Densifying the Sparse Cloud SimSaaS: The need of a Synergy among

Agent-directed Simulation, SimSaaS and HLA

Tiago Azevedo, Rosaldo J. F. Rossetti and Jorge G. Barbosa

Artiﬁcial Intelligence and Computer Science Lab, Department of Informatics Engineering

Faculty of Engineering, University of Porto, Porto, Portugal

Keywords:

Agent-directed Simulation, Agent-supported Simulation, HLA, High Level Architecture, Cloud, SimSaaS,

Simulation Software-as-a-service.

Abstract:

Modelling & Simulation (M&S) is broadly used in real scenarios where making physical modiﬁcations could

be highly expensive. With the so-called Simulation Software-as-a-Service (SimSaaS), researchers could take

advantage of the huge amount of resource that cloud computing provides. Even so, studying and analysing a

problem through simulation may need several simulation tools, hence raising interoperability issues. Having

this in mind, IEEE developed a standard for interoperability among simulators named High Level Architecture

(HLA). Moreover, the multi-agent system approach has become recognised as a convenient approach for

modelling and simulating complex systems. Despite all the recent works and acceptance of these technologies,

there is still a great lack of work regarding synergies among them. This paper shows by means of a literature

review this lack of work or, in other words, the sparse Cloud SimSaaS. The literature review and the resulting

taxonomy are the main contributions of this paper, as they provide a research agenda illustrating future research

opportunities and trends.

1 INTRODUCTION

Modelling & Simulation (M&S) is widely used in real

scenarios such as trafﬁc and transportation networks,

where making physical modiﬁcations could be highly

expensive, dependent on political decisions and very

disruptive to the environment. Its uses could be deci-

sion making and what-if analysis, performance opti-

misations, testing and training, making M&S method-

ologies a huge need for Universities and companies

worldwide.

Nowadays, there is a new paradigm called Sim-

ulation Software-as-a-Service (SimSaaS) where sim-

ulation software is used in the form of services,

thanks to the latest evolutions in cloud computing and

Software-as-a-Service (SaaS). Instead of having the

simulation software installed on their own computers,

researchers could take advantage of the huge amount

of resource that cloud computing provides.

Even so, studying and analysing a problem

through simulation may need several simulation tools,

with different resolutions and domain perspectives,

hence raising interoperability issues that could not be

trivial to solve.

IEEE has already a standard for interoperabil-

ity among simulators named High Level Architecture

(HLA). HLA is covered by many works in the litera-

ture. Moreover, the multi-agent system approach has

become recognised as a convenient approach for mod-

elling and simulating complex systems (Moya and

Tolk, 2007). Indeed, many researchers have devel-

oped work regarding agents.

This paper shows that the M&S community did

not make a complete jump from the simulations in the

local machines to the simulations in the cloud offered

in the form of services. To prove such an assertion,

we conducted a literature review to make the body

of knowledge of the current synergy among agent-

directed simulation, SimSaaS and HLA.

The literature review was conducted using the

methodological and systematic framework proposed

by (vom Brocke et al., 2009). It is not important

in the context of this paper to explicitly describe all

the phases. Yet, some considerations must be made.

The databases sources selected were Scopus, Engi-

neering Village and ACM. These databases are com-

monly known to contain vast work and have been

used by many researchers in software engineering.

The queries made to the databases sources were based

in four main keywords: SimSaaS, Cloud Computing,

172

Azevedo T., J. F. Rossetti R. and G. Barbosa J..

Densifying the Sparse Cloud SimSaaS: The need of a Synergy among Agent-directed Simulation, SimSaaS and HLA.

DOI: 10.5220/0005542801720177

In Proceedings of the 5th International Conference on Simulation and Modeling Methodologies, Technologies and Applications (SIMULTECH-2015),

pages 172-177

ISBN: 978-989-758-120-5

 2015 SCITEPRESS (Science and Technology Publications, Lda.)

HLA and agents. It is considered a time frame from

2004 to 2015. The evolution of knowledge and tech-

nology in the software engineering ﬁeld is tremen-

dous every year. Thus, a time frame of a decade seems

enough.

