Towards Self-Protective Multi-Cloud Applications
MUSA – a Holistic Framework to Support the Security-Intelligent Lifecycle
Management of Multi-Cloud Applications
Erk
uden Rios
1
, Eider Iturbe
1
, Leire Orue-Echevarria
1
, Massimiliano Rak
2
and Valentina Casola
3
1
TECNALIA, ICT-European Software Institute, Parque Tecnológico de Bizkaia, C/ Geldo Edificio 700, E-48160, Derio,
Spain
2
Dipartimento di Ingegneria Industriale e dell’Informazione, Seconda Università di Napoli, via Roma 29, Aversa (CE),
Italy
3
Dipartimento di Ingegneria Elettrica e delle Tecnologie dell’Informazione, da Università di Napoli, via claudio, Napoli,
Italy
Keywords: Multi-cloud, Security-by-design, Cloud SLAs, QoSec, Distributed Deployment, DevOps.
Abstract: The most challenging applications in heterogeneous cloud ecosystems are those that are able to maximise
the benefits of the combination of the cloud resources in use: multi-cloud applications. They have to deal
with the security of the individual components as well as with the overall application security including the
communications and the data flow between the components. In this paper we present a novel approach
currently in progress, the MUSA framework. The MUSA framework aims to support the security-intelligent
lifecycle management of distributed applications over heterogeneous cloud resources. The framework
includes security-by-design mechanisms to allow application self-protection at runtime, as well as methods
and tools for the integrated security assurance in both the engineering and operation of multi-cloud
applications. The MUSA framework leverages security-by-design, agile and DevOps approaches to enable
the security-aware development and operation of multi-cloud applications.
1 INTRODUCTION
Cloud computing is an emerging promising
paradigm for enabling new business models and
economies of scale based on on-demand
provisioning of IT resources (both hardware and
software) over a network as metered services, where
consumers are billed only for what they consume. A
recent IDC Cloud forecast shows that the investment
on public cloud services is expected to be more than
€77,287 million in 2017 (IDC Cloud research,
2013).
Nevertheless, enterprises consider security as the
#1 inhibitor to cloud adoptions (Waidner, 2009)
(Expert Group Report. European Commission,
2010). Companies are reluctant to adopt cloud
computing because of the difficulty in evaluating the
trade-off between cloud benefits and the additional
security risks and privacy issues it may bring. Most
concerns are related to data protection, regulations
compliance (Symantec, 2013) (Bitcurrent cloud
computing survey, 2011) and other issues due to
lack of insight (of controls and governance
processes) in the outsourcing of data and
applications: data confidentiality, trust on
aggregators, control over data and/or code location,
and resource assignment in multi-tenancy (Expert
Group Report. European Commission, 2010).
Businesses that want to exploit cloud computing
need to be vigilant in understanding the potential
privacy and security breaches in this new
environment (Hubbard & Sutton, 2010).
Secure cloud environments are even more
challenging today, since they are becoming more
and more complex in reference to the number of
cloud resource types that are available “as a
service”. Besides the traditional three service models
defined by the NIST (Mell & Grance, 2010) (IaaS,
PaaS and SaaS), new models are showing up such as
Network as a Service specified by the ITU-T or Data
as a Service defined in ISO/IEC 17826:2012
(ISO/IEC 17826:2012, 2012).
As the number of cloud models, cloud resources
and cloud service providers grow in the market, it
becomes theoretically easy (but not necessarily
technically) for the cloud consumer to deploy and
551
Rios E., Iturbe E., Orue-Echevarria L., Rak M. and Casola V..
Towards Self-Protective Multi-Cloud Applications - MUSA – a Holistic Framework to Support the Security-Intelligent Lifecycle Management of Multi-Cloud
Applications.
DOI: 10.5220/0005492905510558
In Proceedings of the 5th International Conference on Cloud Computing and Services Science (CLOSER-2015), pages 551-558
ISBN: 978-989-758-104-5
Copyright
c
2015 SCITEPRESS (Science and Technology Publications, Lda.)
use multiple cloud solutions at the same time in an
integrated way (Miller, 2013). This means that
despite the diverse characteristics of the cloud
resources such as own management APIs and own
service level offerings (both functional and security),
all need to be monitored and managed as an
integrated working entity.
