The Use of Serverless Technologies to Support Data Processing within the

Open Learning and Research Systems

Ihor A. Bezverbnyi

1 a

and Mariya P. Shyshkina

2,3 b

V. M. Glushkov Institute of Cybernetics of the National Academy of Sciences of Ukraine, 40 Academician Glushkov Ave.,

Kyiv, 03187, Ukraine

Institute for Digitalisation of Education of the National Academy of Educational Sciences of Ukraine, 9 M. Berlynskoho

Str., Kyiv, 04060, Ukraine

National University of Life and Environmental Sciences of Ukraine, 15 Heroyiv Oborony Str., 03041, Ukraine

Keywords:

Cloud Computing, Serverless Technologies, Open Science.

Abstract:

The article highlights the promising ways of providing access to cloud-based tools using serverless solutions

to support data processing within the open learning and research environment. It is emphasized that the

implementation of serverless technologies is a current trend in the development of modern ICT open learning

and research systems. The concept of the hybrid serverless cloud is considered. The analysis and evaluation

of the existing experience of using different types of cloud-based solutions to data processing support are

considered and evaluated. The example of the wave ﬁles processing using the lambda-function is examined.

The issues of integration of different services within the open systems of learning and research are covered.

A concept of the cloud-based open learning and research university environment involving the use of the

serverless cloud-based components is considered. The reasonable ways of tools selection are evaluated and

the prospects for their use within the cloud-based open learning and research environment are described.

1 INTRODUCTION

The formation and development of a cloud-based

learning and research environment, taking into ac-

count the principles of open science is an impor-

tant area of modernization of the educational pro-

cess in higher education, the leading trend in the de-

velopment of pedagogical systems of open educa-

tion within the European Research Area (ERA, 2015).

Thanks to the use of cloud technologies there is an

opportunity to build more convenient, ﬂexible, scal-

able systems for access to electronic resources and

services in the process of learning and research, cre-

ating conditions for teamwork with software applica-

tions along with the removal of geographical and time

constraints, providing mobility of all subjects (Bon-

darenko et al., 2019; Bykov and Shyshkina, 2018;

Bykov et al., 2020). This creates a basis for the imple-

mentation of the principles and technologies of open

science for a wider range of users, the creation and

operation of virtual research teams, improving sci-

entiﬁc communication processes, access to data in

https://orcid.org/0000-0001-5569-2700

the research process, implementation of their results,

interaction with society (www.fosteropenscience.eu,

2019). Cloud computing tools and services form

an information technology platform of the modern

educational and scientiﬁc environment, becoming a

network tool for the formation of this environment

(Bargmann, 2018; Hevko et al., 2021; Markova et al.,

2015; Mazorchuk et al., 2020; Munk et al., 2020;

Roberts and Chapin, 2017). Thus, it becomes rele-

vant to analyze trends in the implementation of cloud

data processing services into the activities of a scien-

tist and also the research or educational institution.

2 THE RESEARCH RESULTS

2.1 The Background Issues

Cloud computing in several kinds of available mod-

els, such as IaaS, PaaS and SaaS, plays an important

role in facilitating learning and research data process-

ing. Providing abstraction of resources and simple

automation tools, modern cloud platforms simplify

most routine tasks such as installation, maintenance,

Bezverbnyi, I. and Shyshkina, M.

The Use of Serverless Technologies to Support Data Processing within the Open Learning and Research Systems.

DOI: 10.5220/0010933200003364

In Proceedings of the 1st Symposium on Advances in Educational Technology (AET 2020) - Volume 2, pages 489-494

ISBN: 978-989-758-558-6

489

backup, security, and more (Bykov et al., 2020; Svet-

sky et al., 2017). Moreover, today, the concept of

open science deals with open data and big data pro-

cessing. To ﬁt the requirements of open science sys-

tems design large amounts of data are to be avail-

able for users access and joint processing. Thus the

cloud computing platforms may serve as a reasonable

framework to support open learning and research pro-

cesses both in terms of maintaining and processing a

large amount of data and also to make it available for

the community of scientists for joint processing, re-

trieving and evaluation (Bykov et al., 2020).

The computing capacity becomes crucial for large

amounts of data processing and retrieval, as this kind

of activity is needed at most stages the research pro-

cess, such as data collecting, representation, visual-

ization, analysis, interpretation and discussion. The

possible way to save resources and to provide ﬂexible

use of the cloud-based infrastructure is using lambda-

function within the serverless settings. This leads

to the notion of Function-As-A-Service (FAAS) as a

promising available cloud-based model (Roberts and

Chapin, 2017; Bargmann, 2018; Jonas et al., 2019;

van Eyk et al., 2018).

