P.A. Smirnov, S.V. Kovalchuk, A.V. Boukhanovsky
The article presents Virtual Simulation Objects (VSO) concept which allows to develop domain-specific information and control systems with high-level simulation environments based on cloud computing infrastructure. The concept defines the composition and usage principles of virtual objects. The objects incorporate set of models which can be translated into composite application in form of workflow calling domain-specific applications. Also the objects are related to some real-world objects, which can be explored via the set of models. Object’s composition can be concerned as a structural model, which defines the set of objects and ways of theirs interaction. The VSO concept is based on iPSE (Intelligent Problem Solving Environment) concept which supports building of intelligent domain-specific PSEs for eScience and operates with the knowledge on problem solving with software which is called within composite application presented as abstract workflows.
For the purpose of VSO knowledge storing semantic structure for description of virtual objects was developed. The structure defines classes and relationship of the expert knowledge base. It contains following elements: objects, properties, models, scenarios , and implementations. Instrumental environment working with the knowledge base was developed GUI software using Silverlight technology. The environment was integrated with cloud computing platform CLAVIRE, which can run composite applications constructed after VSO-based systems’ structure. WCF technology is used to interact with CLAVIRE services. The instrumental environment contains two key applications: VSO constructor – GUI software for virtual objects composition, and VSO class editor – GUI software for design and development of virtual objects.
Presented VSO concept is called to try fill in the gap between three large fields of knowledge: information technologies, knowledge engineering, and simulation along with general systems theory. Proposed approach can be concerned as comprehensive solution, which on union of all these subjects for satisfaction of the interest toward system-level scientific simulation. The approach provides support during whole process of simulation-based investigation: modeling, simulation and result analysis.