The most challenging applications in
heterogeneous cloud ecosystems are those that are
able to maximise the benefits of the combination of
the cloud resources in use: multi-cloud applications.
For the context of this paper, a multi-cloud
application is understood as a distributed
application over heterogeneous cloud resources
whose components are deployed in different cloud
service providers and still they all work in an
integrated way and transparently for the end-user.
Multi-cloud application solutions have to deal
with the security of the individual components as
well as with the overall application security
including the communications and the data flow
between the components. Even if each of the cloud
service providers offered its own security controls,
the multi-cloud application has to ensure an
integrated security across the whole composition.
Therefore, the overall security depends on the
security properties of the application components,
which in turn depend on the security properties
offered by the cloud resources they exploit. For
instance, the database component in charge of
storing sensitive data cannot ensure a high
confidentiality if the cloud storage resource in which
it is deployed does not use strong encryption
algorithms. Consequently, the whole multi-cloud
application may be not sufficiently safe.
The paper is structured as follows. Section 2
describes the intended advances over the state of the
art. Section 3 explains the MUSA approach and
introduces the MUSA framework. Section 4 explains
the future validation of the framework in industrial
case studies. Finally, section 5 discusses the future
work.
2 STATE OF THE ART
As outlined in introduction, the main purpose of the
MUSA framework is to offer a solution to build
security-aware multi-cloud applications. This
research activity involves many different open
research aspects, among them we focus on the
following questions: How do we identify the security
requirements of a multi-cloud application? How can
we ensure security of a multi-cloud application even
when control over some of its components is not
granted?? How do we deploy a multi-cloud
application maintaining the promised security
features?
The following three subsections try to offer a
brief summary of the state of the art of the existing
replies to such questions.
2.1 Security-by-design in Multi-Cloud
Applications
Security by design (SbD) was first positioned by
Gartner (Kreizman & Robertson) and pointed out the
importance of incorporating security into the
enterprise architecture process since the beginning,
i.e. by including security requirements in the design
process. In addition, Gartner recently defined the
Runtime Application Self-Protection (RASP)
(Gartner) security concept which is a security
technology capable of controlling the application
execution and detecting and preventing real-time
attacks. The concept behind this idea would be that
the application itself is able to control and manage
security mechanisms embedded in the application or
which can be invoked as a service by the
application.
Security Control frameworks are widely adopted
tools used to identify the security controls required
to ensure the protection of an ICT system. A security
control is a safeguard or a countermeasure
prescribed to protect a system and meet a set of
defined security requirement. Control Frameworks
are a structured list of security controls that help a
security expert to select the checks to perform in
order to guarantee the respect of security
requirements of a given system. Example of such
Control Frameworks are the NIST Control
Framework (NIST 800-53r4, 2013) and the ISO/IEC
27001 (ISO/IEC 27001).
In Cloud environments such frameworks are of
limited use since they mostly miss specific cloud
related security controls. Nevertheless, several
attempts to address these issues have been made in
(NIST SP500, 2010), (Cloud Data Protection Cert,
2013), (Cloud Security Alliance, 2014).
2.2 Security Aware SLAs in
Multi-Cloud Applications
As stated above, in this paper, multi-cloud
applications are distributed applications that run
consuming cloud resources. Such an approach
implies that the multi-cloud application developer
and owner (i.e. the one that runs it and offers its
CLOSER2015-5thInternationalConferenceonCloudComputingandServicesScience
552
services) have no control over the real execution
environment of the application. This inhibits the
correct evaluation of the security controls.
The approach almost universally followed to
define guarantees for users of a service is the
introduction of Service Level Agreements (SLAs).
An SLA is a formal agreement between a service
provider and its end user that describes functional
and non-functional aspects of the provided target
service, together with clearly defined responsibilities
of the involved parties.
