CC BY
9COMPUTER SCIENCE
OPTIMIZATION AND SIMULATION OF COMPLEX NONLINEAR SYSTEMS AND CIRCUITS AS THE COMPOSITION
OF CLOUD SERVICES
Ph. D. Petrenko I.A., DSc Petrenko A. I.
Ukraine, Kiev, National Technical University "Kyiv Polytechnic Institute"
Abstract. Service-oriented computing (SOC) is the new cross-disciplinary paradigm for distributed computing that is changing the way software applications are designed, architected, delivered and consumed. Services are autonomous, platform-independent computational entities that can be used in a platform independent way. Services can be described, published, discovered, and dynamically assembled for developing massively distributed, interoperable, evolvable systems. This paper provides a roadmap of development of the Engineering Design Platform, based on SOC and intended, in particular, for modelling and optimization of Nonlinear Dynamic Systems, based on components of different physical nature and being widely spread in different scientific and engineering fields.
1. Introduction
The distinction between SOC and traditional computing is that application builders no longer construct software using a programming language. Instead, they specify the application logic in a high-level specification language, utilizing standard services as components. Services implement functions that can range from answering simple requests to executing sophisticated processes requiring peer-to-peer relationships between possibly multiple layers of service consumers and providers [1-3].
The delivery of software as a set of distributed services can help to solve problems like software reuse, deployment and evolution. The "software as a service model" will open the way to the rapid creation of new value-added composite services based on existing ones. Although service-oriented computing in cloud computing environments presents a new set of research challenges, their combination provides potentially transformative opportunities.
So any user community, regardless of its discipline, should be supplied with the technological approach to build their own distributed compute-intensive multidisciplinary applications rapidly. This paper contains the roadmap of development of the Engineering Design Platform, based on Service-oriented computing (SOC) and intended, in particular, for modeling and optimization of Nonlinear Dynamic Microsystems, being consisted of components of different physical nature and being widely spread in different scientific and engineering fields.
Service-oriented architecture is used, where the entire simulation process is broken down into a set of loosely-coupled interacting cloud services (specific software components with unified interfaces) that can be performed at different Cloud sites. Web services are representing the basic building blocks of simulation system's functionality: input data preprocessing; mathematical model development and its dimension reduction; DC, AC, TR, STA, FOUR and sensitivities analysis; parametrical optimization, tolerances assignment; statistical analysis and yield maximization; results processing etc. [4,5].
Web applications use a client-server model with a network via an Intranet or the Internet, and are browser-based on the client side. Developed web services can be dynamically orchestrated to execute the workflows composed by the users of the web-based environment. This environment will provide user friendly web interface to the workflow editor and services toolkit for multidisciplinary applications. Beside EDA the simulation, analysis and design can be done for different control systems and dynamic systems composed of electronic, hydraulic, pneumatic, mechanical, electrical, electromagnetic, and other physical phenomena elements. Complex non-linear systems of this type are
widely used in modern aerospace, robotics, NC machine tools, test equipment, highway engineering, agricultural and other applications.
The rest of this paper is organized as follows: Section 2 presents SOC in Engineering Design and innovative features of our proposal. Section 3 describes the impact and exploitation plan and Section 4 concludes this paper.
2. Description of a problem solution
Under pressure of the market the complexity of developed applications is rapidly growing. Modern applications are no more holistic units, as they were in the past. They are not monolithic kernels working on a single computer platform, but rather a set of dynamically changing modules. Service-oriented architecture concept was developed together with development of the Web. This modular approach to software development is based on the use of distributed loosely coupled replaceable components with standardized interfaces for communication via standardized protocols.
Applications are created by several teams of developers using different programming languages, using the set of data that can come "on-line" from several, usually geographically distributed sources. As a result, there is a need for a new style of application development, based on a software services (program services).This style allows programmers not to start work from scratch, and create new applications using ready-made service available on the network. Modern engineering applications are built as a complex network of services offered by different providers, based on heterogeneous resources of different organizational structures. Together, these services form a powerful and versatile tool for the modern e-science: if cloud enables large-scale computation, figuratively speaking, shall make available scientific superpower "virtual supercomputer", the workflow system (workflow management system) can help develop a plan composed of computational procedures, dynamically adjust it to suit the purposes of the simulation and its interim results, monitor the progress of the whole complex computational process.
