Simulation Module Markup Language

Multi-Paradigm Modeling

Simulation Module Markup Language

Home Vision Syntax Semantics Examples Java tools Download Discussion	Summary The development of complex models can be greatly facilitated by the utilization of libraries of reusable model components. In this project we propose an XML-based Simulation Module Markup Language (SMML) for implementing archivable modules in support of continuous spatial modeling. This declarative formalism provides the high level of abstraction necessary for maximum generality, provides enough detail to allow a dynamic simulation to be generated automatically, and avoids the "hard-coded" implementation of space-time dynamics that makes procedural specifications of limited usefulness for specifying archivable modules. A set of these SMML modules can be hierarchically linked to create a parsimonious model specification, or "parsi-model". The parsi-model exists within the context of a modeling environment ( an integrated set of software tools which provide the computer services necessary for simulation development and execution ), which can offer simulation services that are not possible in a loosely-coupled "federated" environment, such as graphical module development and configuration, automatic differentiation of model equations, run-time visualization of the data and dynamics of any variable in the simulation, transparent distributed computing within each module, and fully configurable space-time representations. We believe this approach has great potential for bringing the power of modular model development into the collaborative simulation arena. Objectives Our overall objective is the development of new methods and tools to support collaborative modeling and high performance simulation of complex environmental systems. This research, which builds on a large body of previous work in this area, focuses primarily on developing methods, infrastructure, and interfaces to facilitate the understanding of complex systems through the development of models integrating multiple scales and paradigms. We recognize the need for a variety of approaches and techniques to develop general models of complex natural realities. A major point in this research plan is the development of an integrative framework to effectively connect complementary modeling paradigms. We will be addressing a number of fundamental issues involved in modeling coupled systems with different spatio-temporal scales and representations. The integrative paradigm and modeling infrastructure developed should greatly facilitate the application of computer modeling to the study of environmental systems (and complex systems in general) in support of research, education, and environmental management. Complex Models Development of large-scale models in general has been limited by the ability of any single team of researchers to deal with the conceptual complexity of formulating, building, calibrating, debugging, and understanding complex models. Realistic environmental models are becoming much too complex for any single group of researchers to implement single-handed, requiring collaboration between botanists, zoologists, hydrologists, chemists, land managers, economists, ecologists, and others. Communicating the structure of the model to others can become an insurmountable obstacle to collaboration and acceptance of the model. Policy makers are unlikely to trust a model they don't understand. In addition, it should be noted that the level of analysis and understanding of the current generation of complex, highly non-linear models is often surprisingly poor. A model which is so complex and comprehensive to limit the depth of its own analysis is a waste of research money and opportunity for scientific progress. Tools and techniques aimed at understanding the mechanistic determinants of a model's behavior are needed, as are methods to identify the limits of the "understandable" (or useful) complexity in modeling. Collaborative Infrastructure Collaborative modeling environments, as intended throughout these pages, are software tools designed to operate over a network to allow heterogeneous, disjunct groups of researchers to work together on a project, sharing data and computer resources, and communicating ideas, data and results in real time. As discussed in the previous section, the need for such tools is a consequence of the increasing complexity and multidisciplinarity of the research, and of the increasing delocalization of research groups. Many approaches to collaborative research are possible, depending on which collaborative model is considered and which level of collaboration is required. Approaches vary from the simple, asynchronous sharing of data to the incorporation of conflict resolution algorithms within multi-agent models of individual initiative. The complexity of the collaborative tools reflects the approach. It is also obviously important that the collaborative tool brings simplicity and clarity to the modeling approach, instead of overburdening the users with technical detail. The inevitably non-flat learning curve of any tool and the lack of time to learn more than a few software packages limit the scientist's permeability to innovation in the modeling field. Implications and strategies for feasible collaborative ecological research must be considered in light of the goal of increasing understanding of the model domain, as opposed to building more and more complex simulation models of infinitely complex realities. The importance of integrative, collaborative tools in an increasingly complex modeling scenario has also less obvious justifications. In the case of ecological modeling, there is the distinct danger of tools preventing innovation. The ecological science needs new concepts to formalize and understand nature's complexity. Modeling approaches can be thought of as providing metaphors which are relevant in this regard. Constraints are imposed and limits set by techniques and approaches, the same way that thought is limited by experience, language, and audience. It is very important that collaborative tools do not constrain the researcher within a specific view of natural complexity, but rather allow free space for thought by using knowledge models which allow flexibility and reorganization. Modular Modeling A well-recognized method for reducing conceptual and programming complexity involves structuring the model as a set of distinct modules with well-defined interfaces. Modular design facilitates collaborative model construction, since teams of specialists can work independently on different modules with minimal risk of interference. Modules can be archived in distributed libraries and serve as a set of templates to speed future development. A modeling environment that supports modularity could provide a universal modeling language to promote worldwide collaborative model development. In order to achieve flexibility in knowledge representations, it is important to develop a formalism for coding archivable modules that allows maximal generality and applicability of the modules. This is best achieved by avoiding over-specification of modules, i.e. to achieve maximum generality by including