The most well-known machine-readable SLA
models are the Open Grid Forum’s Web Services
Agreement (WS-Agreement) (Hubbard & Sutton,
2010) and IBM’s Web Service Level Agreement
(WSLA) (VukoliĆ, 2010). The WS-Agreement
specification proposes a domain-independent and
standard way to create SLAs while its predecessor
WSLA seems to be deprecated.
SLAs appear as a successful method to guarantee
common Quality of Service parameters, like
availability and performance indicators. As stated in
many recent works, such as (Kandukuri, Paturi, &
Rakshit, 2009), in order to deal with security
requirements in the Cloud ecosystem, SLAs should
be actually used to define target service security
parameters.
Security Service Level Agreements (often named
SecLA), are recognized as a promising way to model
security issues between Cloud Service Providers and
their users. ENISA, in (Dekker & Hogben, 2011),
has also identified the importance of SecLAs in the
Cloud computing field, pointing out that, in many
circumstances, customers are not aware of many
acquired services security aspects.
As introduced in (Almorsy, Grundy, & Ibrahim,
2011) and in (Luna et al, 2013), the current dearth of
reasoning techniques on Security SLAs is preventing
the diffusion of these approaches in production
environments. Nevertheless, currently, many efforts
are being made to fill this gap. For example, in
(Luna et al, 2013), authors aim to outline techniques
to quantitatively reason about Cloud Security SLAs,
defining security metrics and a proof of concept
semi -automated framework in order to assess cloud
security of different providers.
Several European projects have worked or are
working in this subject focusing mainly on SecSLA
negotiation (SPECS Project, 2014), the creation of a
security-aware SLA based language and related
cloud security dependency model (CUMULUS
project) and on the accountability for cloud-based
services (A4Cloud Project, 2014).
2.3 Security Driven Dynamic
Deployment of Multi-Cloud
Applications
Multi-cloud applications have complex composition,
provisioning and deployment requirements, and the
application design becomes even more complex at
the time an additional aspect such as security enters
in the equation. Therefore, several initiatives are
running in order to support this type of activities.
CloudML (CloudML project, 2013) (Ferry et al,
2013) developed a domain-specific language to
support the specification of provisioning,
deployment and adaptation concerns related to
multi-cloud systems at design-time and their
enactment at runtime. CloudML’s background is
PIM4Cloud language, defined in REMICS project
(REMICS Consortium, 2012) (Ferry, Chauve,
Rossini, Morin, & Solberg, 2013).
Based on CloudML, different approaches
(ARTIST Consortium, 2013) (ModaClouds
consortium, 2013) (PaaSage Consortium, 2014) and
versions of CloudML have been recently released to
provide means to the design of cloud based
applications deployment. In this context where there
are multiple CloudML versions, a joint task force
has been started by MODAClouds, PaaSage and
ARTIST projects which goal is to define a unique
common CloudML specification (ARTIST
Consortium, 2013).
Another approach that can be followed includes
TOSCA (OASIS, 2013). The TOSCA specification
aims to enhance the portability of cloud applications
and services by using a language for defining both
the service components of distributed applications
and the service management interfaces (Antonescu,
Robinson, & Braun, 2012). This approach is
currently being followed by SeaClouds (Seaclouds
consortium, 2013).
3 MUSA APROACH:
THE MUSA FRAMEWORK
Multi-cloud solutions represent a new challenging
field in order to add value to overall cloud client
experience (VukoliĆ, 2010). In order to exploit
multi-clouds potentialities, different architectural
approaches can be adopted (Bohli et al, 2013):
(i) replication of applications, i.e. the same system
is deployed in more than one provider and
malicious attacks can be easily discovered
comparing operation results;
TowardsSelf-ProtectiveMulti-CloudApplications-MUSA-aHolisticFrameworktoSupporttheSecurity-Intelligent
LifecycleManagementofMulti-CloudApplications
553
(ii) partition of application system into tiers, that
allows to separate logic from data;
(iii) partition of application logic into fragments,
that obfuscates the overall application logic to
providers;
(iv) partition of application data into fragments,
that makes impossible to a single provider to
reconstruct data, safeguarding confidentiality.