Thus, workflow management system provides users with the possibility of rapid design, simulation, controlled performance adjustments and debugging reuse of algorithmic sequences ("route" or "map") for many steps of analysis and data visualization.
Our proposal Cloud-OPTIMSIM has the following innovative features:
• Implementation of novel service-oriented design paradigm in Engineering according to which all levels of design are divided into separate loosely coupled stages and procedures for their subsequent transfer to the form of standardized web-services.
• The repository of design web-services will be created which will contain components developed by different producers that support collective design work and globalization of R&D activities in World, either for free or with certain fee.
• Design in Engineering becomes personalized and customized because users can build and adjust their design scenario and workflow by selecting the necessary web-services (as calculation procedures) to be executed on cloud resources.
• Users can also easily introduce new component models and their parameters, which are absent in any existing SPICE-like simulation software.
• Algorithms proposed for additional design web services are novel and unique (multi-criterion optimization, optimal tolerances assignment, yield maximization, etc.) [4,5].
The Cloud-OPTIMSIM project will reflect the "converging" tendency of engineering and research today, as well as the need for company's collaboration across the branches and disciplines, needed to master hybrid technologies. It with its enabling technology will bring innovations upstream (e.g. new software design concepts) and downstream of the Mechatronics and MEMS value (new applications) within the lifetime of the project and thereafter in the lifecycle of the Cloud-OPTIMSIM [6]. World level research and innovation policy has been designed to promote research activities in the MEMS field, through public expenditure in order to develop a strong knowledge base quite successfully.
3. Implementation plan
The industrial relevance of the proposal is based upon two sides: dynamic systems simulation capabilities (functionality of high demand) and coherence with service oriented architecture (modern software development technology). Web applications use a client-server model with a network via an Intranet or the Internet, and are browser-based on the client side. The main objective of Cloud-
OPTIMSIM is the support of the European CloudSME project that promotes the cloud resources in manufacturing and engineering simulations [7].
The Cloud-OPTIMSIM architecture is based on using CloudBroker Platform [8] of CloudSME Project:
• Functionality is distributed across the ecosystem of both web services and cloud services (enabling utilization of cloud computing resources);
• It is compatible with adopted standards and protocols;
• It supports custom user scenario of development and execution functionality which is accessible with lightweight web interface. It hides the complexity of web-service interaction from user with abstract workflow concept and simple graphical workflow editor.
Design in Engineering becomes personalized and customized because users can build and adjust their design scenario and workflow by selecting the necessary web-services (as calculation procedures) to be executed on cloud resources.
Developed web services can be dynamically orchestrated to execute the workflows composed by the users of the web-based environment. This environment (supported by Cloud Platform's WS-PGRADE workflow execution functionalities) will provide user friendly web interface to the workflow editor and services toolkit for multidisciplinary applications [9].
The project in hand adds a new simulation software to those already selected for CloudSME Project (SIMUL8, TransAT, 3D Scan Insole Designer, BFly). It disseminates CloudSME Project, first of all, on EDA (Electronic Design Automation) domain which migration to cloud computing remains slow.
The basic functionality is provided by SPICE-like core procedures, being well known to western engineers. But in comparison with SPICE-like programs our solution offers:
• Comprehensive optimization and tolerance assignment procedures [3-5].
• Original way of generating a system-level model of MEMS from FEM component equations (being received, for example, by means of ANSYS or TransAT) when these equations with boundary conditions are transformed into the equivalent equations of a schematic model, which consist of L, C and G components, and then simplified by means of Y-Д transformation.
• Alternative approach to the secondary response parameters determination (delays, rise and fall times, frequency band, etc.), which can be used in optimization and tolerance assignment procedures.
