Keynote Lectures

Cloud Computing - State-of-the-Art and Future Research Trends
Eleni Karatza, Aristotle University of Thessaloniki, Greece, Greece

The Role of Domain Specific Languages in Modeling and Simulation
Adelinde M. Uhrmacher, University of Rostock, Germany, Germany

Parallel Discrete Event Simulation - Past, Present and Future
Richard Fujimoto, Georgia Institute of Technology, United States, United States

Agent-based Models for Exploring Social Complexity, with an Application of Network Analysis to Agents
Pietro Terna, Università di Torino, Italy, Italy

Abstract
Clouds have been very popular, and their performance becomes more important due to the increase of users and applications. Currently, many enterprises are adopting Clouds to achieve high performance for their applications at low costs.
Because of the nature of these systems, there are important issues that must be addressed, such as: resource allocation, efficient scheduling, energy conservation, reliability, security and trust, cost, availability, quality. Effective management of cloud resources is crucial to use effectively the power of these systems and achieve high system performance. Furthermore, software structures that best exploit clouds capabilities while providing applications compatibility should be examined.
Cloud computing is a concept that has emerged from grid computing; it provides users the ability to acquire computational resources on demand from its virtually infinite pool on a pay-as-you-go basis.
The cloud computing paradigm can offer various types of services, such as computational resources for HPC applications, web services, social networking, etc. Resource allocation and scheduling is a difficult task in clouds where there are many alternative heterogeneous computers. If cloud computing is going to be used for HPC, appropriate methods must be considered for allocating resources to user requests efficiently, VM scalability, as well as effectively scheduling the tasks. The scheduling algorithms must seek a way to maintain a good response time to leasing cost ratio. Furthermore, adequate data security and availability are critical issues that have to be considered along with energy-efficient solutions that are required to minimize the impact of cloud computing on the environment.

Abstract
Simulation, as an experiment performed with a formal model, is meanwhile widely accepted as the third branch of science, complementing theoretical and empirical science. However, “in-silico” experiments often suffer from the lack of reproducibility which has led to the opinion that published simulation results cannot be relied upon. In this talk I will take a closer look what role domain specific languages can play in addressing this problem.  The syntax and the semantics of domain specific modeling languages determine, how and which system can be described, respectively. Only by clearly distinguishing between syntax and semantics, the advantage of syntax can be exploited, i.e., to provide a compact and succinct description of the system of interest. Domain-specific languages are also applied for supporting experiments. The specification in a domain-specific language helps storing, revising and adapting experiments. How both, domain-specific languages for modeling as well as for simulation experiments, may influence modeling and simulation, and may contribute to the reproducibility of simulation, will be discussed based on case studies from cell biology and demography.

Abstract
The parallel discrete event simulation field emerged in the 1970s around the problem of distributing the execution of a discrete event simulation program across multiple processors while still obtaining the same results as a sequential execution. The field has since evolved and addressed many challenges to speeding up and scaling simulation programs, and remains an active area of research to this day. Many impressive successes have been reported in the literature. Today, issues such as the “power wall” facing modern processors and new developments such as massively parallel supercomputers and cloud computing platforms make the technology more important than ever before. Further, broader technology trends such as “big data” and the Internet of Things present new challenges and opportunities.
I will give a retrospective view of the parallel discrete event simulation field dating back to its origins in solving the so-called synchronization problem. I will describe important advances and successes that illustrate the potential offered by this technology. Key impediments that have prevented the technology from achieving more widespread adoption by the general modeling and simulation community are discussed as well as important research challenges that remain.

Abstract
Thinking to agent-based models as artifacts, useful to explore economic complexity, means to introduce three concepts: (i) on the technical side, the agent-based methodology; (ii) in the social science perspective, the idea of building artifacts also in the social domain; (iii) in a more general view, the idea of complexity. With an impressive image, this way of researching involves the application of the Galileo’s method in social sciences.
Remembering the theoretical roots in cybernetics and, more recently, in complexity science, we need also technical roots, with the capability of building models that could be accepted by a wide audience.
Agent-based models can be also particularly useful for policy-making, mainly if they are both quite simple in their structure and highly sophisticated in their outcomes. The pursuing of simplicity and sophistication can be made more efficacious by applying network analysis to the emergent results.