MUSA aims at ensuring the security in all multi-
cloud environments including those that combine
multiple scenarios as described above. To this aim,
MUSA approach combines i) a preventive security
approach, promoting Security by Design practices in
the development and embedding security
mechanisms in the application, and ii) a reactive
security approach, monitoring application runtime to
mitigate security incidents, so multi-cloud
application providers can be informed and react to
them without losing end-user trust in the multi-cloud
application.
In order to ensure the preventive oriented
security to be embedded and aligned with reactive
security measures, MUSA supports an integrated
coordination of all phases in the application lifecycle
management.
3.1 The MUSA Framework
The MUSA framework presented in this paper is
intended to provide support the integration of the
security within the multi-cloud application lifecycle,
as illustrated in Figure 1. MUSA supports the first
phase of the multi-cloud application lifecycle, the
development phase, through the MUSA IDE, which
helps in both specifying the end user security
requirements and integrate such requirements in the
application development.
The MUSA Decision support tool and MUSA
Distributed deployment tool support the multi-cloud
application deployment phase, helping in the choice
of the cloud service provider and deployment of the
multi-cloud application deployment.
The MUSA security assurance platform (SaaS)
supports the last phase of the multi-cloud application
lifecycle (execution phase), monitoring the
application execution and, when needed, applying
correction actions to grant the security features.
The MUSA framework aims to define a set of
best practices and guidelines for the integrated
management of Security by Design mechanisms in
the lifecycle of multi-cloud secure applications,
based on DevOps and agile (AgileManifesto, 2001)
methodologies’ principles. The practices are
supported by the different automation tools provided
in the MUSA framework, which enable the
coordination between programming and deployment
infrastructure worlds, ensuring the continuous
alignment of multi-cloud application security
requirements specification (both at composition and
SLA levels), implementation, monitoring and
enforcement.
Figure 1: MUSA approach.
Following section describes the proposed tools
in more detail.
3.1.1 Multi-cloud Secure Applications
Design
For an effective design of multi-cloud secure
applications an integrated development environment
(IDE) is needed. To solve this requirement, MUSA
framework intends to deliver an IDE that allows the
design of application components taking into
account security requirements. This IDE will be
based on existing open-source solutions and will
include three main modules.
The first module is a security requirements
specification tool for multi-cloud applications,
taking into consideration multi-cloud SLA definition
and composition. The tool will allow expressing in
the multi-cloud application SLA the security
requirements (QoSec) together with functional and
business requirements (QoS). To this aim, the
needed components’ and cloud resources’ SLA
composition shall be computed.
The second module supports the design of the
breakdown of multi-cloud application into
components based on the combination of functional,
business and security properties that the multi-cloud
application should offer. The tool will be based on
existing standards such as CloudML (CloudML
project, 2013) and TOSCA (OASIS, 2013), and will
help application developers to design the
CLOSER2015-5thInternationalConferenceonCloudComputingandServicesScience
554
architecture and the components composition taking
into account the SLAs of the cloud resources in
which the components will be deployed;
The last tool has the goal of design the
provisioning and the deployment configuration of
the needed heterogeneous cloud resources at
multiple clouds layers (IaaS, PaaS).
Moreover, the MUSA IDE will include a set of
security libraries that, embedded in the multi-cloud
application, will enable the activation of security
mechanisms and security controls without modifying
the programming model. The security libraries aim
at proposing a non-intrusive approach to introduce
security in multi-cloud applications. These libraries
will be inserted into the multi-cloud application
components at design time and will be the
responsible for ensuring the overall security at
runtime. This software will include detection of non-
compliant behaviour and enforcement mechanisms
to be executed at runtime.