This delivers some significant benefits: browsers provide a true client side processing capability while running on CloudSME platform from any location, the network provides a real-time collaboration capability that is consistent across users, and finally the web server provides a centralized single source methodology for data as it changes over time. The major outcome of the project will be the service-oriented highly advanced Optimal Engineering Design simulation software. The main objective is divided into quantifiable sub-objectives:
• Implementation of the emerging distributed design paradigms in Engineering, which can be executed in cloud environments.
• Development of tools which enable users to build and adjust scenarios and workflows of their design procedures or mathematical experiments via the Internet by selecting the necessary webservices (as calculation procedures) to be executed on cloud resources, including automatic creation of equations of the object mathematical model (a circuit or a system) based on a description of its structure and properties of the used components in the form of differential-algebraic or differential equations reduced to the form, which is acceptable for solving by other software subsystems.
• Improvement of mathematical background of web-services which cover novel types of analyses and solutions of these equations (Multi-task Optimization, Tolerance Assignment, etc.).
• Development of a structure and component interfaces of the specialized Optimal Engineering Design software, based on the orchestration of web services in service-oriented, distributed computational cloud infrastructure.
• Leveraging cloud computing and parallel computing to overcome resource shortage problem when solving engineering design problems of high complexity and with required accuracy in a reasonable time.
• Development of a set of flexible tools for solving interdisciplinary problems of Optimal Engineering Design, supporting remote collective work.
• Demonstrating developed software on cases of simulation of the Helicopter Fly Control System and MEMS for Acceleration as a Hybrid Dynamic Systems.
Simultaneous possibilities for different research teams to contribute in web-services repository development using different programming languages and planning to implement different data from distributed sources will be provided. Due to loosely coupled web-services features users can modify and adapt a composed application which is preserved when some web-services are changed. Design in Engineering becomes personalized and customized because users can build and adjust their design scenario and workflow by selecting the necessary web-services (as calculation procedures) to be executed on cloud resources.
4. Conclusions
Although there are many existing tools that can be useful in some particular cases (including portals with user interface for grid/cloud applications and sharing resources etc.) there is no single complex solution that sufficiently meets all of requirements with respect to Optimization and Simulation of Complex Non-Linear Engineering Systems in Cloud. Proposed approach, been based on service-oriented computing (SOC), is completely different from present attempts to migrate monolithic large CAE/CAD software systems into the cloud infrastructure as it is done in TINACloud, PartSim, RT-LAB, FineSim Pro [10-14].
The Cloud-OPTIMSIM project promotes interdisciplinary R&I activities by bringing together different research domains and constituencies by creating a community around the Optimal Engineering Design service repository enabling anybody to be either as contributors of specific web services and simulation solutions or as users reusing available solutions for their own products. It improves performance at lower cost: improvements boosting performance and functionality at all levels (device, circuit, system), and in particular in relation to a few critical parameters which drive integration, higher energy efficiency and miniaturization by openness of its software which will provide means for bottom-up emergence and selection of best ideas, designs and technologies for future Engineering development through collaborative and competitive community of research institutes, SME's and industry.
The proposed simulation software as the composition of cloud services is intended primarily for users in SME. We are interested in having interested stakeholders and establishing relations with potential users from industries/SMEs for providing the workflow based approach to Engineering Optimal design "in-silico" with respect to objects performance or energy consumption, using distributed computing resources of Cloud.
REFERENCES
1. M. P. Papazoglou and A. D. Georgakopoulos: Service- Oriented Computing, Communications of the ACM, vol. 46 (10), pp. 24-28, 2003.
2. M. N. Huhns and M. P. Singh: Service-Oriented Computing: Key Concepts and Principles, IEEE Internet Computing, vol. 9, Issue 1, pp. 75-81, 2005.
3. Petrenko A.I. Service-Oriented Computing in a Cloud Computing Environment, Computer Science and Applications, Volume 1, Number 6, 2014, pp. 349-358.
