A Survey on Design of Online Judge System

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 01 | Jan 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 1166
A Survey on Design of Online Judge System
Dr. Arati A. Dandavate1 Sanket Kad2, Jayesh Palande3, Rushikesh Jadhav4, Prathamesh Wankar5
1Guide, Department of Computer Engineering, JSPM’s Jayawantrao Sawant College of Engineering,
2,3,4,5Student, Department of Computer Engineering, JSPM’s Jayawantrao Sawant College of Engineering, Pune
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract - Online judges are systems developed for judging
the source code submitted by users, which isnextcompiled and
tested in a homogenous environment. Due to its applications
online judges are becoming very popular and are being
developed for various purposes. Based on the objectives ofOJ’s
we classify them into systems that support organization of
competitive programming contests, enhance education and
recruitment processes, facilitate in solving of data mining
challenges, online compilers and development platforms
integrated as components of other custom systems. Thus, we
would like to implement an efficient system which will help
students to gain knowledge of data structure and
programming skills. The system includes user interface,
sandbox judging environment, functions for evaluatingscores
and updating reports into database. To prevent malicious
codes from damaging the system, the online judge system
utilizes sandbox, ensuring the safety of the system. The system
uses parallel execution and concurrency. These new features
make the system more impactful by giving high performance,
high reliability and high stability.
Key Words: Online Judge, Competitive Programming,
Contests, Programming Skills, OJ Architecture
1.INTRODUCTION
Programming is the heart of computer science, subject in
computer science and computing related courses, and there
are increasing demands on improving the experience,
motivation and efficiencyof programminglanguageteaching
and learning. In short Programming is the heart of the
Computer Science. Many students from the school and
colleges lacks in the programming which is required for the
computer science related courses and for trendy
technologies like artificial intelligence, machine learning,
data science and many more. To gain the knowledge of data
structure and programming skillsplatformslikeonlinejudge
systems are helpful for the students. But many students
cannot survive in this growing competition in competitive
programming.
There are some problems or system lacks some features
or system can be modify to make students to improve their
coding/programming skills.
 Most of the online judges doesn’t have the facilityof
teacher guidance.
 Students writes the poor code.
 Some systems don’t have plagiarism check or some
systems have plagiarism checking mechanism but
which is not adding value in OJ system.
 System doesn’t provide personalized feedback to
the students.
 Low interest in program learning and weak
initiative of some students. Some studentscontinue
their learning habits in high school, lack of the
ability of active learning, self-learning.
 Poor programming abilities of students which
reflect when they can’t meet the needs of relevant
courses and also during employers common
evaluation of computer science major and software
engineering major students in our university.
 Obvious polarization of students during actual
performance.
 Some students have low interest program learning,
lack self-learning ability and ability of active
learning.
To solve some of the problem mentioned above, many
online judges has implementedorproposed newtechniques,
for example. Instead of plagiarism which is not a good
practise to check the code, code similarity check is
performed on the code [1]. Also analysing source files,
personalized feedback, teaching adjustment are there.
2. METHODOLOGIES
System Design and Sandbox environment
System implemented by is based on B/S
architecture. It has two services Web part and Sandbox
environment. System is based on micro servicearchitecture.
For the faster development of the system’s UI, they have
used Spring MVC/Spring Boot which is best to use [1].
User is expected to submit thecode/programwhich
is next compiled and executed in the system.Submittedcode
might be a malicious which can bring the system down by
large resources, making the compile time more, infinite
looping, and many more. To avoid malicious code from
damaging system Sandbox Models are used [3] provides
security sandbox design which is built in JAVA [1].