This paper will start to brieﬂy explain some pre-

liminary background concepts regarding the agent-

oriented paradigm, HLA and cloud for a better under-

standing of the scope of this work. After that, the re-

sults of the literature review are broadly indicated and

a taxonomy of the research work is presented. The

literature review and resulting taxonomy are the main

contributions of this paper, as they provide a research

agenda illustrating future research opportunities and

trends.

2 PRELIMINARY BACKGROUND

For the sake of clariﬁcation in future references, this

section will brieﬂy describe what is the agent-directed

simulation paradigm, the HLA standard and the cloud

paradigm.

Yilmaz and

Oren (Yilmaz and

Oren, 2007) in-

dicated that the agent-directed simulation paradigm

consists in three main areas: (1) simulation for agents

(simulation of agent systems, that is, the simula-

tion model is one or more agents), (2) agent-based

simulation (model behaviour generation or monitor-

ing of this process by using agents) and (3) agent-

supported simulation (improving simulation by using

agents as support facilities). There are several re-

searchers which consider agent simulation and agent-

based simulation the same principle as they do not

take into account the contribution of agents in model

generation. In this work, it is adopted the same per-

spective in which the two principles are seen as the

same.

In order to have a structural basis for interoper-

ability among simulators, IEEE developed HLA, a

software standard that provides a common techni-

cal architecture for distributed M&S. A federate is

the name given to every participant of the simula-

tion, whereas each one can interact within a federa-

tion. Communication between simulators is possible

thanks to a Run-Time Infrastructure (RTI). HLA base-

line components are: (1) Federate Interface Speciﬁca-

tion (IEEE, 2010c); (2) Framework and Rules (IEEE,

2010a); and (3) Object Model Template (OMT) Spec-

iﬁcation (IEEE, 2010b). The ﬁrst is a deﬁnition of

the services that each federate can use for communi-

cation. The second is a set of rules that ensure the

proper interaction within a federation. The latter is a

speciﬁcation of the format and syntax of the data that

is exchanged among federates.

Cloud computing is a fresh and on-going recent

buzzword where more and more work is being done

not only in the industry but also among academics.

Nonetheless, there is no general consensus on an un-

ambiguous deﬁnition (Geelan, 2009). A problem in

deﬁning cloud computing is that it overlaps with other

domains in distributed systems. Foster et al. (Foster

et al., 2008) deﬁne the ﬁelds of distributed systems

according to scale and domain (application-oriented

versus service-oriented). Web 2.0 covers the spec-

trum of service-oriented applications, opposing to the

Supercomputing and Cluster Computing, which have

been more focused on traditional local applications.

Cloud Computing lies at the large-scale side, being

more scalable than Grid Computing. Grid Computing

overlaps with all these ﬁelds, and because of that it is

normal to exist wrong deﬁnitions.

Despite the overlapping of cloud computing with

other domains, it is possible to distinguish it from

grid computing. Indeed, there are three aspects that

are new in cloud computing (Armbrust et al., 2010):

(1) the appearance of inﬁnite computing resources

available on demand; (2) the elimination of an up-

front commitment by cloud users; and (3) the abil-

ity to pay for use of computing resources on a short-

term basis as needed. Two new buzzwords emerged

from Cloud Computing, trying to extend it even fur-

ther: Fog Computing (Bonomi et al., 2012) and Cloud

2.0 (Miluzzo, 2014). Concluding, we adopt the deﬁ-

nition provided by The National Institute of Standards

and Technology (Mell and Grance, 2011): Cloud

computing is a model for enabling ubiquitous, con-

venient, on-demand network access to a shared pool

of conﬁgurable computing resources (e.g., networks,

servers, storage, applications, and services) that can

be rapidly provisioned and released with minimal

management effort or service provider interaction.

