STATE OF THE ART SURVEY ON DSPL SECURITY CHALLENGES

International Journal of Computer Science & Engineering Survey (IJCSES) Vol.7, No.5, October 2016
DOI:10.5121/ijcses.2016.7501 1
STATE OF THE ART SURVEY ON DSPL SECURITY
CHALLENGES
Mohamed AMOUD, Ounsa ROUDIES
SIWeb Team - École Mohammadia d’Ingénieurs (EMI)
Mohammed V University in Rabat, Morocco
ABSTRACT
The Dynamic Software Product Line (DSPL) is becoming the system with high vulnerability and high
confidentiality in which the adaptive security is a challenging task and critical for it to operate. Adaptive
security is able to automatically select security mechanisms and their parameters at runtime in order to
preserve the required security level in a changing environment. This paper presents a literature review of
security adaptation approaches for DSPL, and evaluates them in terms of how well they support critical
security services and what level of adaptation they achieve. This work will be done following the Systematic
Review approach. Our results concluded that the research field of security approaches for DSPL is still
poor of methods and metrics for evaluating and comparing different techniques. The comparison reveals
that the existing adaptive security approaches widely cover the information gathering. However,
comparative approaches do not describe how to decide on a method for performing adaptive security DSPL
or how to provide knowledge input for adapting security. Therefore, these areas of research are promising.
Keywords
Dynamic Software Product Lines; DSPL; Security; Systematic Review; State of the Art
1. INTRODUCTION
For more than twenty years, companies have introduced product lines successfully to build
cheaper and faster software with top quality. Several experiences have demonstrated the benefits
of the adoption of a product line approach. Software product lines (SPL) have been used
successfully in industry for building families of systems of related products, maximizing reuse,
and exploiting their variable and configurable options. Software product line engineering (SPLE)
practices offer desirable characteristics such as reduced time-to-market, rapid product
development and more affordable development costs as a result of systematic representation of
the variabilities of a domain of discourse that leads to methodical reuse of software assets. Giant
companies reveal that the use of an approach to the SPL development can generate significant
quantitative and qualitative improvements in productivity and customer satisfaction. This practice
can also efficiently satisfy the current need for mass customization of software. Their growing
success is due to their ability to offer companies ways to exploit their software products
commonalities to achieve economies of production. The development lifecycle of a product line
consists of two main phases: domain engineering, which deals with the understanding and
formally modeling of the target domain; and application engineering that is concerned with the
configuration of a product line into one concrete product based on the preferences and
requirements of the stakeholders.
DSPL extend the concept of conventional SPL by enabling software-variant generation at runtime
and produce software capable of adapting to such fluctuations. In contrast with traditional SPLs,
DSPL bind variation points at runtime, when software is launched to adapt to the current
environment, as well as during operation to adapt to changes in the environment.
Building a product line that dynamically adapts itself to changing requirements implies a
deployment of the product configuration at runtime. It also means that the system requires
monitoring capabilities for detecting changes in the environment. As a response to these changes,
the system adapts by triggering a change in its configuration, providing context-relevant services
or meeting quality requirements. Dynamic software reconfiguration is concerned with changing
the application configuration at runtime after it has been deployed.