3.1.2 Multi-cloud Secure Applications
Deployment
Once the application has been successfully designed,
it has to be deployed. Deploying multi-cloud
applications on distributed and heterogeneous
resources encompasses several challenges that are
not addressed currently in the existing tools. To
solve this challenge, MUSA aims to provide a secure
multi-cloud deployment tool that offers a distributed
deployment service based on the dynamic selection
of the cloud service providers (CSPs) that match
with the application risk analysis, the subsequent
security requirements as well as functional and
business needs. In order to achieve it, the MUSA
framework bases its offering mainly over three
components. The first one is the cloud resource
categorization of CSPs based on the measures of the
security and functional properties at real time. The
second component, a decision support tool, allows
the selection of the cloud resources which
combination is compliant with the security and
functional requirements specified in the multi-cloud
application composite SLA, after a previous
simplified process of risk analysis. Finally, the third
component allows an automated deployment of the
multi-cloud secure application, distributing each of
the application components’ packages towards the
matched cloud resource.
3.1.3 Multi-cloud Secure Applications
Runtime
Monitoring multi-cloud applications at runtime
involves collecting metrics of QoS and QoSec
parameters of both the components of the
application and the cloud resources provisioned.
MUSA aims to provide a monitoring service capable
of collecting such measurements by using standard
APIs (if they are used by the cloud service providers
(CSP)), cloud interoperability frameworks such as
jclouds (Apache, 2012), or measures provided by
MUSA security embedded libraries.
Whenever an incident occurs, MUSA sends
alerts to notify the application provider about
detected security relevant incidents. Moreover,
MUSA send alerts when an application is in risk of
not fulfilling its SLA, and some preventive action
needs to be taken in order to keep security
parameters well counterweighted with performance
or within the margins specified in the SLA, e.g. a
redeployment of application components across a
different combination of cloud resources.
Finally, the enforcement service offered by
MUSA ensures that the multi-cloud application
respects the security requirements in its SLA.
All three services (monitoring, notification and
enforcement) are delivered in the form of a security
assurance platform, packaged as a SaaS product. The
MUSA SaaS security assurance platform
collaborates closely with the embedded libraries to
enforce the security protection of the multi-cloud
application user’s data, through mechanisms such as
authentication, authorisation, data encryption, data
location assurance, etc. when the cloud resources
used do not offer such mechanisms.
The MUSA SaaS will store monitored security
parameters over the cloud resources and
components, and manage the necessary notifications
and alerts, so as the multi-cloud application provider
can early react to possible security breaches.
Contract verification processes will require a
mapping between low-level resource metrics and
high-level security parameters of the cloud services.
This process will be done at runtime by MUSA
SaaS, which will provide real-time assessment
supported by complex processing of composed
measures of low-level metrics.
4 MUSA FRAMEWORK
VA L I D AT I O N
The economic viability, user acceptance and
practical usability of the MUSA framework is
expected to be validated through piloting the
solution in realistic industry environments
representing highly relevant services for the
TowardsSelf-ProtectiveMulti-CloudApplications-MUSA-aHolisticFrameworktoSupporttheSecurity-Intelligent
LifecycleManagementofMulti-CloudApplications
555
European economy: airline flight scheduling systems
and urban smart mobility services.
In the following, we summarize the research
challenges faced in both case studies.
4.1 Case Study A:
Airline Flight Scheduling
Multi-Cloud Application
For our first case study we have selected
NetLine/Sched product by Lufthansa Systems to
demonstrate how MUSA framework benefits the
integrated security management of this application
that exploits a number of heterogeneous cloud
resources.
The product NetLine/Sched supports all aspects
of flight schedule development and management.
In this case study, we are particularly interested
in researching on how to:
(i) allow NetLine/Sched developer declare the
options regarding data localisation (e.g.
location country of the files), data retention and
deletion, data integrity, confidentiality, access
control and availability, etc. and make possible
that such policies are embedded in the
application specification.
(ii) enable security properties are embedded into
the deployed application artefacts (security-by-
design) for their continuous control at
operation.
(iii) allow deployment into secure multi-cloud and
multi-provider environments.
(iv) provide automated security assurance,
supported by continuous monitoring,
enforcement and notification mechanisms.
(v) keep the NetLine/Sched operator informed
about the discrepancies and/or adapts to such
requirements even in those cases that a change
in the architecture or composition of the clouds
underneath is needed.
4.2 Case Study B:
Smart Mobility Multi-Cloud
Application
Our second case study is an urban smart mobility
multi-cloud application in Tampere city in Finland.