3 AN UNEXPLORED CLOUD

SimSaaS

This section will show the unexplored Cloud

SimSaaS by means of a literature review. It starts by

referring SimSaaS and Cloud generically, following

the HLA standard and the agent-directed simulation

topics.

In our research, it was clear that SimSaaS is a

very modern topic. The great majority of the papers

were published after 2011 and the ﬁrst ones do not

directly use the term SimSaaS, vaguely mentioning

simulation and web. Although there is an increasing

number of works per year, papers about SimSaaS are

DensifyingtheSparseCloudSimSaaS:TheneedofaSynergyamongAgent-directedSimulation,SimSaaSandHLA

173

not many. Nevertheless, they are wide concerning the

domains of application. It is possible to see works

in the biomedical domain (Sawicki et al., 2012), in

crowd and pedestrian ﬁeld (Wang and Wainer, 2015)

as well as works regarding ontology learning (Wang

and Wainer, 2014), trafﬁc and transportation (H

arri

et al., 2010), scheduling parallel discrete event simu-

lation jobs (Liu et al., 2012a) and a cloud simulation

in manufacturing (Taylor et al., 2014a), just to cite

some.

Beyond these speciﬁc domain works, there are

also some generic ones concerning frameworks for

development, for example (Tsai et al., 2011) (Guo

et al., 2011). Moreover, Cayirci refers to SimSaaS

using the term Modelling and Simulation-as-a-

service (MSaaS). He clariﬁes MSaaS, including top

threats (Cayirci, 2013c). He also talks about the no-

tions and relations of accountability, risk and trust

modelling (Cayirci, 2013a), as well as MSaaS com-

position in multi-datacenter or multi-cloud scenar-

ios (Cayirci, 2013b). Cayirci is not the only one using

the term MSaaS: (Siegfried et al., 2014) illustrate po-

tential beneﬁts that may be achieved by MSaaS and

challenges that remain to be solved.

Since cloud simulation started to be studied, an

overall picture took some time to arrive. So, in 2012

Liu et al. (Liu et al., 2012b) proposed a general ar-

chitecture of cloud simulation in the form of a SaaS

type cloud. Figure 1 summarises the main blocks of

this architecture. The simulation services which are

offered through a website to very different users, are

divided into three self-explanatory groups: Modelling

as a Service, Execution as a Service and Analysis as

a Service. All these services are possible due to the

physical and virtual resources in the bottom, as well

as the so-called Cloud Operating System which man-

ages and connects the baseline infrastructure to the

top.

Figure 1: The main blocks of the Cloud Simulation general

architecture (Liu et al., 2012b).

Although HLA has already been used in a variety

of works like agent-based simulations, there is almost

no clear references about interoperability among sim-

ulators when talking about SimSaaS. A ﬁrst approach

in extending HLA to support grid-wide distributed

simulation dates back to 2005 (Xie et al., 2005).

A very relevant work from 2012 discusses how

HLA can be integrated with Service Oriented Ar-

chitecture (SOA) in the context of a smart building

project (Dragoicea et al., 2012). The Simulation En-

gine Service is exposed by RESTful services. In-

side this Engine, there is also a RESTful API that ex-

poses access to the RTI’s federation management and

deals with the creation, initialisation, deletion, start-

ing, stopping, and execution of simulations. Still in

this context, the authors refer another paper (Wang

et al., 2008), where a comparison between HLA and

SOA concluded that:

• HLA has good interoperability, synchronization

and effective and uniform information exchange

mechanism between the communicating compo-

nents (federates), but lacks several features of web

services, such as: the integration of heterogeneous

resources, web-wide accessibility across ﬁrewall

boundaries;

• SOA beneﬁts from loose coupling, component

reuse and scalability but lacks a uniform data ex-

change format and time synchronization mecha-

nisms;

• The combination of HLA and SOA can extend the

capabilities of the two technologies and thus en-

able integrated simulated and real services.

Like HLA, the agent-directed simulation

paradigm is used in a huge variety of ﬁelds but,

when it comes to SimSaaS, few examples exists.

