Presentation is loading. Please wait.

Presentation is loading. Please wait.

CSCE 313 Embedded System Design Department of Computer Science and Engineering University of South Carolina.

Similar presentations

Presentation on theme: "CSCE 313 Embedded System Design Department of Computer Science and Engineering University of South Carolina."— Presentation transcript:

1 CSCE 313 Embedded System Design Department of Computer Science and Engineering University of South Carolina

2 Page 2 Syllabus Highlights Instructor’s Info SWGN 3A15 7-5872 Office Hour: Mon Wed 2pm-3:30pm or by appointment

3 Page 3 About this course Text T. Noergaard, Embedded Systems Architecture: A Comprehensive Guide for Engineers and Programmers, Elsevier, 2005 Exam One exam Closed book No makeup without valid reasons and prearrangement with instructor Homework

4 Page 4 About this course Projects Two types  instructor designed with substantial guidance  mainly by yourselves Team work Grading  Demonstration  Reports Supplementary readings

5 Page 5 About this course Grading 15% homework, 20% exam + quizzes, 65% projects Academic Dishonesty Academic Courtesy No chatting, cell phone, food/drink please

6 Page 6 About this course Prerequisites CSCE211/CSCE212 Things you should already know Digital logic design  Number representation –2 ’ s complement (int and fp)  Binary arithmetic –add, subtract, shift, multiply, division,  Combinational logic –and, or, xor, nand, nor, mux, …  Sequential logic –Flip-flops, registers  Read/draw a schematic

7 Page 7 About this course Things you should have already known  Basic computer organization –Instruction execution –Simple assembly programming –Processor/memory/buses  Programming –C/C++ –Compilation and debugging

8 Page 8 Topics The principles of embedded system design and its challenges Embedded hardware architecture Embedded software development Hardware/software interfacing and debugging Processes and operating systems Advanced design techniques Other topics such as (low power design and real-time computing) may be covered

9 Page 9 Things you will learn What is the embedded system and how to design such systems Practical skills and experiences Commercial design softwares  Quartus II, Nios II IDE Commercial hardware evaluation board  Altera DE2 board

10 Page 10 Embedded System Design When we talk about the computing system PC Laptop Workstation Mainframe Super Computers … In fact, computing systems are everywhere …

11 Page 11 Computing systems are everywhere …wake up … …have breakfast … …set home safety system …

12 Page 12 A late model car can have as many as 65+ processors for engine control, A/C control, cruise control, ABS, audio, etc More than 30% of the cost of a car is now in electronics 90% of all innovations will be based on electronic systems Examples In Your Daily Life (cont’) …get into your car …

13 Page 13 …on your way to your office… Examples In Your Daily Life (cont’)

14 Page 14 …in your office … Examples In Your Daily Life (cont’)

15 Page 15 …back home … Examples In Your Daily Life (cont’) Several hundred processors can be involved in the course of one day for one person !

16 Page 16 Other Examples Mission critical controls  Nuclear plant control, aircraft navigation, military equipment Medical equipment Communications Toy, etc Embedded systems have been deeply ingrained in our life !

17 Page 17 What What is the embedded system? Embedded System  Processor based –General processors –Micro controllers –DSP  A subsystem  With special purpose –Not a general programming computer

18 Page 18 Processor(s) Auxiliary Systems (power, cooling,…) Memory Other Hardware ASIC/FPGA Software A/D D/A Sensors Actuators Interconnections Input Output Keyboards, Mouse, Displays, Switches, etc Clock circuit, timer, etc Major components

19 Page 19 Example: Digital Camera Processor

20 Page 20 Some common characteristics Application specific Executes a single program, repeatedly Tightly-constrained Low cost, low power, small, fast, etc. Reactive and real-time Continually reacts to changes in the system’s environment Timing is as important as the correctness

21 Page 21 Why Processors Flexibility Easy to upgrade Easy to build complex system behavior Maintainability

22 Page 22 Design Challenges Moore’s Law Productivity Gap Design Cost Stringent Time-to-market Design requirements (constraints)

23 Page 23 Moore’s Law The transistor density of semiconductor chips would double roughly every 18 months. --by Gordon Moore,1965 (co-founder of Intel)

24 Page 24 Image courtesy of Intel Corporation

25 Page 25 Source: S. Borkar, Thousand Core Chips – A Technology Perspective, DAC, 2007 Power Consumption Challenges

26 Page 26 Fact In 1975, an IBM mainframe computer that could perform 10,000,000 instructions per second cost around $10,000,000. 33mhz Intel 80486 can perform better than 11mips. How much money you would like to pay for a computer with Intel 80486 processor today?