2
From the security point of view, dynamically changing DSPLs are a challenge, as static security
mechanisms are not able to offer an optimal security level for the varying situations. Moreover, it
is impossible at design-time to anticipate all situations in which a DSPL application will be
utilized. These challenges cause a need for self-adaptive security, which is able to select security
mechanisms and tune their parameters at runtime.
Currently, several security adaptation approaches exist. On the one hand, approaches concentrate
on adapting a particular security mechanism or supporting a specific security attribute. On the
other hand, some approaches are generic; that is, they support different attributes and
mechanisms. Hence, it is difficult to select the most suitable adaptation approach for different
usages. Moreover, it is difficult to know what research steps are needed in the future.
The objective of this paper is to give an overview of the state of the art in the adaptive security
issues for a DSPL by doing a Systematic Literature Review (SLR) on simple and clear question in
this regard. In particular, we identify and compare different security adaptation approaches for
DSPL, and evaluate them in terms of how well they support critical security services and what
level of adaptation they achieve.
In the section 2 we describe our method for conducting the review. Results are presented in
Section 3. Section 4 answers our questions. Finally, the Conclusion and future work Section close
the paper.
2. METHOD
The aim of this study is identifying and comparing different security adaptation approaches for
DSPL. We used guidelines proposed by Barbara Kitchenham [1] for performing our study. The
main steps are explained in the next parts of this section.
2.1. SYSTEMATIC LITERATURE REVIEWS (SLR)
Systematic Literature Reviews (SLR) is a rigorous method for assessing, reviewing and
aggregating research results. Unlike an ordinary literature review consisting of an annotated
bibliography, SLR analyzes existing literature with reference to specific research questions on a
topic of interest. Furthermore, it can be considered as much more effort prone than an ordinary
literature survey.
2.2. RESEARCH QUESTIONS
RQ1: What is the focus of research in adaptive security of DSPL?
RQ2: What are the claimed benefits of self-adaptive security in DSPL and what are the tradeoffs
implied by self-adaptive?
RQ3: how can DSPLs autonomously evaluate changes and threats in their environment in order to
adaptively reconfigure themselves?
RQ4: What are the limitations of the existing approaches, and interesting areas for future
research?
Regarding to RQ1, We were looking for researches and case studies need to get insight in the
research trends in adaptive security of DSPL, providing context for the study.
Regarding to RQ2, it was important for us to know the claims associated with adaptive security,
the evidence that exists for these claims and the tradeoffs implied by this adaptive security.
Regarding to RQ3, we want to know how applications with stringent safety requirements require
security mechanisms that reduce human intervention when DSPL features change.
The goal of RQ4 is to help deriving conclusions from the study.

3
2.3. RESEARCH PROCESS
Our search process for review was based on online searching in famous online databases which
are addressed as table1. Since these databases cover almost all major journals and conference
proceedings, manually review of journal was not required. Review has been carried on by mean
of search facilities in these databases and using appropriate logical expressions. In first stage, our
focus was on title and abstract of articles found in search process and select appropriate and
relevant studies. If there was any doubt, our decision was based on reviewing it at one glance.
Table 1. Studies Resource
Source Address
Scopus www.scopus.com
IEEE Xplore ieeexplore.ieee.org
ACM Digital Library Portal.acm.org
Springer Link www.springerlink.com
Science Direct www.sciencedirect.com
2.4. INCLUSION AND EXCLUSION CRITERIA
Our primary goal is to understand the claims and supporting evidence of adaptive security in
DSPL, we excluded papers about theoretical aspects, as well as surveys and roadmap papers. We
also excluded short papers of 1 or 2 pages.
There were some papers that were relevant to our study indirectly in our defined process. This
will strengthen our review, because all relevant documents were included and our review covered
sufficiently direct and indirect studies in this research.
All studies are assessed through a quality check, which is an inherent part of a thorough literature
study. Checking the originality and quality of the studies is important for data synthesis and
interpretation of results later on.
2.5. QUALITY ASSESSMENT
For assessing studies we defined the following questions:
QA1: Does study agree with existence of the focus of research in adaptive security for DSPL?
QA2: Does the security of DSPL recognise the need for adaptation?
QA3: Does study report any similar practices in the claimed benefits of adaptive security in DSPL
and the tradeoffs implied by self-adaptive security?
QA4: Does study report show how to face securely unexpected risks and activate appropriate
countermeasures to respond to new threats?
QA5: Is it possible to avoid specifying all the behaviours in advance for Autonomous and
Adaptive Security?
We scored questions as bellow:
QA1. Y (Yes) study explicitly agrees with existence of any objectives; P (Partially) study
implicitly agrees and N (No) study disagrees with existence of any objectives.
QA2. Y, study explicitly agrees with existence of any needs; P, study implicitly agrees and N,
there is no need for adaptation.
QA3. Y, the authors address one or more similar practices; P, some of the ones practices could be
tailored and customized in the second and N, there is no similar and adaptable practices in them.
QA4. Y, the authors report provide sufficient arguments; P, so not enough and N, there is no
argument.

