**3. Prototyping optimal design platform for engineering**

The analysis of a state‐of‐art scientific platforms shows that there is a need of distributed computing‐oriented platform. This obliges to redesign similar environments in the terms of separate interacting software services. So the designers should specify a workflow of the interaction of services.

Based on PRCC facilities, the Institute of Applied System Analysis (IASA) of NTUU "Kiev Polytechnic Institute" (Ukraine) has developed the user case WebALLTED<sup>1</sup> as the Web‐ enabled engineering design platform, intended, in particular, for modeling and optimization of nonlinear dynamic systems, which consist of the components of different physical nature and which are widely spread in different scientific and engineering fields. It is the cross‐disci‐ plinary application for distributed computing.

Developed engineering service‐oriented simulation platform consists of the following layers (**Figure 2**). The most important features of this architecture are the following: Web accessibil‐ ity, the distribution of the functionality across the software services in e‐infrastructure, the compatibility with existing protocols and standards, the support of user‐made scenarios in

**Figure 2.** Main elements of SOA in the engineering simulation system.

<sup>1</sup> ALLTED means ALL TEchnologies Designer [7, 8].

development‐time and in run‐time, and the encapsulation of the software services interac‐ tion complexity.

Based on PRCC facilities, the Institute of Applied System Analysis (IASA) of NTUU "Kiev

enabled engineering design platform, intended, in particular, for modeling and optimization of nonlinear dynamic systems, which consist of the components of different physical nature and which are widely spread in different scientific and engineering fields. It is the cross‐disci‐

Developed engineering service‐oriented simulation platform consists of the following layers (**Figure 2**). The most important features of this architecture are the following: Web accessibil‐ ity, the distribution of the functionality across the software services in e‐infrastructure, the compatibility with existing protocols and standards, the support of user‐made scenarios in

as the Web‐

Polytechnic Institute" (Ukraine) has developed the user case WebALLTED<sup>1</sup>

plinary application for distributed computing.

76 Recent Progress in Parallel and Distributed Computing

1

ALLTED means ALL TEchnologies Designer [7, 8].

**Figure 2.** Main elements of SOA in the engineering simulation system.

The following functions are accessible via user interface: authentication, workflow editor, artefacts repository management environment, task monitoring, and more. The server side of the system is designed as multitier one in order to implement the workflow concept described early. First‐access tier is the portal supporting user environment. The purpose of its modules is the following: the user‐input‐based generation of abstract workflow specification; the tran‐ sition of task specification to lower tiers, where the task will be executed; and the postprocess‐ ing of results including saving the artefacts in DB.

The workflow manager works as second‐execution tier. It is deployed in the execution server. The purpose of this tier is the mapping of the abstract workflow specification to particular software services. The orchestration is done using the specific language similar to WS‐BPEL for BPEL instruments. The workflow manager starts executing particular workflow with the external orchestrator as well as observes the state of workflow execution and procures its results.

Particular workflow is working with functional software services and performs the following actions: data preprocessing and postprocessing, simulation, optimization, and so on. If high demand for resources is forecasted, only one node could be loaded to heavy. So the computa‐ tion is planned on separate nodes and hosting grid/cloud services. These services give pos‐ sibility to use widespread infrastructure (such as grid or cloud). It is possible to modify and to introduce of new functions to the system. This is done by the user by selection or registration of another Web or grid/cloud services.

The user is able to start the task in an execution tier. Task specification is transient to the service of workflow management. This abstract workflow is transformed to the particular implementation on execution server. Then, the workflow manager analyses the specifica‐ tion, corrects its possible errors (in some extent), demands the data about the services from the repository, and performs binding of activity sequence and software services calls. For the arrangement of software services in correct invocation order, the Mapper unit is used in the workflow. It initializes XML messages, variables, etc., and provides the means for the control during a run‐time including the observing of workflow execution, its cancelling, early results monitoring, and so on. Finally, the orchestrator executes this particular "script."

User is informed about the progress of the workflow execution by monitoring unit communi‐ cating with workflow manager. When execution is finished, the user can retrieve the results, browse and analyze them, and repeat this sequence if needed.

The architecture hides the complexity of web‐service interaction from user with abstract workflow concept and simple graphical workflow editor (**Figure 3**).

Web services are representing the basic building blocks of simulation system's functionality, and they enable customers to build and adjust scenarios and workflows of their design proce‐ dures or mathematical experiments via the Internet by selecting the necessary Web services, including automatic creation of equations of a mathematical model (an object or a process)

**Figure 3.** WebALLTED graphical workflow editor.

based on a description of its structure and properties of the used components, operations with large‐scale mathematical models, steady‐state analysis, transient and frequency‐domain analysis, sensitivity and statistical analysis, parametric optimization and optimal tolerance assignment, solution centering, yield maximization, and so on [3].

Computational supporting services are based mostly on innovative numeric methods and can be composed by an end user for workflow execution on evaluable grid/cloud nodes [3]. They are oriented, first of all, on Design Automation domain, where simulation, analysis, and design can be done for different electronic circuits, control systems, and dynamic systems composed of electronic, hydraulic, pneumatic, mechanical, electrical, electromagnetic, and other physical phenomena elements.

The developed methodology and modeling toolkit support collective design of various micro‐ electro‐mechanical systems (MEMS) and different microsystems in the form of chips.