The serverless computing applications in different

areas and their estimation are among the most cur-

rent issues considered nowadays, for example for ma-

chine learning (Kurz, 2021), network functions vir-

tualization (Aditya et al., 2019), geospatial architec-

tures (Bebortta et al., 2020). Casale et al. (Casale

et al., 2020) propose the platform for decomposi-

tion and orchestration for serverless computing. Or-

tiz (Ortiz, 2019) present architecting serverless mi-

croservices on the cloud with AWS and also issues

of instructors training to use these technologies. Still

the area of educational application of serverless tech-

nologies to provide better use and implementation for

learning and research within the university sector is

poorly investigated and needs further research. There

is a need to consider methodological issues and possi-

ble ways of serverless technology application within

the open learning and research university environ-

ment.

The article aims to consider and evaluate a hy-

brid cloud-based serverless architecture as the pos-

sible open learning and research platform to sup-

port data processing and research collaboration. The

main idea is that design and development of learn-

ing and research environment due to the proposed ap-

proach will result in more efﬁcient use of the cloud-

based resources, better access to learning and research

data and collaboration support. The case study of

the sound signal processing as possible example of

serverless approach application for learning and re-

search is considered.

2.2 The Conceptual Basis

The conceptual and terminological body of investiga-

tion and the main principles of designing and develop-

ing university cloud-based learning and research envi-

ronment such as the principles of open science, open

education, as well as the speciﬁc principles inherent

in cloud-oriented systems are considered by Bykov

and Shyshkina (Bykov and Shyshkina, 2018).

The cloud-based learning and research environ-

ment (LRE) of a higher education institution is

the environment in which the virtualized computer-

technological (corporate or hybrid-based) infras-

tructure is purposefully built for the realization

of computer-procedural functions (such as content-

technological and information-communication func-

tions) (Bykov and Shyshkina, 2018).

Serverless technologies are used to build ap-

plications that are hardly predictable as for the

amount of the computer capacities necessary for

their processing. The serverless hybrid cloud archi-

tecture is designed to deploy the lambda-functions

(aws.amazon.com, 2019).

The lambda-function is a cloud-based service

model when the computing is fulﬁlled within the

cloud-based infrastructure of a provider on user de-

mand and the user needn’t create and manage the

server architecture.

2.3 The Model and Approach

In ﬁgure 1, the conﬁguration of the serverless appli-

cation architecture is shown.

The overall approach is to access Lambda-

function trough API Gateway, avoiding server man-

agement as Lambda-function returns the values into

the static HTML format, the data retrieved on S3-

basket, that may be outputted and processed.

In this case, a user refers to certain electronic re-

sources and a computing capacity set on a hybrid

serverless architecture from any device using the In-

ternet connection.

The advantage of the proposed approach is that, in

learning or research processes, it is necessary to use

computing capacities for special purposes that may

appear eventually due to the current need. In par-

ticular, in the course of the experimental research,

big data processing may be needed that require much

computer capacity for a short period. It may be re-

dundant to maintain and manage the cloud server for

these purposes. At the same time, there is a possi-

bility of designing special lambda functions so the

AET 2020 - Symposium on Advances in Educational Technology

490

Figure 1: The serverless application architecture (retrieved from https://app.cloudcraft.co/).

learner or researcher can access them via the Internet

and use the server with the powerful processing capa-

bilities without deploying it any time as the function

is needed. The necessary resources can be supplied

more ﬂexibly on demand.

2.4 Current Developments and

Implementation

The cloud-based LRE was implemented at the In-

stitute for Digitalisation of Education of the Na-

tional Academy of Educational Sciences of Ukraine

in the course of research projects and pedagogical

experiments conducted during 2012–2017. During

that period, cloud-based services for open education

and open science support were introduced in the re-

search and educational process (Bykov and Shyshk-

ina, 2018).

In 2018 the V4+ Academic Research Consortium

Integrating Databases, Robotics and Language Tech-

nologies was established, which aimed to address

regional issues related to EU ICT research priori-

ties. The BOX Cloud shared work-space – the shared

work-space for all partners was built on the IBM BOX

Cloud for storage and transfer of documents that net-

worked researchers’ computers. Virtual machine with

Windows 10 – this virtual machine is simply a shared

computer with Windows 10 in the form of a remote

desktop was used to support open learning and re-

search collaboration (Bykov et al., 2020).