only information essential to definition of a module. Every bit of spurious information included in a module definition becomes a constraint which can reduce the applicability of the module. Graphical Modeling Tools A second step toward reducing the complexity of the modeling process involves the utilization of graphical, icon-based module interfaces, wherein the structure of the module is represented diagramatically, so that new users can recognize the major interactions at a glance. Scientists with little or no programming experience can begin building and running models almost immediately. Inherent constraints make it much easier to generate bug-free models. Built-in tools for display and analysis facilitate understanding, debugging, calibration and analysis of the module dynamics. In light of the previous comments on generality, it is essential that the simplification inherent in the adoption of familiar graphical metaphors for the model building blocks does not turn into a new set of constraints on the model structure, but is chosen in order to allow greater freedom for development and reorganization of components. One major advantage of this graphical approach to modeling is that the process of modeling can become a consensus building tool. The graphical representation of the model can serve as a blackboard for group brainstorming, allowing students, educators, policy makers, scientists, and stakeholders to all be involved in the modeling process. New ideas can be tested and scenarios investigated using the model within the context of group discussion as the model grows through a collaborative process of exploration. When applied in this manner the process of creating a model may be more valuable than the finished product. Multiple Spatio-Temporal Representations and Scales Building realistic spatially explicit ecological generally requires the integration of multiple spatial data structures in a single model, and the coupling of data and models designed to operate at different spatio-temporal scales. Entities such as vegetation cover, river/canal networks, and individual animals may co-exist within a single model and require different classes of spatial representation ( e.g. raster, vector, or agent-based ). The coupling of multiple scales requires methods for extrapolating data and models from one scale of observation and aggregation to other scales. Thus the implementation of the concept of space in the modeling environment must be general enough to allow the instantiation of a wide range of specific space-time representations, and the details of linking, transferring data between, and decomposing (over multiple processors) these spatial representations should be invisible to the modelers. In addition, realistic environmental models require the integration of multiple temporal representations in a single model. For example, many processes are best represented using differential equations, others are best represented using event-based simulation, and others, such as input-output economic models, use a "black-box" or look-up table implementation. Some processes, such as storm events, are best handled with a hybrid approach. Entity-Based Models The need for supporting both mobile entities and landscape processes in management scale models is being increasingly realized by environmental modelers. Mobile entities, such as individual animals, move in the landscape while retaining their identities as unique individuals. In contrast, landscape processes, such as erosion or plant growth, describe changing characteristics of the landscape in a fixed area. In modeling endangered species, for example, it is important to keep track of individuals over time and space, interactions among individuals, landscape processes over time and space, interactions among landscape processes, and interactions of individuals with landscape processes. Individual based simulation modeling provides an essential complement to cell based process models. Complex Modeling using a Module Markup Language As discussed earlier, the development of complex models can be greatly facilitated by the utilization of libraries of reusable model components. To this end we have proposed a Simulation Module Markup Language (SMML) for implementing archivable modules in support of complex, multi-paradigm modeling. Our basic premise is that the most common approach to model integration, which involves linking procedural models using distributed object formalisms, is greatly limited by the fact that the various sub-models are, by their nature, overspecified as modules. In the process of implementing a sub-model in a procedural programming language, the implementer generally "hard-codes" many choices such as programming language, spatio-temporal representation, model control and IO interfaces, computing paradigm ( serial / parallel-message passing / parallel-shared memory ), etc. When the sub-model enters the collaborative domain as a reusable module these fixed aspects are seen to be extremely limiting and irrelevant to the essential dynamics of the model. In the SiMMaL project we are proposing a more flexible approach, which defines archivable modules using abstract model specifications which contain only enough information to specify the essential dynamics of the module and allow a wide range of customized procedural implementations to be generated automatically. This purely declarative formalism provides the high level of abstraction necessary for maximum generality, provides enough detail to allow a dynamic simulation to be generated automatically, and avoids the hard-coded implementation of space-time dynamics that makes procedural specifications of limited usefulness for specifying archivable modules. A set of these modules can be hierarchically linked within the SMML formalism to create an SMML model. We believe this approach has great potential for bringing the power of modular model development into the collaborative simulation arena. Global Benefits of Program We believe that it is now possible, because of recent developments in data acquisition and display, information-sharing technologies, and large-scale computing and modeling, to build computer simulation models that cover the salient features of the world's climate, economy, and ecosystems, and the major interactions among them. The combined models, ranging in scale and extent from a single "patch" to the globe, will provide both an integrated conceptual framework, and a practical tool allowing researchers from many disciplines to collaborate effectively in order to produce effective answers to the critical problems facing humanity. To achieve this goal it will be necessary to mobilize this scholarly community within a worldwide "collaboratory" based on new electronic information sharing technologies, bringing together leaders in advanced computation and software development with leaders in ecological and economic data collection and modeling. The SiMMaL project is intended to support this collaborative spatial modeling effort. The development of this paradigm should greatly facilitate the application of computer modeling to the study of spatial systems in support of research, education, and policy making.