4
QA5. Y, study addresses possibility to avoid specifying all the behaviours in advance for
Autonomous and Adaptive Security; P, study partly agrees (or implicitly) and N, there is no
possibility.
We defined Y=1, P=0.5 and N=0 or Unknown where information is not clearly specified. All
authors assessed every article and if there is no agreement in scoring, we discussed enough to
reach agreement.
We defined Y=1, P=0.5 and N=0 or Unknown where information is not clearly specified. All
authors assessed every article and if there is no agreement in scoring, we discussed enough to
reach agreement.
2.6. DATA COLLECTION
These data were extracted from each article:
• The full source and references
• The author(s) information and details
• Research issues
• Main ideas
All articles were reviewed and data was extracted and checked. This idea was chosen for better
consistency in reviewing all papers and improving quality of review.
2.7. DATA ANALYSIS
Our collect data was organized to address:
-Whether study agrees with existence of the objectives of adaptive security for DSPL or not?
(Addressing RQ1)
-Whether study agrees with existence of any needs for adaptation or no? (Addressing RQ1, RQ2)
-Whether study mentions similar practice/concept in either methods or no? (Addressing RQ3)
-Whether study provides sufficient arguments to face securely unexpected risks and activate
appropriate countermeasures or no? (Addressing RQ2 and RQ4)
-Whether study agrees with existence of possibility to avoid specifying all the behaviours in
advance for Autonomous and Adaptive Security or not? (Addressing QR5)
-Whether authors believe that this area is promising or no? (Addressing QR5)
3. RESULTS
In this section we explain results of our review.
3.1. SEARCH RESULTS
Table 2 shows the results of our selection procedure. In this table, results of searching in all
databases are provided, but, some of the studies were repeated in more than one online database,
so, final number of unique studies selected for our review was distinguished after elimination of
repeated articles. Final selected studies are listed in table 3.
3.2. QUALITY EVALUATION OF STUDIES
During this phase, we found that some of the selected articles discussing security in general or
only the SPL, but, they do not provide any valuable information to our research, so, we decided to
delete them from scope of our study. Assessment of each study was done by means of criteria
explained in section 2.4 and the scores for each of them are shown in table 4.

5
3.3. QUALITY FACTORS
For assessing results of our quality questions, we use average of total scores. This average is
useful for some questions, but it is not useful for some other. For instance, we cannot answer the
question about possibility of integration with average of scores because of the nature of the
question; instead, we use negative ideas for rejecting possibility.
Table 2. Results of Study Selection Procedure
Source Search Results Selected Studies
Scopus 91 7
IEEE Xplore 85 10
ACM Digital Library 22 5
Springer Link 36 7
Science Direct 04 1
Total 238 30
Repeated articles 12
Finally selected articles 18
Table 3. Selected Studies For Conducting Review
ID Title Authors Main Topic Year
S1
Strategies for Variability
Transformation at Run-time
O. Haugen
and al. [2]
the security of DSPL
recognises the need for
adaptation
2009
S2
Self-Adaptive Software:
Landscape and Research
Challenges
M. Salehie
and al. [3]
Self-protecting is the
capability of recovering from
their effects and detecting
security breaches
2009
S3
Security Requirements Engineering
Framework for Software Product
Lines
D. Mellado
and al. [4]
To describe a security
requirements engineering in
order to facilitate the
development of secure SPLs
2010
S4
A Security Requirements
Engineering Tool For Domain
Engineering In SPL
J.
Rodríguez
and al. [5]
how to provide automated
support through which to
facilitate the application of
the security quality
requirements engineering
process for SPL
2011
S5
Claims and Supporting Evidence
for Self-Adaptive Systems: A
Literature Study
D. Weyns
and al. [6]
Claims versus the tradeoffs
of adaptive security 2012
S6
Non-functional Properties in
Software Product Lines - taxonomy
for classification
M. Noorian
[7]
The adaptive security
attribute should be measured
at runtime.
2012
S7
Automated Planning for Feature
Model Configuration based on
S. Soltani
and al. [8]
Adaptive security is a non-
functional requirements in
2012