Nevertheless, already in 2006, it was mentioned the

importance of agents on simulation by exploring the

relationship of software agents to simulation and

games (Yilmaz et al., 2006).

Some authors (Tolk and Diallo, 2010)(Tolk et al.,

2011) defend that most of the current simulation in-

teroperability standards are inadequate as they just

support the federation by focusing on information ex-

change without providing the necessary introspective.

HLA provides more ﬂexibility as it only standardises

how to structure the data, not the exchanging of in-

formation. However, the focus remains on the infor-

mation exchanged within a system. Consequently, a

formal approach to simulation interoperability using

agent-supported simulation tries to solve this prob-

lem.

JAVA Agent DEvelopment Framework (JADE) is

a common software framework that simpliﬁes the im-

plementation of multi-agent systems. Web Services

Integration Gateway (WSIG) is an add-on for JADE

which performs two-way translations between service

SIMULTECH2015-5thInternationalConferenceonSimulationandModelingMethodologies,Technologiesand

Applications

174

requests and responses and JADE agent requests and

responses. Thanks to this, it was possible to design

a service-oriented simulation software framework as

part of a broader approach towards generating im-

proved levels of actionable views of situation aware-

ness (Shao and McGraw, 2009).

Shao and McGraw referred that the great beneﬁt

of using JADE as the underlying agent development

framework is that JADE agent entities can invoke

web service functionality hosted outside the JADE

run-time environment using normal JADE agent pro-

tocols, and that external entities can invoke JADE

agent functionality from outside the JADE environ-

ment using normal web service protocols. Although

the framework is very relevant, the applications were

not in the cloud nor in a grid.

A truly implementation of agents in the cloud

showed that agent-based M&S can beneﬁt a lot from

cloud computing, making it easier to have more accu-

rate and faster results, as well as timely experimenta-

tion and optimisation (Taylor et al., 2014b). Even so,

agent-based M&S in the cloud may be highly com-

plex due to the very different clouds, cloud middle-

wares and service approaches.

Federated simulation environments have some

limitations in supporting dynamic model and simulat-

ing updating, as it was pointed in 2004 and 2006 (Yil-

maz et al., 2006). An example is HLA federation

development, as it requires complete speciﬁcation of

object models and information exchanges before the

simulation begins. It was also argued that there is a

fundamental roadblock because of a lack of machine

processable formal annotations describing behaviour,

assumptions and obligations of federates.

4 TAXONOMY OF THE

RESEARCH WORK

SimSaaS is a trendy term which has been growing

considerably in recent years. Thus, it is the ideal time

to take advantage of this hype. However, there are

some concerns: there is a lack of automation and in-

tegration of tools in M&S (Wang and Wainer, 2015),

and research dissemination methods suffer as they

do not allow publishing simulation code and scripts

along with the published paper (Sliman et al., 2013).

Herewith, HLA is another term referenced a lot

in the literature since the ﬁrst complete version (HLA

1.3) was published in 1998, but once again, when it

comes to SimSaaS, almost nothing focuses on this and

there is few work regarding extension of HLA to al-

low simulation services in general and in the cloud.

Indeed, HLA solely has some disadvantages (Yilmaz

et al., 2006)(Tolk and Diallo, 2010).

In a 2014 panel about the future of research in

M&S (Yilmaz et al., 2014) it is referred as a future re-

search topic the distribution of SimSaaS in the cloud.

So, once more SimSaaS is still mentioned as an un-

explored area, now speciﬁcally in distributed simula-

tion.

Although a truly implementation of agents in the

cloud showed that agent-based M&S can beneﬁt from

cloud computing (Taylor et al., 2014b), there is a lack

of work putting together agents and cloud in order to

support SimSaaS.

Summing up all the discoveries of the described

literature review, it is possible to see a lot of gaps in

the literature concerning SimSaaS, SimSaaS in spe-

ciﬁc domains of application, SimSaaS in the cloud,

HLA in the cloud, solutions to HLA restrictions,

agents to support SimSaaS and agents in the cloud.