Home

Summary

The development of complex models can be greatly facilitated by the utilization of libraries of reusable model components. In this project we propose an XML-based Simulation Module Markup Language (SMML) for implementing archivable modules in support of continuous spatial modeling. This declarative formalism provides the high level of abstraction necessary for maximum generality, provides enough detail to allow a dynamic simulation to be generated automatically, and avoids the "hard-coded" implementation of space-time dynamics that makes procedural specifications of limited usefulness for specifying archivable modules. A set of these SMML modules can be hierarchically linked to create a parsimonious model specification, or "parsi-model". The parsi-model exists within the context of a modeling environment ( an integrated set of software tools which provide the computer services necessary for simulation development and execution ), which can offer simulation services that are not possible in a loosely-coupled "federated" environment, such as graphical module development and configuration, automatic differentiation of model equations, run-time visualization of the data and dynamics of any variable in the simulation, transparent distributed computing within each module, and fully configurable space-time representations. We believe this approach has great potential for bringing the power of modular model development into the collaborative simulation arena.

Objectives

Our overall objective is the development of new methods and tools to support collaborative modeling and high performance simulation of complex environmental systems. This research, which builds on a large body of previous work in this area, focuses primarily on developing methods, infrastructure, and interfaces to facilitate the understanding of complex systems through the development of models integrating multiple scales and paradigms. We recognize the need for a variety of approaches and techniques to develop general models of complex natural realities. A major point in this research plan is the development of an integrative framework to effectively connect complementary modeling paradigms. We will be addressing a number of fundamental issues involved in modeling coupled systems with different spatio-temporal scales and representations. The integrative paradigm and modeling infrastructure developed should greatly facilitate the application of computer modeling to the study of environmental systems (and complex systems in general) in support of research, education, and environmental management.

Complex Models

Development of large-scale models in general has been limited by the ability of any single team of researchers to deal with the conceptual complexity of formulating, building, calibrating, debugging, and understanding complex models. Realistic environmental models are becoming much too complex for any single group of researchers to implement single-handed, requiring collaboration between botanists, zoologists, hydrologists, chemists, land managers, economists, ecologists, and others. Communicating the structure of the model to others can become an insurmountable obstacle to collaboration and acceptance of the model. Policy makers are unlikely to trust a model they don't understand.

