Welcome to CENG! What is Computer Engineering Computer Engineering - PowerPoint PPT Presentation

0.1 0.2 Welcome to CENG! What is Computer Engineering • Computer Engineering is… – Developing efficient systems that combine hardware, software, and networking to interact with the physical world and/or solve information-based problems – Learning how to design, analyze, implement, and test such systems Goal of CECS/CENG: To develop engineers who can span the complex inter-relationship of computer hardware and software, creating and designing system solutions http://blog.tmcnet.com/blog/rich-tehrani/uploads/facebook-datacenter-electrical-large.jpg http://www.cmu.edu/news/image-archive/Boss.jpg http://prisonerofclass-5933.zippykid.netdna-cdn.com/wp-content/uploads/2013/05/iphone.jpg http://firstcallappliance.com/wp-content/uploads/image/microwave.jpg http://www.engadget.com/2011/02/19/intel-to-spend-5-billion-on-new-14nm-fab-in-arizona-creating-4/ https://www.staples-3p.com/s7/is/image/Staples/s0978329_sc7?$splssku$ http://www.spacex.com/sites/spacex/files/spacex_default.jpg 0.3 0.4 Today's Digital Environment Computer Engineering as Abstraction Levels if (x > 0) then Applications x = x + y - z; Software a = b*x; Code Networks 1110010101… SW Algorithms C++ / Java / Chips Python (Processors) OS / Assembly / Libraries Machine Code A Applications Processor / Memory / S Functional Networks GPU / FPGAs / Networks + B Units C++ / Java / C++ / Java / Algorithms Algorithms Python Python Digital Logic HW OS / OS / Assembly / Assembly / x AND Libraries Libraries Logic Machine Code Machine Code y F gate z Processor / Memory / Processor / Memory / Transistors / Circuits GPU / FPGAs GPU / FPGAs - - - - Output Digital Logic Digital Logic (Drain ) Voltage / Currents - - - Controlling Transistors Transistors / Circuits Transistors / Circuits Input (Gate ) Source Voltage / Currents Voltage / Currents

0.5 0.6 Goals of this Course You Can Do That With Computer Engineering • Embedded Systems Cloud & Distributed Computing Applications Applications (CyberPhysical, Databases, Data • Programming Mining, Machine Learning, etc.) Networks Networks – C language SW SW Applications Algorithms Algorithms C++ / Java / C++ / Java / (AI, Robotics, Graphics, Mobile) • Digital logic Python Python OS / – Hardware design OS / Assembly / Assembly / Systems & Networking Libraries Libraries Embedded Machine Code Machine Code (Embedded Systems, IoT, • Computer Performance Networks) Processor / Memory / Processor / Memory / GPU / FPGAs / Networks GPU / FPGAs / Networks – Understanding and Architecture improving performance of (Processors & Embedded Digital Logic Digital Logic HW HW HW) systems Transistors / Circuits Transistors / Circuits • Electronics Devices & Integrated Circuits (Semiconductors & Voltage / Currents Voltage / Currents – Voltage, current, basic Fabrication) circuit theory 0.7 0.8 Full Timeout - Syllabus Course Advice • Catch the wave! – Start early and lay a good foundation in the first few weeks (even on the "easy" stuff) – Don’t let shame or embarrassment keep you from the help you need • Put in the time – Achieving mastery takes ______ hours? 1 • Play! – Be active, curious, and engaged • Don't let "class" get in the way of learning – Do not be afraid to just try, even if you fail • Failure breeds far more learning than success 1 From Malcolm Gladwell's Outliers

0.9 0.10 CECS Curriculum EE Curriculum Core Classes EE/CS EE 155 EE 355 EE 250 EE 301 EE 364 EE 202 EE 330 Robotics 109 Networks & Web Programming Linear Probability and Linear Circuits Electromagnetics Software Embedded IoT CS 445 Programming Design Systems Systems Statistics AI CS 360 CS 353 Area Electives Graphics CS 420 EE 354 EE 457 EE Digital System Computer 109 EE 477 CS 270 Design Architecture EE 250 CS 104 CS CS 170 EE 479 CS 476 CS 201 103 EE 354 EE 450 EE 454 CS 350 Integrated Algorithms, Theory, Circuits Cryptography Core Concepts Networks Intro. To System- Operating (Programming Languages, on-Chip Systems Data Structures, CS 310 Digital Logic Design) EE 451 CS 360 CS 445 EE 477 EE 459 EE 457 CS 477 AI Robotics Parallel & Intro. To VLSI Distributed Software CS 350 Computing Engineering EE 454 EE 451 EE 459 Operating Computer Systems Architecture Capstone Circuits, Signals, and Electrical Sciences Computer Engineering Systems 0.11 0.12 Careers Careers • Information Technology … and presidents of USC. Our last two presidents, Steven • General & High Performance Computing Sample and Max Nikias were both Electrical Engineers. • Mobile and Networking • Media & Entertainment • Automotive • Robotics • Aerospace / Defense