27 Page 27 Productivity Gap The gap between the availability of the IC technology (increasing computing power) and the application of the IC technology.

28 Page 28 Fact: 1981 leading edge chip required 100 designer months. 2002 leading edge chip requires 30,000 designer months Productivity Gap

29 Page 29 Design Challenges (cont’d) Moore’s Law Productivity Gap Design cost Reduce non-recurring engineering (NRE) cost  A superior human engineer may outperform the CAD tools in designing simple embedded systems but not for systems with hundred millions to billions gates Stringent time-to-market Design requirements (constraints)

30 Page 30 Design Cost Total cost = NRE cost + Unit cost * Unit # NRE (Non-recurring engineering cost) cost  one time monetary cost for the products Unit cost  the monetary cost of manufacturing each copy of the system, excluding NRE cost Ex: 1000 Units, NRE cost $5,000,000, unit cost $1000  Cost/product = (5000000+1000*1000)/1000 = 6000$

31 Page 31 Design Challenges (cont’d) Moore’s Law Productivity Gap Design cost Reduce non-recurring engineering (NRE) cost Stringent time-to-market Design requirements (constraints)

32 Page 32 Time-to-market Time required to develop a product to the point it can be sold to customers Market window –Period during which the product would have highest sales Delays can be costly –Each week delay translates to 1% loss in relative performance Revenues ($) Time (months)

33 Page 33 Time-to-Market 1997199819992002 Technology 350nm250nm180nm130nm Cost $1.5-2.0billion$2-3billion$3-4billion$4+ billion Design cycle 18-12mo12-10mo10-8mo8-6mo Complexity 200-500k1-2M4-6M10-25M

34 Page 34 Design Challenges (cont’d) Moore’s Law Productivity Gap Design cost Reduce non-recurring engineering (NRE) cost Stringent time-to-market Design requirements (constraints)

35 Page 35 Design Requirements (Constraints) Functionality Timing Power consumption Cost Size & Weight Safety & Reliability Low cost reliability with minimal redundancy Others: component acquisition, upgrades, compatibility, etc.

36 Page 36 The General Design Flow In the past: Hardware and software design technologies were very different Recent maturation of synthesis enables a unified view of hardware and software Hardware/software “codesign” High abstraction level design requirements specification architecture component design system integration

37 Page 37 Layout Transistor 1970 ’ s

38 Page 38 1980’s Transistor Gate

39 Page 39 1990’s Gates Reg ALU MUX ALUs,MUXes, Regs

40 Page 40 1990’s-2000s’ Reg ALU MUX ALUs,MUXes, RegsProc,FPGA,ASICs uPASICFPGA MEMA/DD/A

41 Page 41 Software Design Instruction Set Architecture Assembly Programming (Compiler, Linker) VHDL, C programming Embedded C, Hardware C, … Graphical Language, behavioral synthesis …

42 Page 42 Abstraction Conceptual interpretation of a system Abstraction Level System Complexity

43 Y-chart Behavioural Structural Physical System Level RT Level Logical Level Circuit Level Processors,ASICs,memory ALU,Reg,MUX Gates,Flip-Flops Transistors Macro-cells,chips Cells, wiring Layout Differential Eqn. Boolean Eqn. RT Specifications Algorithms,processes Boards,SOCs

44 BehavioralStructure Physical Synthesis Analysis Optimization Extraction Generation Refinement Abstraction

45 Page 45 … Abstraction Level

46 Page 46 State-of-the-Art Design Address the complex system at high abstraction levels Explore all degrees of design freedom Making some critical decisions at an early design phrase Reducing production cost, design cost, and development time Emphasize on design use

47 Page 47 IP Reuse IP (Intellectual Property) A family of components that can perform certain clusters of functions  Hard IP, e.g. layout, netlist  Soft IP, e.g. RTOS, device driver, libraries Reuse Not something new Reduce NRE cost Reduce development time

48 Page 48 Platform Based Design A good implementation strategy for IP reuse Platform A platform is, in general, an abstraction that covers a number of possible refinements into a lower level. For every platform, there is a view that is used to map the upper layers of abstraction into the platform and a view that is used to define the class of lower level abstractions implied by the platform.

49 Page 49 Platforms Architectural Space (Performance) Application Space (Features) Platform Instance Application Instances System Platfor m Platform Design Space Exploration Platform Specification

50 Page 50 Summary Course Syllabus What is embedded system and the major components The challenges in embedded system design The design methodology

Download ppt "CSCE 313 Embedded System Design Department of Computer Science and Engineering University of South Carolina."

Similar presentations

Ads by Google