As the metaphor in the title of this paper tries to ad-

dress, SimSaaS in the cloud is currently too sparse

since it has so many gaps in research. It is necessary

to make it less sparse (densifying) in order to augment

the scientiﬁc and technological knowledge among re-

searchers in the ﬁeld.

Densifying the sparse cloud SimSaaS is not just

putting together cloud and SimSaaS, but also the syn-

ergies among them and agent-directed simulation and

HLA, which could bring so many advantages. Fog

Computing and Cloud 2.0, which were previously

mentioned, could also help in this evolutionary pro-

cess. Concluding, making these synergies a reality

will be the front research opportunities for the next

years.

A taxonomy of the research work could make the

gaps identiﬁed and the research agenda for the next

years more clear. Figure 2 illustrates the taxonomy

in the form of a Venn Diagram. Every work is about

M&S, more precisely SimSaaS. So, there are works

that simply mention SimSaaS. Then, inside SimSaaS

topic, research can focus on Cloud, HLA or Agent-

directed simulation. In the particular case of Agent-

directed simulation, there is a subset regarding Agent-

supported simulation. As some works can address

more than just one term, the representation in the form

of a Venn Diagram was chosen to illustrate these pos-

sible synergies.

The presented taxonomy could be used as a con-

ceptual framework for future developments. Re-

searchers should look to these opportunities in the sci-

entiﬁc community to orient their work, taking advan-

tage of the beneﬁts that this trends can give to their

daily investigations.

DensifyingtheSparseCloudSimSaaS:TheneedofaSynergyamongAgent-directedSimulation,SimSaaSandHLA

175

Cloud AdSim

AsSim

HLA

SimSaaS

M&S

(AdSim: Agent-directed Simulation, AsSim: Agent-supported Simulation)

Figure 2: Diagram representing the taxonomy of the re-

search work.

5 CONCLUSIONS

This paper started to brieﬂy expose important con-

cepts for a better understanding of its contents. Af-

ter that, a literature review is presented, focusing in

four distinct yet related topics: SimSaaS, Cloud Com-

puting, HLA and Agents. Finally, a taxonomy of the

research work is presented in the form of a Venn Dia-

gram for a clear visualisation about which topics can

(and should) have synergies among them. This tax-

onomy could be used as a conceptual framework for

future developments.

With the front research opportunities for the next

years and current research work identiﬁed, we hope

this paper can leverage the scientiﬁc activity in the

ﬁeld, with researchers actually ﬁnding it useful to

make the jump to the cloud. That jump will bring

advantages not only to each researcher in particular,

but also to the overall simulation scientiﬁc commu-

nity seeking for more knowledge.

ACKNOWLEDGEMENT

This work has been partially supported by MIEIC,

Faculty of Engineering, University of Porto.

REFERENCES

Armbrust, M., Fox, A., Grifﬁth, R., Joseph, A. D., Katz,

R., Konwinski, A., Lee, G., Patterson, D., Rabkin, A.,

Stoica, I., et al. (2010). A view of cloud computing.

Communications of the ACM, 53(4):50–58.

Bonomi, F., Milito, R., Zhu, J., and Addepalli, S. (2012).

Fog computing and its role in the internet of things. In

Proceedings of the ﬁrst edition of the MCC workshop

on Mobile cloud computing, pages 13–16. ACM.

Cayirci, E. (2013a). A joint trust and risk model for MSaaS

mashups. pages 1347–1358, Washington, DC.

Cayirci, E. (2013b). Conﬁguration schemes for modeling

and simulation as a service federation. Simulation,

89(11):1388–1399.

Cayirci, E. (2013c). Modeling and simulation as a cloud

service: A survey. pages 389–400, Washington, DC.

Dragoicea, M., Bucur, L., Tsai, W.-T., and Sarjoughian, H.

(2012). Integrating HLA and Service-Oriented Archi-

tecture in a Simulation Framework. In Proceedings

of the 2012 12th IEEE/ACM International Symposium

on Cluster, Cloud and Grid Computing (Ccgrid 2012),

CCGRID ’12, pages 861–866, Washington, DC, USA.

