TDT4255 Computer Design. Lecture 1. Magnus Jahre

Transcription

1 1 TDT4255 Computer Design Lecture 1 Magnus Jahre

2 2 Outline Practical course information Chapter 1: Computer Abstractions and Technology

3 3 Practical Course Information

4 4 TDT4255 Computer Design TDT4255 Computer Design is a thorough review of scalar processor core design: Detailed discussion of single cycle, multi-cycle and pipelined CPUs Introduction to advanced techniques like out-of-order execution, branch prediction and speculative execution Hands-on approach Exam counts 50% and exercises counts 50% towards the final grade Complementary to TDT4260 Computer Architecture TDT4255 goes through the scalar processor core in detail TDT4260 treats t the scalar processor core as a black box and covers the rest of the computer system

5 5 Course Staff Lecturer/Coordinator Magnus Jahre Associate Professor, Computer Architecture and Design Group (CARD), IDI, NTNU PhD 2010: Managing Shared Resources in Multicore Memory Systems Lecturer Ian Bratt 20% University Lecturer, CARD, IDI, NTNU Has designed processors for Tilera, will start at ARM this fall Teaching Assistant (TA) Vinay Kumar Gautam PhD student t at the Intelligent t Systems group at IDI

6 6 Preliminary Reading List Patterson and Hennessy, Computer Organization and Design The Hardware/Software Interface, 4th Edition: Chapter 1 Chapter 2 Chapter 3 Chapter 4 Chapter Appendix C Appendix D Hennessy and Patterson, Computer Architecture A Quantitative Approach, 4th Edition: Chapter 2 (will be provided) All exercises and provided exercise material

7 7 Exercises Three exercises Exercise 1: Design you own ALU (counts 10% of final grade) Exercise 2: Design you own multi-cycle CPU (counts 20% of final grade) Exercise 3: Design you own pipelined CPU (counts 20% of final grade) TA will introduce the exercises in detail in the first exercise lecture

8 8 Information Distribution Course web page no/emner/tdt4255/ Most of the information will be available here It s Learning Broadcast messages to all students Remember: set up notification of news! Information that should not be publicly available (e.g. supplementary files for exercises) Delivery of exercises

9 9 Preliminary Lecture Plan See PDF on the course web page (shown in class) Lectures will be on Wednesdays and Fridays Note: Irregular schedule

10 10 Chapter 1 Computer Abstractions and Technology Acknowledgement: Slides are adapted from Morgan Kaufmann companion material

11 11 The Computer Revolution Progress in computer technology Underpinned by Moore s Law Makes novel applications feasible Computers in automobiles Cell phones Human genome project World Wide Web Search Engines Computers are pervasive

12 12 Classes of Computers Desktop computers General purpose, variety of software Subject to cost/performance tradeoff Server computers Network based High capacity, performance, reliability Range from small servers to building sized Embedded computers Hidden as components of systems Stringent power/performance/cost constraints

13 13 The Processor Market ed n Units Shippe Million

14 14 Understanding Performance Algorithm Determines number of operations executed Programming language, compiler, architecture Determine number of machine instructions executed per operation Processor and memory system Determine how fast instructions are executed I/O system (including OS) Determines how fast I/O operations are executed

15 15 Chapter 1.2 Below Your Program

16 16 Below Your Program Application software Written in high-level language System software Compiler: translates HLL code to machine code Operating System: service code Handling input/output Managing memory and storage Scheduling tasks & sharing resources Hardware Processor, memory, I/O controllers

17 17 Levels of Program Code High-level language Level of abstraction closer to problem domain Provides for productivity and portability Assembly language Textual representation of instructions Hardware representation Binary digits (bits) Encoded instructions and data

18 18 Chapter 1.3 Under the Covers

19 19 Components of a Computer Same components for all kinds of computer Desktop, server, embedded Input/output t t includes User-interface devices Display, keyboard, mouse Storage devices Hard disk, CD/DVD, flash Network adapters For communicating with other computers

20 20 Inside the Processor (CPU) Datapath: performs operations on data Control: sequences datapath, memory,... Cache memory Small fast SRAM memory for immediate access to data

21 21 Inside the Processor AMD Barcelona: 4 processor cores

22 22 Abstractions Abstraction helps us deal with complexity Hide lower-level detail Instruction set architecture (ISA) The hardware/software interface Application binary interface The ISA plus system software interface Implementation The underlying details and interface

23 23 Technology Trends Electronics technology continues to evolve Increased capacity and performance Reduced cost DRAM capacity Year Technology Relative performance/cost 1951 Vacuum tube Transistor Integrated circuit (IC) Very large scale IC (VLSI) 2,400, Ultra large scale IC 6,200,000,