Tampere Region has almost half a million
inhabitants with a modal share of: 16% public
transport, 27% pedestrians and cyclists, 57% private
cars.
Tampere City Council has a number of services
exposed to allow companies and individual
developers to develop, test and productize own
traffic applications using public data. The services
can be publicly accessed via Intelligent Transport
Systems and Services (ITS) platform (Wikipedia
ITS, 2014), which includes the public transport
services APIs, other traffic related APIs, traffic data,
etc.
In this case study Tampere University (TUT)
will take the role of many entrepreneur citizens and
companies (generally SMEs) that create innovative
applications by combining freely available open
services and datasets in the Web to create business.
The multi-cloud application by TUT aims at
supporting the energy efficient and sustainable
multi-modal transit of Tampere citizens when
commuting from home to work and vice versa.
The major challenges for MUSA in this case
study are the following:
(i) Enhance security capabilities of innovative
services in transportation and public
infrastructure in Tampere.
(ii) Enable entrepreneurs and citizens willing to
develop innovative services based on IST
Factory to be able to easily integrate security-
intelligence into their applications through the
use of MUSA IDE.
(iii) Empower operators of the multi-cloud
applications that integrate IST Factory cloud-
based services to ensure security of data
storage and exchange at runtime through the
use of MUSA assurance tools.
(iv) Allow evaluating new service multi-cloud
deployment implications by checking service
dependencies on other network and cloud
resources.
5 FUTURE WORK
Application growth, rise in complexity and need for
interoperability create market opportunity for cloud
integrators and multi-cloud providers by offering
new capabilities in the existing complex cloud
landscape (North Bridge in partnership with
GigaOM Research, 2013).
Taking profit of this opportunity window,
MUSA aims at contributing to building up the
innovation capacity and technology excellence of
the European software and service industry by
proposing a solution to master the security-
intelligent lifecycle of multi-cloud applications
based on novel DevOps and security-by-design
approaches.
CLOSER2015-5thInternationalConferenceonCloudComputingandServicesScience
556
In this paper we have presented the MUSA
framework whose main goal is to support the
security-intelligent lifecycle management of multi-
cloud applications. There are a number of major
challenges in the path:
(i) Enable the security aware design of distributed
applications over heterogeneous cloud
resources.
(ii) Automatic discovery and decision support
system of combinations of cloud services that
best match the required balance between
security and functional properties.
(iii) Security assurance though continuous
monitoring and integrated methods in both
engineering and operation of multi-cloud
applications.
The MUSA project which will lead to the
development and validation of MUSA framework
was launched on January 2015 and will last 36
months. Future publications on the progress of the
framework are expected both online (www.musa-
project.eu) and in future papers.
ACKNOWLEDGEMENTS
The project leading to this paper has received
funding from the European Union’s Horizon 2020
research and innovation programme under grant
agreement No 644429.
REFERENCES
A4Cloud Project. (2014). Accountability For Cloud and
Other Future Internet Services. Retrieved from
Accountability For Cloud and Other Future Internet
Services.: www.a4cloud.eu/
AgileManifesto. (2001, February 17). Manifiesto for Agile
Development. Retrieved December 8, 2013, from
Manifiesto for Agile Development:
http://agilemanifesto.org/
Almorsy, M., Grundy, J., & Ibrahim, A. S. (2011).
Collaboration-based cloud computing security
management framework. IEEE International
Conference on Cloud Computing (CLOUD) (pp. 364-
371). IEEE.
Antonescu, A.-F., Robinson, P., & Braun, T. (2012).
Dynamic Topology Orchestration for Distributed
Cloud-Based Applications. NCCA, (pp. 116 - 223).