IEEE Computer Society.

Foster, I., Zhao, Y., Raicu, I., and Lu, S. (2008). Cloud

computing and grid computing 360-degree compared.

In Grid Computing Environments Workshop, 2008.

GCE’08, pages 1–10. IEEE.

Geelan, J. (2009). Twenty-one experts deﬁne cloud com-

puting. Cloud Computing Journal, 4:1–5.

Guo, S., Bai, F., and Hu, X. (2011). Simulation software as

a service and Service-Oriented simulation experiment.

pages 113–116, Las Vegas, NV.

arri, J., Killat, M., Tielert, T., Mittag, J., and Hartenstein,

H. b. (2010). DEMO: Simulation-as-a-service for ITS

applications. Taipei.

IEEE (2010a). IEEE Standard for Modeling and Simulation

(M &S) High Level Architecture (HLA)– Framework

and Rules. IEEE Std 1516-2010 (Revision of IEEE Std

1516-2000), pages 1–38.

IEEE (2010b). IEEE Standard for Modeling and Simula-

tion (M &S) High Level Architecture (HLA)– Ob-

ject Model Template (OMT) Speciﬁcation. IEEE

Std 1516.2-2010 (Revision of IEEE Std 1516.2-2000),

pages 1–110.

IEEE (2010c). IEEE Standard for Modeling and Simulation

(M&S) High Level Architecture (HLA)– Federate In-

terface Speciﬁcation. IEEE Std 1516.1-2010 (Revision

of IEEE Std 1516.1-2000), pages 1–378.

Liu, X., Qiu, X., Chen, B., He, Q., and Huang, K. (2012a).

Scheduling parallel discrete event simulation jobs in

the cloud. volume 2012, London.

Liu, X., Qiu, X., Chen, B., and Huang, K. (2012b).

Cloud-based simulation: The state-of-the-art com-

puter simulation paradigm. In Proceedings of the 2012

ACM/IEEE/SCS 26th Workshop on Principles of Ad-

vanced and Distributed Simulation, PADS ’12, pages

71–74, Zhangjiajie. IEEE Computer Society.

Mell, P. and Grance, T. (2011). The NIST deﬁnition of

cloud computing.

Miluzzo, E. (2014). I’m Cloud 2.0, and I’m Not Just a Data

Center. Internet Computing, IEEE, 18(3):73–77.

Moya, L. J. and Tolk, A. (2007). Towards a taxonomy of

agents and multi-agent systems. In Proceedings of

the 2007 spring simulation multiconference-Volume 2,

pages 11–18. Society for Computer Simulation Inter-

national.

SIMULTECH2015-5thInternationalConferenceonSimulationandModelingMethodologies,Technologiesand

Applications

176

Sawicki, B., Chaber, B., Starzy

nski, J., and Szmuro, R.

(2012). Internet application concept to trivialize EMF

biomedical computing. COMPEL - The International

Journal for Computation and Mathematics in Electri-

cal and Electronic Engineering, 31(4):1190–1197.

Shao, G. and McGraw, R. (2009). Service-oriented Sim-

ulations for Enhancing Situation Awareness. In Pro-

ceedings of the 2009 Spring Simulation Multiconfer-

ence, SpringSim ’09, pages 48:1–48:7, San Diego,

CA, USA. Society for Computer Simulation Interna-

tional.

Siegfried, R., Van Den Berg, T., Cramp, A., and Huiskamp,

W. (2014). M&S as a service: Expectations and chal-

lenges. pages 248–257. SISO - Simulation Interoper-

ability Standards Organization.

Sliman, L., Charroux, B., and Stroppa, Y. (2013). A new

collaborative and cloud based simulation as a service

platform: Towards a multidisciplinary research simu-

lation support. pages 611–616, Cambridge.