The cloud-based components that had been elabo-

rated and tested within this period of research were

implemented in the learning process. The learning

course “Cloud Computing Technologies” was devel-

oped and introduced in National University of Life

and Environmental Sciences of Ukraine for train-

The Use of Serverless Technologies to Support Data Processing within the Open Learning and Research Systems

491

Figure 2: The result of the lambda-function processing the wave ﬁle.

ing computer science bachelors. The students were

trained to build cloud-based components on virtual

machines using AWS and Azure platforms. The

methodology of open learning and research platform

implementation proved to be useful.

The next step of the research was the creation

of the serverless hybrid cloud architecture to support

collaborative research with Kyiv Glushkov Institute

of Cybernetics of NAN of Ukraine. The script was

used for uploading and output of the sound signal os-

cillogram using Python programming language with

the library Matplotlib within the framework Flask.

The lambda-function was used to build the sound sig-

nal oscillogram and make the image (ﬁgure 2).

The serverless environment was used for the solv-

ing of tasks of wave rows analysis:

1. A Python-based Internet application tested on lo-

calhost using the Flask framework was created.

2. In the AWS console, a user account with nec-

essary permissions to protect future applications

was created. On this user account, an S3 bucket

and an EC2 server were created. On S3 the work-

ing folder with the script in Python (or another

working language for AWS Lambda might be

used) was upload.

3. To provide the processing, it was necessary to at-

tach one or more layers with the environment li-

AET 2020 - Symposium on Advances in Educational Technology

492

braries installed. The virtual environment with the

libraries was installed on the EC2 server. An ad-

ditional layer was formed from this environment.

Also, AWS Lambda may contain additional freely

distributable layers. They can also be included in

future applications.

4. The electronic table of available resources in the

YAML format was formed using CloudFormation

tool. The YAML script creates a separate role for

working with the future application.

4.1. Using the role, a Lambda-function was cre-

ated, its codes were downloaded from the zip

ﬁle created on the previous step in S3.

4.2. Using this role a Gateway API was created

allowing to call the Lambda-function from a

browser.

5. Debugging was fulﬁlled.

Using this sequence of steps the hybrid environ-

ment with lambda-function was created and tested.

Using the proposed architecture the problem of sound

signal processing was solved.

3 CONCLUSIONS AND

DISCUSSION

Pedagogically balanced and expedient introduction of

cloud technologies in the educational and research

process of higher education institutions, formation

and development of the learning and research envi-

ronment on this basis are factors of expanding ac-

cess to electronic educational resources, increasing

the effectiveness of ICT infrastructure. The use of

serverless technologies to provide cloud services of

data processing, visualization and retrieve is a rele-

vant and promising area of development and modern-

ization of the university open learning and research

environment.

This experience can be used for the development

of new cloud-based components for educational and

scientiﬁc purposes based on the proposed architecture

of the hybrid cloud-based environment with Lambda-

function.

This approach still needs further implementation

and evaluation.

REFERENCES

Aditya, P., Akkus, I. E., Beck, A., Chen, R.,

Hilt, V., Rimac, I., Satzke, K., and Stein, M.

(2019). Will Serverless Computing Revolution-

ize NFV? Proceedings of the IEEE, 107(4):667–

678. https://www.ruichuan.org/papers/serverless-

ieee19.pdf.

aws.amazon.com (2019). Building appli-

cations with serverless architectures.

https://aws.amazon.com/lambda/serverless-

architectures-learn-more/.

Bargmann, C. (2018). Serverless & FaaS. https:

//users.informatik.haw-hamburg.de/

∼

ubicomp/

projekte/master2018-gsem/Bargmann/folien.pdf.

Bebortta, S., Das, S. K., Kandpal, M., Barik, R. K., and

Dubey, H. (2020). Geospatial serverless comput-

ing: Architectures, tools and future directions. ISPRS

International Journal of Geo-Information, 9(5):311.

https://www.mdpi.com/2220-9964/9/5/311.

Bondarenko, O. V., Pakhomova, O. V., and Zaselskiy, V. I.

(2019). The use of cloud technologies when studying

geography by higher school students. CEUR Work-

shop Proceedings, 2433:377–390.