Apache. (2012). Apache jclouds. Retrieved April 2014,
from Apache jclouds: http://jclouds.apache.org/
ARTIST Consortium. (2012). ARTIST Projec . Retrieved
April 15th, 2014, from ARTIST Project:
http://www.artist-pro-ject.eu/
ARTIST Consortium. (2013, September). Deliverable
7.2.1. Cloud services modelling and performance
analysis framework. Retrieved April 2014, from
Deliverable 7.2.1. Cloud services modelling and
performance analysis framework: http://www.artist-
project.eu/sites/default/files/D7.2.1%20Cloud%20serv
ices%20modeling%20and%20performance%20analysi
s%20framework_M12_30092013.pdf.
ARTIST Consortium. (2013, September). Deliverable
D4.3.1 Dissemination report. Retrieved April 2014,
from Deliverable D4.3.1 Dissemination report:
http://www.artist-
project.eu/sites/default/files/D4.3.1%20Dissemination
%20report_M12_01102013.pdf.
Bitcurrent cloud computing survey. (2011). Bitcurrent
cloud computing survey 2011. Bitcurrent cloud
computing survey 2011.
Bohli, J. et al. (2013). Security and Privacy Enhancing
Multi-Cloud Architectures.
Cloud Security Alliance. (2014). Cloud Controls Matrix.
Retrieved April 2014, from Cloud Controls Matrix:
https://cloudsecurityalliance.org/research/ccm.
Cloud Data Protection Cert. (2013). Cloud Data Protection
Cert. Retrieved April 2014, from Cloud Data
Protection Cert: http://clouddataprotection.org/cert.
CloudML project. (2013). Model-based provisioning and
deployment of cloud based systems. CloudML project.
Retrieved April 2014, from Model-based provisioning
and deployment of cloud based systems. CloudML
project: http://cloudml.org.
CUMULUS project. (n.d.). Certification infrastrUcture for
MUlti-Layer cloUd Services. Retrieved from
Certification infrastrUcture for MUlti-Layer cloUd
Services: http://cumulus-project.eu/
Dekker, M., & Hogben, G. (2011). Survey and analysis of
security parameters in cloud SLAs across the
European public sector. Retrieved April 2014, from
Survey and analysis of security parameters in cloud
SLAs across the European public sector:
http://www.enisa.europa.eu/activities/Resilience-and-
CIIP/cloud-computing/survey-and-analysis-of-
security-parameters-in-cloud-slas-across-the-
european-public-sector.
Expert Group Report. European Commission, I. S. (2010).
The Future of Cloud Computing: Opportunities for
European Cloud Computing Beyond 2010.
Ferry, N. et al. (2013). Towards model-driven
provisioning, deployment, monitoring, and adaptation
of multi-cloud systems. CLOUD 2013: IEEE 6th
International Conference on Cloud Computing, (pp.
887-894).
Ferry, N., Chauve, F., Rossini, A., Morin, B., & Solberg,
A. (2013). Managing multi-cloud systems with the
CloudML framework. NordiCloud’13: 2nd Nordic
Symposium on Cloud Computing & Internet
Technologies. Oslo, Normay.
Gartner. (n.d.). Gartner IT Glossary - Runtime Application
Self-Protection (RASP). Retrieved April 2014, from
http://www.gartner.com/it-glossary/runtime-
TowardsSelf-ProtectiveMulti-CloudApplications-MUSA-aHolisticFrameworktoSupporttheSecurity-Intelligent
LifecycleManagementofMulti-CloudApplications
557
application-self-protection-rasp (Retrieved April
2014).
Hubbard, D., & Sutton, M. (2010). Top Threats to Cloud
Computing V1. 0. Cloud Secuirty Alliance.
IDC Cloud research. (2013, September). IDC Cloud
research. Retrieved March 2014, from IDC Cloud
research:
http://www.idc.com/getdoc.jsp?containerId=prUS2429
8013.
ISO/IEC 17826:2012. (2012). ISO/IEC 17826:2012
Information technology -- Cloud Data Management
Interface (CDMI).
ISO/IEC 27001. (n.d.). ISO/IEC 27001 Information
Technology – Security Techniques – Information
Security management Systems – requirements.
Kandukuri, B., Paturi, V. R., & Rakshit, A. (2009). Cloud
security issues. SCC'09. IEEE International
Conference on Services Computing, 2009., (pp. 517-
520).