Taylor, S., Kiss, T., Terstyanszky, G., Kacsuk, P., and Fan-

tini, N. (2014a). Cloud computing for simulation

in manufacturing and engineering: Introducing the

CloudSME simulation platform. volume 46, pages

89–96, Tampa, FL. The Society for Modeling and

Simulation International.

Taylor, S. J. E., Anagnostou, A., Kiss, T., Terstyanszky, G.,

Kacsuk, P., and Fantini, N. (2014b). A Tutorial on

Cloud Computing for Agent-based Modeling & Sim-

ulation with Repast. In Proceedings of the 2014 Win-

ter Simulation Conference, WSC ’14, pages 192–206,

Piscataway, NJ, USA. IEEE Press.

Tolk, A. and Diallo, S. Y. (2010). Using a Formal Approach

to Simulation Interoperability to Specify Languages

for Ambassador Agents. In Proceedings of the Win-

ter Simulation Conference, WSC ’10, pages 359–370.

Winter Simulation Conference.

Tolk, A., Diallo, S. Y., Padilla, J. J., and Herencia-Zapana,

H. (2011). Model Theoretic Implications for Agent

Languages in Support of Interoperability and Com-

posability. In Proceedings of the Winter Simulation

Conference, WSC ’11, pages 309–320. Winter Simu-

lation Conference.

Tsai, W.-T., Li, W., Sarjoughian, H., and Shao, Q. (2011).

SimSaaS: Simulation Software-as-a-service. In Pro-

ceedings of the 44th Annual Simulation Symposium,

ANSS ’11, pages 77–86, San Diego, CA, USA. Soci-

ety for Computer Simulation International.

vom Brocke, J., Simons, A., Niehaves, B., Riemer, K., Plat-

tfaut, R., and Cleven, A. (2009). Reconstructing the

Giant: On the Importance of Rigour in Documenting

the Literature Search Process. In Proceedings of the

European Conference on Information Systems,

Verona, Italy.

Wang, S. and Wainer, G. (2014). Semantic Mashups for

Simulation As a Service with Tag Mining and Ontol-

ogy Learning. In Proceedings of the Symposium on

Theory of Modeling & Simulation - DEVS Integrative,

DEVS ’14, pages 25:1–25:8, San Diego, CA, USA.

Society for Computer Simulation International.

Wang, S. and Wainer, G. (2015). A Simulation As a Service

Methodology with Application for Crowd Modeling,

Simulation and Visualization. Simulation, 91(1):71–

95.

Wang, W., Yu, W., Li, Q., Wang, W., and Liu, X. (2008).

Service-oriented High Level Architecture. In Pro-

ceedings of the 2008 Summer Computer Simulation

Conference, SCSC ’08, pages 16:1–16:12, Vista, CA.

Society for Modeling & Simulation International.

Xie, Y., Teo, Y. M., Cai, W., and Turner, S. J. (2005). Ser-

vicing Provisioning for HLA-Based Distributed Sim-

ulation on the Grid. In Proceedings of the 19th Work-

shop on Principles of Advanced and Distributed Sim-

ulation, PADS ’05, pages 282–291, Washington, DC,

USA. IEEE Computer Society.

Yilmaz, L.,

Oren, T., and Aghaee, N.-G. (2006). Intelligent

agents, simulation, and gaming. Simulation & Gam-

ing, 37(3):339–349.

Yilmaz, L. and

Oren, T. I. (2007). Agent-directed sim-

ulation systems engineering. In Proceedings of

the 2007 Summer Computer Simulation Conference,

pages 897–904. Society for Computer Simulation In-

ternational.

Yilmaz, L., Taylor, S. J. E., Fujimoto, R., and Darema, F.

(2014). Panel: The Future of Research in Modeling &

Simulation. In Proceedings of the 2014 Winter Simu-

lation Conference, WSC ’14, pages 2797–2811, Pis-

cataway, NJ, USA. IEEE Press.

DensifyingtheSparseCloudSimSaaS:TheneedofaSynergyamongAgent-directedSimulation,SimSaaSandHLA

177