In addition, it should be noted that the level of analysis and understanding of the current generation of complex, highly non-linear models is often surprisingly poor. A model which is so complex and comprehensive to limit the depth of its own analysis is a waste of research money and opportunity for scientific progress. Tools and techniques aimed at understanding the mechanistic determinants of a model's behavior are needed, as are methods to identify the limits of the "understandable" (or useful) complexity in modeling.

Collaborative Infrastructure

Collaborative modeling environments, as intended throughout these pages, are software tools designed to operate over a network to allow heterogeneous, disjunct groups of researchers to work together on a project, sharing data and computer resources, and communicating ideas, data and results in real time. As discussed in the previous section, the need for such tools is a consequence of the increasing complexity and multidisciplinarity of the research, and of the increasing delocalization of research groups. Many approaches to collaborative research are possible, depending on which collaborative model is considered and which level of collaboration is required. Approaches vary from the simple, asynchronous sharing of data to the incorporation of conflict resolution algorithms within multi-agent models of individual initiative. The complexity of the collaborative tools reflects the approach.

It is also obviously important that the collaborative tool brings simplicity and clarity to the modeling approach, instead of overburdening the users with technical detail. The inevitably non-flat learning curve of any tool and the lack of time to learn more than a few software packages limit the scientist's permeability to innovation in the modeling field. Implications and strategies for feasible collaborative ecological research must be considered in light of the goal of increasing understanding of the model domain, as opposed to building more and more complex simulation models of infinitely complex realities.

The importance of integrative, collaborative tools in an increasingly complex modeling scenario has also less obvious justifications. In the case of ecological modeling, there is the distinct danger of tools preventing innovation. The ecological science needs new concepts to formalize and understand nature's complexity. Modeling approaches can be thought of as providing metaphors which are relevant in this regard. Constraints are imposed and limits set by techniques and approaches, the same way that thought is limited by experience, language, and audience. It is very important that collaborative tools do not constrain the researcher within a specific view of natural complexity, but rather allow free space for thought by using knowledge models which allow flexibility and reorganization.

Modular Modeling

A well-recognized method for reducing conceptual and programming complexity involves structuring the model as a set of distinct modules with well-defined interfaces. Modular design facilitates collaborative model construction, since teams of specialists can work independently on different modules with minimal risk of interference. Modules can be archived in distributed libraries and serve as a set of templates to speed future development. A modeling environment that supports modularity could provide a universal modeling language to promote worldwide collaborative model development.

In order to achieve flexibility in knowledge representations, it is important to develop a formalism for coding archivable modules that allows maximal generality and applicability of the modules. This is best achieved by avoiding over-specification of modules, i.e. to achieve maximum generality by including only information essential to definition of a module. Every bit of spurious information included in a module definition becomes a constraint which can reduce the applicability of the module.

Graphical Modeling Tools

A second step toward reducing the complexity of the modeling process involves the utilization of graphical, icon-based module interfaces, wherein the structure of the module is represented diagramatically, so that new users can recognize the major interactions at a glance. Scientists with little or no programming experience can begin building and running models almost immediately. Inherent constraints make it much easier to generate bug-free models. Built-in tools for display and analysis facilitate understanding, debugging, calibration and analysis of the module dynamics. In light of the previous comments on generality, it is essential that the simplification inherent in the adoption of familiar graphical metaphors for the model building blocks does not turn into a new set of constraints on the model structure, but is chosen in order to allow greater freedom for development and reorganization of components.

One major advantage of this graphical approach to modeling is that the process of modeling can become a consensus building tool. The graphical representation of the model can serve as a blackboard for group brainstorming, allowing students, educators, policy makers, scientists, and stakeholders to all be involved in the modeling process. New ideas can be tested and scenarios investigated using the model within the context of group discussion as the model grows through a collaborative process of exploration. When applied in this manner the process of creating a model may be more valuable than the finished product.