0.13 0.14 Companies Research/Extracurriculars at USC • Information Technology • Facebook, Apple, Google, MS • Autonomous Networks Research Group • Semiconductor & General • Intel, IBM – Wireless & mobile networks, embedded computing sensing and processing • Telecomm and Networking • Information Sciences Institute • Qualcomm, Cisco – AI, Internet, Advanced Processing Systems research • Media & Entertainment • Institute for Creative Technologies • EA, Disney, Riot Games • Automotive – Virtual Reality, Graphics, Animation, Games • Ford, Uber, etc. • Robotics • Student Orgs • JPL, iRobot • Aerospace – Makers (http://viterbimakers.usc.edu/) • SpaceX, Boeing, Raytheon • Startups – HackSC, AthenaHacks, SparkSC • Embark (now Apple), Zboard – IEEE, ACM 2 of the top 10 fastest-growing job markets will be Computer Science and System (CENG) Engineers * U.S. Department of Labor 0.15 0.16 Digital System Spectrum Application Specific Hardware (no software) • Key idea: Any “algorithm” can be implemented in HW or SW or some mixture of both Computing System Flexibility, Design Time • A digital systems can be located anywhere in a spectrum of: Spectrum Performance – ALL HW: (a.k.a. Application-Specific IC’s) Cost – ALL SW: An embedded computer system • Advantages of application specific HW – Faster, less power • Advantages of an embedded computer system (i.e. general purpose HW for executing SW) General Purpose – Reprogrammable (i.e. make a mistake, fix it) HW w/ Software EMBEDDED SYSTEMS – Less expensive than a dedicated hardware system (single computer system can be used for multiple designs) • One of the main points for the whole course: – Applications and processing can be done in software, custom-designed HW, or some mixture of both – Each one has pros/cons – We will give you the knowledge for how it can be done in both http://www.xbitlabs.com/images/news/2008-06/nvda_tegra_chip_scheme.jpg

0.17 0.18 Processing Logic Approaches Embedded Systems • Custom Hardware X • An embedded system is… + – Logic that directly implements a Y Out – A special purpose computer that is designed into a larger * specific task device to perform some amount of dedicated tasks – Example above may use separate A + adders and a multiplier unit • Utilize a microcontroller… B • General Purpose (GP) Custom HW – Laptops or desktops contain a microprocessor Processor Implementation – Embedded systems contain microcontrollers – Logic designed to execute SW – What's the difference? instructions Data storage Instruc. – Provides basic processing Store • Microprocessor is part of a larger computer system w/ RAM and resources that are reused by ADD X,Y general purpose I/O + * ADD A,B each instruction • Microcontroller is a single-chip with RAM and I/O to control MUL X,A • What if I want to perform: specific electro/mechanical devices GP Proc. Implementation – X*Y + A*B of (X+Y)*(A+B) – What's easiest to redesign? 0.19 0.20 A Comparison Engineering Design • Different microprocessor and microcontroller based systems Engineering is PC iPhone MIPS32 Arduino (Core i7) (A6 chip) (PIC32MX…) (ATMega328) Design with Constraints Clock speed 3 GHz 1.3 GHz 80 MHz < 20 MHz Data size 64-bits 32-bit 32-bits 8-bits RAM 8 GB 1 GB 16 KB 2 KB Storage 1 TB 32 GB 128 KB 32 KB Cost $1,000 $650 $6.04 $2.88

0.21 0.22 Engineering Design Engineering Design Artists and engineers both design things, but differently. What constraints does an engineer face when designing something? Artist: “We’ll create a spaceship powered by a warp drive.” Engineer: “We need to build a rocket engine that works.” 0.23 0.24 Engineering Design Engineering Design What constraints does an engineer The constraints may differ depending face when designing something? on the product. Medical implant device: Reliability > Cost Cost Performance Reliability Durability Ease of use Weight Toy: Cost > Reliability Power Size Safety Environmental Political Testability