4. M. Zgurovsky, A. Petrenko, V. Ladogubets, O. Finogenov, B. Bulakh. WebALLTED: Interdisciplinary modeling in grid and cloud, Computer Science • 14 (2) 2013, pp.295-306.
5. Petrenko A.,Ladogubets V., Tchkalov V., Pudlowski Z."ALLTED - a Computer-Aided System for Electronic Circuit Design", UICEE (UNESCO), Melbourne, 1997, p. 204.
6. Petrenko A.I. "Macromodels of Micro-Electro-Mechanical Systems (MEMS) in Book "Microelectromechanical Systems and Devices", Dr. Nazmul Islam (Ed.), Chapter 8. InTech, 2012, pp. 155-190, ISBN: 978-953-51-0306-6.
7. CloudSME project - OCEAN: http://www.ocean-project.eu/bin/download/ Services/Plugfests/ CloudSME_Ocean_20140515.pdf
8. CloudBroker Platform : http://cloudbroker.com/platform/
9. WS - PGRADE/gUSE Tutorial: http://www.training.prace-ri.eu/uploads/tx_ pracetmo/WS-PGRADE_gUSE_training_x.pdf
10. Cloud computing: an opportunity for EDA: http://www.ocoudert. com/ blog/2011/03/16/cloud-computing-an-opportunity-for-eda/
11. TINA Cloud project home: http://www.tina.com /English/tina/
12. PartSim project home: http://www.partsim.com/
13. RT-LAB project home: http://www.opal-rt.com/company/company-profile/
14. FineSim Pro project home: http://news.synopsys.com/2013-02-07-Latest-Advances-in-FineSim-Deliver-Up-to-2X-Performance-and-Capacity-Improvements
THE USE OF VIRTUAL COMPUTERS IN TEACHING OF INFORMATION DISCIPLINES
Teacher Toyirov A. X., Senior teacher Zaripova M. J., Teacher Jumaev F.T.
Uzbekistan, Termiz, Termiz State University.
Abstract. The questions of possibility of use of virtual computers in teaching of information disciplines are considered in the article. The main stages and technology of creation of virtual computers are described.
Keywords: virtual computer, VirtualBox, virtualization.
Main difference of informatics from other technical disciplines studied at the higher school, consists that its subject of studying changes with accelerated rates. Today the number of computers in the world exceeds 500 million units, moreover each computing system is in own way unique. To find two systems with identical equipment rooms and program configurations very difficult and therefore effective operation of computer facilities requires from experts rather wide level of knowledge and practical skills. At the same time, in a quantitative sense rate of numerical growth of computing systems considerably exceeds rate of the training of specialists, capable effectively to work with them.
But from the moment of introduction at the higher schools of a course of informatics disputes don't stop: what to teach and how to teach. The purpose of training of informatics in particular, demands the following:
1. to be able to work with the computer equipment which becomes daily reality as the TV or
phone;
2. to be able to be guided at the huge amount of information which increases in sizes with an enormous speed;
3. to be able to adapt and live in the becoming complicated world.
Now the students who are training in "Applied mathematics and informatics" study a set of disciplines of subject preparation, among which: "Bases of the theory of information", "The theory of algorithms", "Programming", "Numerical methods", "Architecture of the computer", "Computer networks", "The Internet and technology multimedia", "Information and communication technologies in education", "Computer modeling", "A technique of training in informatics". Besides, for the students who are training in "The applied mathematics and informatics" are provided by educational programs of discipline of subject preparation at the choice of the student, established by educational institution.
Among the listed disciplines, most dynamically "Computer networks, the Internet and technology multimedia" and "Software" develop in recent years. Constantly there are new technologies (Wi-Fi, WiMax, new network services (ip-television, IP-telephony)), the software develops. Mastering students skills of work with operating systems and their network opportunities within these disciplines is very important.
However very often in educational institutions for safety access to the computer from under the account with the limited rights and opportunities is provided to students. On the other hand, in the majority of educational institutions computers work under control of operating systems of Windows family. This system has many shortcomings. Today to a dress with Windows there is the alternative full-fledged Linux operating system meeting all modern requirements.