Bykov, V., Mikulowski, D., Moravcik, O., Svetsky, S., and

Shyshkina, M. (2020). The use of the cloud-based

open learning and research platform for collaboration

in virtual teams. Information Technologies and Learn-

ing Tools, 76(2):304–320. https://journal.iitta.gov.ua/

index.php/itlt/article/view/3706.

Bykov, V. Y. and Shyshkina, M. P. (2018). The con-

ceptual basis of the university cloud-based learning

and research environment formation and development

in view of the open science priorities. Information

Technologies and Learning Tools, 68(6):1–19. https:

//journal.iitta.gov.ua/index.php/itlt/article/view/2609.

Casale, G., Arta

c, M., van den Heuvel, W.-J., van Hoorn,

A., Jakovits, P., Leymann, F., Long, M., Papaniko-

laou, V., Presenza, D., Russo, A., Srirama, S. N.,

Tamburri, D. A., Wurster, M., and Zhu, L. (2020).

RADON: rational decomposition and orchestration

for serverless computing. SICS Software-Intensive

Cyber-Physical Systems, 35(1):77–87.

ERA (2015). European Research Area Roadmap 2015-

2020. https://era.gv.at/era/era-roadmap/european-era-

roadmap-2015-2020/.

Hevko, I. V., Lutsyk, I. B., Lutsyk, I. I., Potapchuk, O. I.,

and Borysov, V. V. (2021). Implementation of web

resources using cloud technologies to demonstrate and

organize students’ research work. Journal of Physics:

Conference Series, 1946(1):012019.

Jonas, E., Schleier-Smith, J., Sreekanti, V., Tsai, C.-

C., Khandelwal, A., Pu, Q., Shankar, V., Car-

reira, J. M., Krauth, K., Yadwadkar, N., Gonzalez,

J., Popa, R. A., Stoica, I., and Patterson, D. A.

(2019). Cloud Programming Simpliﬁed: A Berke-

ley View on Serverless Computing. Technical Report

UCB/EECS-2019-3, University of California, Berke-

ley. https://www2.eecs.berkeley.edu/Pubs/TechRpts/

2019/EECS-2019-3.pdf.

Kurz, M. S. (2021). Distributed double machine learn-

ing with a serverless architecture. In Companion of

the ACM/SPEC International Conference on Perfor-

mance Engineering, ICPE ’21, page 27–33. Associa-

tion for Computing Machinery, New York, NY, USA.

The Use of Serverless Technologies to Support Data Processing within the Open Learning and Research Systems

493

Markova, O. M., Semerikov, S. O., and Striuk, A. M.

(2015). The cloud technologies of learning: Ori-

gin. Information Technologies and Learning Tools,

46(2):29–44. https://journal.iitta.gov.ua/index.php/

itlt/article/view/1234.

Mazorchuk, M. S., Vakulenko, T. S., Bychko, A. O.,

Kuzminska, O. H., and Prokhorov, O. V. (2020).

Cloud technologies and learning analytics: Web ap-

plication for PISA results analysis and visualization.

CEUR Workshop Proceedings, 2879:484–494.

Munk, R., Marchant, D., and Vinter, B. (2020). Cloud en-

abling educational platforms with corc. CEUR Work-

shop Proceedings, 2879:438–457.

Ortiz, A. (2019). Architecting Serverless Microservices on

the Cloud with AWS. In Proceedings of the 50th ACM

Technical Symposium on Computer Science Educa-

tion, SIGCSE ’19, page 1240. Association for Com-

puting Machinery, New York, NY, USA.

Roberts, M. and Chapin, J. (2017). What Is Serverless?

O’Reilly Media, Inc.

Svetsky, S., Moravcik, O., and Tanuska, P. (2017). The

Knowledge Management IT Support: From Personal-

ized to Collaborative Approach. In Proceedings of the

14th International Conference on Intellectual Capital,

Knowledge Management & Organisational Learning,

pages 253–260. Academic Conferences and Publish-

ing International Limited.

van Eyk, E., Toader, L., Talluri, S., Versluis, L.,

a, A., and Iosup, A. (2018). Serverless

is More: From PaaS to Present Cloud Com-

puting. IEEE Internet Computing, 22(5):8–

17. https://michael.tsikerdekis.com/downloads/10.

1109.MIC.2017.265102442.pdf#page=9.

www.fosteropenscience.eu (2019). FOSTER.

https://www.fosteropenscience.eu/.

AET 2020 - Symposium on Advances in Educational Technology

494