6
functional and non-functional
requirements
DSPL
S8
A Systematic Review of Model-
Driven Security
H. Nguyen
[9]
How to improve the
productivity of the
development process and
quality of the resulting secure
systems
2013
S9
Runtime Monitoring and Auditing
of self-adaptive systems
D. H.
Carmo and
al. [10]
How to avoid specifying all
the behaviours in advance for
Autonomous and Adaptive
Security
2013
S10
Comparison of Adaptive
Information Security Approaches
A. Evesti
and al. [11]
Limitations and prospects of
adaptive security
2013
S11
Architecture and Knowledge-
Driven Self-Adaptive Security in
smart space
A. Evesti
and al. [12]
Self-adaptive security as an
applicable solution to
anticipate all the possible
changes at design-time.
2013
S12
An overview of Dynamic Software
Product Line architectures and
techniques
R. Capilla
and al. [13]
Challenges and solutions are
necessary to support runtime
variability and adaptive
security mechanisms in
DSPL models and software
architectures.
2014
S13
A Systematic Survey of Self-
Protecting Software Systems
E. Yuan
and al. [14]
autonomic systems capable
of detecting and mitigating
security threats at runtime
2014
S14
Policy -Based Language for
Autonomous and Adaptive
Security
F. Cuppens
[15]
how to simultaneously
address both adaptive and
autonomy in DSPL
2014
S15
Dynamic Reconfiguration of
Security Policies in Wireless
Sensor Networks
Mónica
Pinto and
al. [16]
self-protection solution based
on the combination of
dynamic adaptation and
reconfiguration of security
2015
S16
Representing and Configuring
Security Variability in Software
Product Lines
V.
Myllärniem
i [17]
Security variability can be
represented and distinguished
as countermeasures
2015
S17
Security Systems Engineering
Approach in Evaluating
Commercial and Open Source
Software Products
Jesus
Abelarde
[18]
The amount of security
resources and time necessary
to accommodate proper
security evaluations is
underestimated.
2016
S18
TRUSTWORTHY VARIANT
DERIVATION WITH TRANSLATION
VALIDATION FOR SAFETY
CRITICAL PRODUCT LINES
J.
Almendros-
Jiménez
and al. [19]
Propose a general technique
of checking correctness
through translation validation
to automatically verify runs
of a variant derivation tool.
2016

7
Table 4. Quality Evaluation
4. DISCUSSION
In this part, the answers to our study questions will be discussed.
4.1. WHAT IS THE FOCUS OF RESEARCH IN ADAPTIVE SECURITY OF DSPL?
Most of the articles agree that there are the objectives of research in adaptive security for DSPL.
By reviewing them, it seems that this research focus is derived from the following concerns:
category of the study, subject, concrete focus and application domain. Overall, fifty eight percent
of the studies focus on one or more activities of adaptive security (monitoring, analyzing,
planning, execution), runtime models, multiple control loops and on reflection [10]- [19].
More than two thirds of the articles agree that the security of DSPL recognise the need for
adaptation in order to achieve the required security level. On the one hand, a survey by D. Weyns
et al. [6] reveals that the existing security approaches DSPL are not generic, but rather approaches
focus on specific security objectives. Furthermore, a study of Yuan et al. [14] compares over 30
self-protection approaches and shows that most existing approaches focus on the part of the
adaptive control loop, instead of covering the entire adaptation loop. In the adaptive security
approaches, such as: Self-Adaptive Security in smart space [12], Self-Adaptive Software [3],
Strategies for Variability Transformation at Run-time [2], A Security Requirements Engineering
Tool For Domain Engineering In SPL [5] and Runtime Monitoring and Auditing of self-adaptive
systems [10], Architectural Approach for Self-managing Security Services in [17]- [18], authors
notice that any of these approaches support all security objectives but concentrate on specific and
pre-selected objectives
4.2. WHAT ARE THE CLAIMED BENEFITS OF SELF-ADAPTIVE SECURITY IN DSPL AND
WHAT ARE THE TRADEOFFS IMPLIED BY SELF-ADAPTIVE?
Eight studies examine the claims versus the tradeoffs of adaptive security and clearly demonstrate
that the researches mainly report on claimed benefits, while little attention is given to the
Source QA1 QA2 QA3 QA4 QA5
1 Y P N Y Y
2 Y Y P Y Y
3 P Y Y P Y
4 P Y P Y P
5 P Y Y P Y
6 Y Y Y Y Y
7 N P P Y Y
8 P Y Y N Y
9 Y P P Y Y
10 P N P Y P
11 N P P Y P
12 P Y N Y P
13 N P P Y Y
14 P N Y Y Y
15 P Y N P Y
16 N P Y Y P
17 Y Y P P Y
18 Y Y Y Y P
Average 0,58 0,72 0,61 0,83 0,83

