shahin, M. Ali Babar, and L. Zhu, “Continuous integration, delivery and deployment: a systematic review on approaches, tools, challenges and practices,” IEEE Access, vol. 5, pp. 3909–3943, 2017, doi: 10.1109/ACCESS.2017.2685629. [25] I. Sommerville, Software engineering, 9th Editio. Pearson, 2010. [26] “Stack overflow trends,” Stack Overflow. https://insights.stackoverflow.com/trends?tags=bitbucket%2Cgithub-actions%2Cgitlab (accessed Sep. 17, 2022). BIOGRAPHIES OF AUTHORS Pooja Mittal obtained her Ph.D. degree from Maharshi Dayanand University. Her area of research and specialization include data mining, data warehousing, and computer science. She had published more than 50 research papers in renowned International and National Journals and attended more than 30 Conferences. Currently she is working as Assistant Professor in the Department of Computer Science and Applications, Maharshi Dayanand University, Rohtak (Haryana). She can be contacted at mpoojamdu@gmail.com. Poonam Narang Research Scholar, pursuing Ph.D. from the Department of Computer Science and Applications, Maharshi Dayanand University, Rohtak, Haryana under the supervision of Respected Dr. Pooja Mittal (Research Guide and Second Author). Author’s Qualification is M.Phil. (CS), MCA. She had attended many National and International Conferences including Springer and IEEE and also published many research papers. She can be contacted at email: poonam.mehta20@gmail.com.