A Presentation on Architectual Design by Students of Engineering
- 1. A PRESENTATION ON ARCHITECTURAL DESIGN By Drishya Phuel
- 2. Acknowledgement I would like to express my sincere gratitude to our respected teacher Engineer Ramesh Tamang Sir for providing me with this wonderful opportunity and helpful materials to share my ideas on my today’s topic.
- 3. Topics to be discussed vArchitectural design decisions vSystem Organization vModular Decomposition Styles vControl Styles
- 4. Initial queries • What are Architecture? • What is Architectural designing? • What is organization?
- 5. An Overview of Architectural Design • Architectural design in software engineering is about • decomposing the system into interacting components. • defining an overview of the system structure, features of the components • how these components communicate with each other to share data
- 6. System properties of Architectural Design • Security: uses a layered architecture with critical assets in the innermost layers. • Performance: -meantime taken between request and response of the page -can be improved by avoiding critical operations and reducing communication between • Maintainability: easily modifiable and replaceable components
- 7. Architectural Design Decisions • How will the system be distributed across a number of cores, processors; network? • What approach will be used to structure the system? • Which architectural styles are appropriate? • How can software architecture be documented? • How can the system be decomposed into modules? • What control strategy must be used to control the operation of the components in the system?
- 8. • What architectural organization is best for delivering the non-functional requirements of the system? What comes into play once we have talked about organization?
- 9. System Organization What does organization mean in literal words in terms of software development? ØStructural Relationship ØImplementation of architectural design ØComponents relationships ØRealizing architectural specifications
- 10. Software Architectural Models q static type of architectural structural model represents the major system components. qDynamic type of architectural process model represents the process structure of the system. qDistribution type of architectural model represents how the component is distributed across various computers. qInterface type of architectural model represents the interface of the components. qRelationships type of architectural model represent models such as data flow diagram to represent the component interrelationship.
- 11. Could this design process BE more perfect? To document the architectural design process, architectural models are used.
- 12. • Architectural design models are application domain-specific and the most common two types of domain-specific models are: • Generic model: abstractions derived from a number of real systems and encapsulated the characteristics of these systems. -follow bottom-up approach • Reference model: -derived from the application domain rather than from existing systems -top-down approach -provides a comparison between different software architecture
- 13. Architectural Styles 1. Data-flow Architecture: Each component is known as filter -filter transforms the data and sends this to other filters using connectors -pipes are connectors - pipe is a unidirectional channel -Each filter works as an independent entity
- 14. Data Flow Architecture diagram
- 15. Aspects Some advantages • It supports reusability. • It is maintainable and modifiable. • It supports concurrent execution. Some disadvantages • It often degenerates to batch sequential system. • It does not provide enough support for applications requires user interaction • It is difficult to synchronize two different but related streams.
- 16. 2. Object Oriented Architecture • components are represented as objects • objects interact with each other through methods. • methods are connectors • Objects maintain the integrity of the system. • An object is not aware of the representation of other objects.
- 17. Advantages • allows designers to decompose a problem into a collection of independent objects. • implementation detail of objects is hidden from each other and hence, they can be changed without affecting other objects.
- 18. 3.Layered Architecture • layers are components • layers are arranged in a hierarchical manner, each one built upon the one below it • protocols are used as connectors • The interaction between layers is provided through protocols
- 19. Layered Architecture representation and communication • also known as n-tiered architecture • Closed: The previous layer cannot bypass a layer with a “closed” label, and the request must go through it before it proceeds. • Open: The previous layer can bypass the layer with the “open” label
- 20. 4.Data-centered Architecture • a central data structure or data store (central repository) • represents the current state of the data • a collection of client software client software performs several operations like add, delete, update, etc., onthe data stored
- 21. Advantages • Clients operate independently of one another. • Data repository is independent of the clients. • It adds scalability (that is, new clients can be added easily). • It supports modifiability. • It achieves data integration in component-based development using blackboard.
- 22. Data-repository
- 23. 5.Call and Return Architecture -enables software designers to achieve a program structure, which can be easily modified. -function is decomposed into a control hierarchy where the main program invokes a number of program components, which in turn may invoke other components. -Remote procedure call architecture: In this, components of the main or subprogram architecture are distributed over a network across multiple computers
- 24. Control Styles • concerned with how a system is decomposed into sub-systems • sub-systems must be controlled so that their services are delivered to the right place at the right time • Structural models do not and should not include control information
- 25. Generic control styles that are used in software systems: • Centralized Control Style • - control - start - stop
- 26. Centralized control style: Call-return model
- 27. • Example: When a Java object requests a service from another object, it does so by calling an associated method
- 28. Centralized Control style:The manager model. • One system component is designated as a system manager • controls the starting, stopping and coordination of other system processes. • may also be applied in sequential systems
- 30. EVENT-BASED CONTROL • Rather than control information being embedded in a subsystem, • each sub-system can respond to externally generated events. • events may be command inputs, certain range of values Eg: active objects where changing a value of an object's attribute triggers some actions. We will discuss two even-driven models
- 31. Broadcast model -When these events occur, control is transferred to the sub-system that can handle the event.
- 32. Interrupt-driven models • exclusively used in real-time systems where external interrupts are detected by an interrupt handler.
- 33. Aspects of interrupt-driven models Advantages • allows very fast responses to events to be implemented • may be combined with the centralized management model. Disadvantages • complex to program and difficult to validate • may be impossible to replicate patterns of interrupt timing during system testing. • interrupt mapping may be impractical if a very fast response to individual interrupts is required.
- 34. Important Questions asked in IOE Examination 1. Write short notes on a)Modular Decomposition 2. Compare the following: i) Client Server vs Distributed object architecture 3. What is software design architecture and what is its significanc in software engineering? 4.Explain multi-tier architecture with example. 5. What are the differences between thin client and thick clients?
- 35. Oh My God!
- 36. Differences between thin client and thick client • Thin Client a. Thin clients have browser based installation. b. There is complete processing on the server side c. are easily deployable d. data verification is done required from server • Thick Client a. Thick clients have locally based installation b. Thick clients make use of computer resources more than server. c. expensive to deploy d. data verification is done from clients
- 37. References: i. geeksforgeeks.org ii. tutorialspoint iii. technology.purplepedia.com iv. educba.com v. coursera.org vi. compilations by Dr. Sunil Chaudhary vii. Ian Sommerville, Software Engineering, Latest edition •