8
implications of adaptive security. It is remarkable that the efficiency/performance ratio is almost
the only quality attribute with a negative effect due to the adaptive security [11]. We also
evaluated the type of claims that have been made to the quality attributes and found that the
dominant demand is improving quality attributes of the software. The main reported tradeoff
implied by the adaptive security is the top performance [6]-[18].
4.3. HOW CAN DSPLS AUTONOMOUSLY EVALUATE CHANGES AND THREATS IN THEIR
ENVIRONMENT TO ADAPTIVELY RECONFIGURE THEMSELVES?
Most of the articles agree that it is possible to have an autonomous and adaptive security in DSPL
[10]-[15]-[19]. Mónica Pinto and al. [16] present an approach to building adaptive security at
runtime. They extend SPLs by adding the ability to automatically derive changed configurations
by monitoring the context, and to automatically reconfigure the security application while it is
running. The adaptation platform of this approach provides a conceptual model and reference
architecture for adaptive system. A. Evesti and al. [12] address SPL that allow mobile devices in
smart space to download software configurations on-demand. When a device enters a particular
context, the application provider service must deduce and create a variant for the device. As
devices enter a context, their unique capabilities must be discovered and dealt with efficiently and
correctly. D. H. Carmo and al. [10] present a new approach where they meet challenges in
adaptive security construction and execution by combining certain aspect oriented and model
driven techniques in order to deal with complexity through abstractions used both to specify the
dynamic variability at design time and to manage run time adaptations.
4.4. WHAT ARE THE LIMITATIONS OF THE EXISTING APPROACHES AND INTERESTING
AREAS FOR FUTURE RESEARCH?
The issue of security in SPL is long, but most solutions are based on the assumption that the SPL
is a closed environment. Given current trends, where the SPL is dynamic and open system, these
solutions are not sufficient to ensure the adaptive security [4]. Although there are researchers
working in this field and solutions are provided to be better, but the mechanism of adaptive
security for DSPL is not yet mature. In addition, the existing security solutions are based on the
features of the current DSPL; since DSPL reveals more and more new features that may be
supported in the future; the adaptive security mechanism has to be modernized and new security
issues have to be identified [1]-[9]-[11].
The majority of articles are optimistic about the potential prospects of this promising area. The
research should focus on the following areas: -1- Policy, model and design of the security
architecture -2- Securing the management and sharing of knowledge...etc.
5. CONCLUSION AND FUTURE WORK
The objective of this literature study was to summarize existing research on engineering self-
adaptive software systems and shed light on the claimed benefits and provided evidence of
adaptive security in DSPL. The study shows that the existing adaptive security approaches widely
cover the information gathering. However, comparative approaches do not describe how to decide
on a method for performing adaptive security DSPL or how to provide knowledge input for
adapting security. Therefore, these areas of research are promising.
Future research is still necessary to provide more efficient mechanisms able to manage the
dynamic and adaptive characteristics of security in DSPL and also to determine how security can
be deployed and redeployed automatically using variability mechanisms and multiple binding
times, reducing human intervention and effort required by engineering operations when certain
system features change. As an emerging topic, we expect that promising new research will bring
better and integrated a self-adaptive security solution for Mobile Devices based on the
combination of the MAPE- K reference model and DSPL approach.