Multiple Spatio-Temporal Representations and Scales

Building realistic spatially explicit ecological generally requires the integration of multiple spatial data structures in a single model, and the coupling of data and models designed to operate at different spatio-temporal scales. Entities such as vegetation cover, river/canal networks, and individual animals may co-exist within a single model and require different classes of spatial representation ( e.g. raster, vector, or agent-based ). The coupling of multiple scales requires methods for extrapolating data and models from one scale of observation and aggregation to other scales. Thus the implementation of the concept of space in the modeling environment must be general enough to allow the instantiation of a wide range of specific space-time representations, and the details of linking, transferring data between, and decomposing (over multiple processors) these spatial representations should be invisible to the modelers.

In addition, realistic environmental models require the integration of multiple temporal representations in a single model. For example, many processes are best represented using differential equations, others are best represented using event-based simulation, and others, such as input-output economic models, use a "black-box" or look-up table implementation. Some processes, such as storm events, are best handled with a hybrid approach.

Entity-Based Models

The need for supporting both mobile entities and landscape processes in management scale models is being increasingly realized by environmental modelers. Mobile entities, such as individual animals, move in the landscape while retaining their identities as unique individuals. In contrast, landscape processes, such as erosion or plant growth, describe changing characteristics of the landscape in a fixed area. In modeling endangered species, for example, it is important to keep track of individuals over time and space, interactions among individuals, landscape processes over time and space, interactions among landscape processes, and interactions of individuals with landscape processes. Individual based simulation modeling provides an essential complement to cell based process models.

Complex Modeling using a Module Markup Language

As discussed earlier, the development of complex models can be greatly facilitated by the utilization of libraries of reusable model components. To this end we have proposed a Simulation Module Markup Language (SMML) for implementing archivable modules in support of complex, multi-paradigm modeling. Our basic premise is that the most common approach to model integration, which involves linking procedural models using distributed object formalisms, is greatly limited by the fact that the various sub-models are, by their nature, overspecified as modules. In the process of implementing a sub-model in a procedural programming language, the implementer generally "hard-codes" many choices such as programming language, spatio-temporal representation, model control and IO interfaces, computing paradigm ( serial / parallel-message passing / parallel-shared memory ), etc. When the sub-model enters the collaborative domain as a reusable module these fixed aspects are seen to be extremely limiting and irrelevant to the essential dynamics of the model.

In the SiMMaL project we are proposing a more flexible approach, which defines archivable modules using abstract model specifications which contain only enough information to specify the essential dynamics of the module and allow a wide range of customized procedural implementations to be generated automatically. This purely declarative formalism provides the high level of abstraction necessary for maximum generality, provides enough detail to allow a dynamic simulation to be generated automatically, and avoids the hard-coded implementation of space-time dynamics that makes procedural specifications of limited usefulness for specifying archivable modules. A set of these modules can be hierarchically linked within the SMML formalism to create an SMML model. We believe this approach has great potential for bringing the power of modular model development into the collaborative simulation arena.

Global Benefits of Program

We believe that it is now possible, because of recent developments in data acquisition and display, information-sharing technologies, and large-scale computing and modeling, to build computer simulation models that cover the salient features of the world's climate, economy, and ecosystems, and the major interactions among them. The combined models, ranging in scale and extent from a single "patch" to the globe, will provide both an integrated conceptual framework, and a practical tool allowing researchers from many disciplines to collaborate effectively in order to produce effective answers to the critical problems facing humanity. To achieve this goal it will be necessary to mobilize this scholarly community within a worldwide "collaboratory" based on new electronic information sharing technologies, bringing together leaders in advanced computation and software development with leaders in ecological and economic data collection and modeling. The SiMMaL project is intended to support this collaborative spatial modeling effort. The development of this paradigm should greatly facilitate the application of computer modeling to the study of spatial systems in support of research, education, and policy making.

Site maintained by: Thomas Maxwell

Last modified: Tue Mar 9 15:53:51 EST 1999