[1] This paper has shown full system architecture of system
with key technologies which are listed as follows.
Parallel Judgement [1]– Multipletestcasesofa program runs
on multiple thread without mutual interference.
Splot input/output stream [1] – To avoid the problem of
concurrency due to I/O streams, system had implemented
std io stream agent which will handle input/output
concurrency problem.
Signal interaction mechanisms [1] – This Signal interaction
mechanism provides information interaction between web
and sandbox side.
Message box mechanism[1] – Will takecareofasynchronous
behavior of getting result and synchronous behavior of
submitting requests.
Sandbox connecting manager[1] – Itmanagesthesandboxes
and achieve the load balancing between sandboxes.
Sandbox Business Manager [1] – It handles and checks the
state of the sandbox.
Docker Container - Sandbox Environment
Docker Container is standard unit of software, for running
quickly and reliably it packages up all the dependencies and
the code. Docker Container includes everything as it
becomes independent to run an application, it is standalone,
lightweight software it includes code, runtime, systemtools,
libraries and settings [4]. Malicious code cannot harm the
system due to isolation and the security to the resources
provided by Docker Container. Operating System layers
provides virtualization that’s why container does not need
the OS overhead while running.
Key Features
Fork the Solution; Fork lets you make changes to a solution
without affecting the original. You can fetch and use or
submit changes to the original and also give points as a
royalty.
If a participant has just solved a problem partially and has
submitted it, he will only get points on the basis of
effectiveness of that solution. But another participant can
use this partially solved solution as a boiler plate to build
their logic for the solution. In this way both of them will get
bonus points as both the solutions were effective to some
extent. This can help both participants to improve their
programming logic and together can build a moreoptimized
and more effective solution.
Test Case Generation: Existing technique of Custom input
field is not helpful as the size of testcases is very large. Basic
idea is to generate the testcases for better accuracy and
debugging. Contestant can write the program for the
generation of testcases as per the constraints of problem.
Output of this program will be the input for the solution of a
problem.
3. ADVANTAGES
Docker Container – It provides reliable and efficient
evaluation of a solution. Container does not use a lot of
memory in order to execute and test the correctness of the
program. After the evaluation container is destroyed and
resources are released.
Plagiarism check cannot be implemented successfully,that’s
why Code Similarity Check is the best option for the
Plagiarism [3].
Code Quality, Teaching Adjustment, Personalized feedback
helps the better evaluation and guidance [3].
4. APPLICATIONS
Competitive Programming: - Online judge platforms are
used to host competitive programming contests which can
help students to improve their skills in data structures and
develop programming logic. It also benefits students' career
preparation and advancements, ability to deliver results
under pressure, ability to self-assess and rate others work
and develop new ways of solving problems withhigh quality
work and time limitation.
Education [3]: - Programming ability cannot be developed
just by listening to teachers lectures but should be grasped
only with massive and repeated training. In traditional
education conductors assess students by grading their
homework manually. When it comes to online programming
education, only after they submit their programs and
programs have been judged by conductors will they get
feedback about their programs. It reduces students'
enthusiasm for programming training. Proposed system
returns feedback to users aftercompilingand runningresult.
Online Compiler: - Online compilers are debugging tools
which allows you to compile source code and execute it
online in a number of programming languages. The online
judge system can also be used as online compilers where
users can submit the source code in various programming
languages and can remotely compile and execute it via a
browser.
Recruitment: - Online judges are used as recruitment
platforms where freshers as well as experienced candidates
can be hired by the employers. Students can compete, build
their profiles, showcase their coding skills and based on
good results they can get hired. HackerEarth is an online
platform whose main objective ishiringtalenteddevelopers,
organizing hackathons, and hosting crowdsourcing-based
ideas.
Data-Mining services: - Online judge systems are used to
evaluate data-mining algorithms where users execute their
code locally using test data and submit the resultsgenerated

by the algorithm. MLcomp is an example of such an online
platform where it stores datasets, thus any user can upload
his/her own datasets and algorithms. It is a cloud-based
platform.
5. CONCLUSIONS
OJ systems provide a platform for students to master
programming ability and can automatically judge the
solutions submitted byusers inreal-timefeedback.Themain
focus of the system is on improving the judgment speed and
accuracy, performance optimization to database. The
implementations show that the system has high reliability,
efficiency, safety and good extensibility.
Docker container being a lightweight virtualization
technology, has advantages of high security, high startup
speed, faster development and deployment cycles. Our
proposed system is course oriented, combined with
advantages of OJ’s and Docker container to evaluate
submissions made by students with the methods such as
grading tool, a code similarity checker and a code quality
evaluation tool alsoallowsstudentstologinanytimethrough
the proposed experimental system and its implementation.
The system can exercise problems, edit, save, debug, run,
evaluate the source code of the program,gettimelyfeedback
and based on the result leaderboard is also displayed.
With the proposed system, students' interest and
enthusiasm in programming learning can be motivatedwith
the features such as leaderboard, teacher guidance and
competitive problems. The system greatly reduces the time
of guidance, review and feedback of students' programming
training, devotes more time and energy to organizing
teaching contents efficiently, designing program-training
topics, thus improving the teaching quality. Online judge
systems are ofinterestforeducational organizationsbecause
they support fully automatic and accurate evaluation of
student assignments, allowing the teacher to focus on
exercise quality and the teaching process. But this can also
increase the workload of teaching assistants.
REFERENCES
[1] Liang, Haohui & Chen, Chaojie & Zhong, Xiuyu & Chen,
Yuefeng. (2017). IOP Conference Series: Earth and
Environmental Science. 69. 012091. 10.1088/1755-
1315/69/1/012091.
[2] Han, Y., Zhang, Z., Yuan, B., Bi, H., Shahzad, M.N., & Liu, L.
(2019). An Experimental Online Judge System Based on
Docker Container for Learning and TeachingAssistance.
2019 IEEE Smart World, Ubiquitous Intelligence &
Computing, Advanced & Trusted Computing, Scalable
Computing & Communications, Cloud & Big Data
Computing, Internet of People and Smart City
Innovation
(SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI),
1462-1467.
[3] Zhou, Wenju; Pan, Yigong; Zhou, Yinghua; Sun,
Guangzhong (2018). Proceedings of ACM Turing
Celebration Conference - China on - TURC '18 - The
framework of a new online judge system for
programming education., (), 9–14.
doi:10.1145/3210713.3210721.
[4] [Online] : Available
https://www.docker.com

A Survey on Design of Online Judge System

More Related Content

Similar to A Survey on Design of Online Judge System (20)

More from IRJET Journal (20)

Recently uploaded (20)

A Survey on Design of Online Judge System