9
REFERENCES
[1] B. Kitchenham, O. Pearl Brereton, David Budgen, Mark Turner, John Bailey and Stephen Linkman,”
Systematic literature reviews in software engineering- A Systematic literature reviews”, Keele
University 2008
[2] O. Haugen C. Cetina, X. Zhang, F. Fleurey, V. Pelechano : ‘’ Strategies for variability transformation
at run-time’’, SPLC '09 Proceedings of the 13th International Software Product Line Conference,
Pages 61-70, Carnegie Mellon University Pittsburgh, PA, USA, 2009
[3] M. Salehie and L. Tahvildari, “Self-adaptive software: Landscape and research challenges,” ACM
TAAS, vol. 4, 2009.
[4] D. Mellado, E. Fernández-Medina, M. Piattini: ‘’ Security Requirements Engineering Framework for
Software Product Lines’’, Information and Software Technology, 2010. Volume 52: p. 1094-1117.
Oct. 2010
[5] Jesús Rodríguez, Eduardo Fernández-Medina, Mario Piattini, Daniel Mellado:’’ A Security
Requirements Engineering Tool for Domain Engineering in Software Product Lines ’’, part of the
ESFINGE Project of the Ministry of Science and Innovation (Spain), 2011
[6] D. Weyns1, M. Usman Iftikhar, Sam Malek, J. Andersson, ‘’ Claims and Supporting Evidence for Self-
Adaptive Systems: A Literature Study’’, 2012 ICSE Workshop on SEAMS, Zurich, pp.89-98, 4-5
June 2012.
[7] M. Noorian, E. Bagheri, W. Du:’’ Non-functional Properties in Software Product Lines: A Taxonomy
for Classification’’, In Proceedings of "SEKE'12", Peges 663-667, 2012
[8] S. Soltani, M. Asadi, D. Gasevic, M. Hatala, E. Bagheri :’’ Automated planning for feature model
configuration based on functional and non-functional requirements’’, SPLC '12 Proceedings of the
16th International Software Product Line Conference - Volume 1 Pages 56-65 ACM New York, NY,
USA, 2012
[9] Phu H. Nguyen:’’ A Systematic Review of Model-Driven Security’’, Software Engineering Conference
(APSEC, 2013 20th Asia-Pacific (Volume: 1) , IEEE, Univ. of Luxembourg, Dec. 2013
[10] D. H. Carmo, Sergio T. Carvalho, Leonardo G. P. Murta, Orlando Loques, ‘’ Runtime Monitoring and
Auditing of Self-Adaptive Systems’’,8th IEEE International Conference on Global Software
Engineering (ICGSE), Brazil 2013.
[11] A. Evesti; E. Ovaska, ‘’ Comparison of Adaptive Information Security Approaches’’, ISRN Artificial
Intelligence, Article ID 482949, 18 pages. Volume 2013.
[12] A. Evesti. ‘’ Adaptive security in smart spaces’’. PhD's thesis, University of Oulu, on the 31st
of
January 2014.
[13] Capilla, R., et al., ‘’An overview of Dynamic Software Product Line architectures and techniques:
Observations from research and industry’’. J. Syst. Software, 2014,
[14] E. Yuan, N. Esfahani, S. Malek:’’ A Systematic Survey of Self-Protecting Software Systems’’, ACM
Transactions on Autonomous and Adaptive Systems (TAAS), Volume 8 Issue 4, New York, NY,
USA . 2014.
[15] F. Cuppens:’’ Policy -Based Language for Autonomous and Adaptive Security’’, the Concordia
Institute for Information Systems Engineering, Canada, Jan. 2014
[16] Mónica Pinto and al.,’’ Dynamic Reconfiguration of Security Policies in Wireless Sensor Networks’’,
Sensors 2015, 15, 5251-5280; doi:10.3390/s150305251, 2015
[17] Varvana Myllärniemi : ‘’Representing and Configuring Security Variability in Software Product
Lines’’, Proceedings of the 11th International ACM SIGSOFT Conference on Quality of Software
Architectures, pp: 1-10, ACM New York, NY, USA, 2015
[18] Jesus Abelarde, ‘’ Security Systems Engineering Approach in Evaluating Commercial and Open
Source Software Products’’, SANS Institute InfoSec Reading Room, January 25, 2016
[19] Jesús M. Almendros-Jiménez, Luis Iribarne, Jesús López-Fernández, Ángel Mora-Segura, ‘’
Trustworthy variant derivation with translation validation for safety critical product lines’’, Journal of
Logical and Algebraic Methods in Programming, Vol. 85, Issue 2, February 2016

STATE OF THE ART SURVEY ON DSPL SECURITY CHALLENGES

More Related Content

What's hot (18)

Viewers also liked (20)

Similar to STATE OF THE ART SURVEY ON DSPL SECURITY CHALLENGES (20)

Recently uploaded (20)

STATE OF THE ART SURVEY ON DSPL SECURITY CHALLENGES