Kreizman, G., & Robertson, B. (n.d.). Incorporating
Security into the Enterprise Architecture Process.
Retrieved April 2014, from Incorporating Security into
the Enterprise Architecture Process:
http://www.gartner.com/DisplayDocument?ref=g_sear
ch&id=488575.
Luna, J., et al. (2013). Negotiating and Brokering Cloud
Resources based on Security Level Agreements.
CLOSER 2013, (pp. 533-541).
Mell, P., & Grance, T. (2010). The NIST definition of
cloud computing. In ACM (Ed.), Communications of
the ACM, 53, no. 6, p. 50.
Miller, P. (2013, September). Sector RoadMap:
Multicloud management in 2013.
ModaClouds consortium. (2013, September). Deliverable
4.2.1 MODACloudML development – Initial version.
Retrieved April 2014, from Deliverable 4.2.1
MODACloudML development – Initial version:
http://www.modaclouds.eu/wp-
content/uploads/2012/09/MODAClouds_D4.2.1_MO
DACloudMLDevelopmentInitialVersion.pdf.
NIST 800-53r4. (2013). 291 NIST Security and Privacy
Controls for Federal Information Systems and
Organizations. Retrieved April 2014, from 291 NIST
Security and Privacy Controls for Federal Information
Systems and Organizations:
nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.S
P.800-53r4.pdf.
NIST SP500. (2010). 291 NIST Cloud Computing
Standards Roadmap. Retrieved April 2014, from 291
NIST Cloud Computing Standards Roadmap:
http://www.nist.gov/itl/cloud/upload/NIST_SP-500-
291_Version-2_2013_June18_FINAL.pdf.
North Bridge in partnership with GigaOM Research.
(2013). The future of cloud computing, 3rd annual
survey 2013. Retrieved March 2014, from The future
of cloud computing, 3rd annual survey 2013:
http://www.northbridge.com/2013-cloud-computing-
survey.
OASIS. (2013). Topology and Orchestration Specification
for Cloud Applications Standard. Retrieved April
2014, from TOSCA standard by OASIS: www.oasis-
open.org/committees/tc_home.php?wg_abbrev=tosca.
PaaSage Consortium. (2014, April 30). Deliverable
D2.1.2: CloudML Implementation Documentation
(First version). Retrieved from Deliverable D2.1.2:
CloudML Implementation Documentation (First
version):
http://www.paasage.eu/images/documents/paasage_d2
.1.2_final.pdf.
REMICS Consortium. (2012). Deliverable 4.1
PIM4Cloud. Retrieved March 2014, from Deliverable
4.1 PIM4Cloud:
http://www.remics.eu/system/files/REMICS_D4.1_V2
.0_LowResolution.pdf.
Seaclouds consortium. (2013). Seaclouds project.
Seamless adaptive multi-cloud management of
service-based applications. Retrieved from Seaclouds
project. Seamless adaptive multi-cloud management of
service-based applications: http://www.seaclouds-
project.eu/project.html.
SPECS Project. (2014). Secure Provisioning of Cloud
Services based on SLA management. Retrieved from
Secure Provisioning of Cloud Services based on SLA
management: http://specs-project.eu/
Symantec. (2013). Choosing a Cloud Hosting Provider
with Confidence. Retrieved April 2014, from
Choosing a Cloud Hosting Provider with Confidence:
http://www.itwhitepapers.com/content20287.
VukoliĆ, M. (2010). The Byzantine empire in the
intercloud. 41(3), 105-111.
Waidner, M. (2009, November). Cloud computing and
security. Lecture Univ. Stuttgart (November 2009).
Retrieved from Cloud computing and security. Lecture
Univ. Stuttgart (November 2009).
Wikipedia ITS. (2014). Intelligent Transport Systems and
Services (ITS) Factory Wiki. Retrieved April 2014,
from Intelligent Transport Systems and Services (ITS)
Factory Wiki:
http://wiki.itsfactory.fi/index.php/ITS_Factory_Develo
per_Wiki.
CLOSER2015-5thInternationalConferenceonCloudComputingandServicesScience
558