24 24 Chapter 1.4 Performance

25 25 Defining Performance Which airplane has the best performance? Boeing 777 Boeing 777 Boeing 747 BAC/Sud Concorde Douglas DC-8-50 Boeing 747 BAC/Sud Concorde Douglas DC Passenger Capacity Cruising Range (miles) Boeing 777 Boeing 777 Boeing 747 BAC/Sud Concorde Douglas DC-8-50 Boeing 747 BAC/Sud Concorde Douglas DC Cruising Speed (mph) Passengers x mph

26 26 Response Time Book definition: Time from issuing a command to its completion This is often referred to as the turn-around dtime More common response time definition: Time from issue to first response Execution time is the time the processor is busy execution the program Turn-around time includes the time the process waits to be executed, execution time does not Also: user execution time vs. system execution time

27 27 Response Time and Throughput Throughput Total work done per unit time How are response time and throughput affected by Replacing the processor with a faster version? Adding more processors?

28 28 Relative Performance Define Performance = 1/Execution Time X is n time faster than Y Performance X Execution time Performance Y Y Execution time X n Example: time taken to run a program 10s on A, 15s on B Execution Time B / Execution Time A = 15s / 10s = 1.5 So A is 1.5 times faster than B

29 29 Measuring Execution Time Elapsed time/wall clock time Total turn-around time, including all aspects Processing, I/O, OS overhead, idle time Determines system performance CPU time Time spent processing a given job Discounts I/O time, other jobs shares Comprised of user CPU time and system CPU time Different programs are affected differently by CPU and system performance

30 30 CPU Clocking Operation of digital hardware governed by a constant-rate t t clock Clock (cycles) Data transfer and computation Update state Clock period Clock period: duration of a clock cycle e.g., 250ps = 0.25ns = s Clock frequency (rate): cycles per second e.g., 4.0GHz = 4000MHz = Hz

31 31 CPU Time CPU Time CPU Clock Cycles Clock Cycle Time CPU Clock Cycles Clock Rate Performance improved by Reducing number of clock cycles Increasing clock rate Hardware designer must often trade off clock rate against cycle count

32 32 CPU Time Example Computer A: 2GHz clock, 10s CPU time on benchmark X Designing i Computer B Design Goal: 6s CPU time on benchmark X Faster clock causes 20% increase in the number of clock cycles How fast must Computer B clock be? Clock Rate B Clock Cycles CPU Time B B Clock Cycles 6s A Clock Cycles A CPU Time A Clock Rate A 10s 2GHz Clock Rate B s s 9 4GHz

33 33 Instruction Count and CPI Clock Cycles Instruction Count Cycles per Instruction CPU Time Instruction Count CPI Clock Cycle Time Instruction Count CPI Clock Rate Instruction Count for a program Determined by program, ISA and compiler Average cycles per instruction Determined by CPU hardware If different instructions have different CPI Average CPI affected by instruction mix Under which conditions is it valid to compare CPIs directly?

34 34 CPI Example Computer A: Cycle Time = 250ps, CPI = 2.0 Computer B: Cycle Time = 500ps, CPI = 1.2 Same ISA, program and compiler Which is faster, and by how much? CPU Time A CPU Time B CPU Time B CPU Time A Instruction Count CPI Cycle Time A A I ps I500ps Instruction Count CPI B I1.2500ps I 600ps I 600ps 1.2 I500ps Cycle Time B A is faster by this much

35 35 CPI in More Detail If different instruction classes take different numbers of cycles Clock Cycles n i1 (CPIi Instruction Counti) Weighted average CPI CPI Clock Cycles Instruction Count n i 1 CPI i Instruction Counti Instruction Count Relative frequency

36 36 CPI Example Alternative compiled code sequences using instructions in classes A, B, C Class Cass A B C CPI for class IC in sequence IC in sequence Sequence 1: IC = 5 Sequence 2: IC = 6 Clock Cycles Clock Cycles = = = 10 = 9 Avg. CPI = 10/5 = 2.0 Avg. CPI = 9/6 = 1.5

37 37 Performance Summary CPU Time Instructions Program Clock cycles Instruction Seconds Clock cycle Performance depends on Algorithm: affects IC, possibly CPI Programming language: affects IC, CPI Compiler: affects IC, CPI Instruction set architecture: affects IC, CPI, Clock cycle time

38 38 Chapter 1.5 The Power Wall

39 39 Power Trends In CMOS IC technology Power Capacitive load Voltage 2 Frequency 30 5V 1V 1000

40 40 Reducing Power Suppose a new CPU has 85% of capacitive load of old CPU 15% voltage and 15% frequency reduction 2 Pnew Cold 0.85(Vold 0.85) Fold P C V F old old od old od old od 0.52 The power wall: We can t reduce voltage further We can t remove more heat How else can we improve performance?

41 41 Chapter 1.6 Uniprocessors to Multiprocessors

42 42 Uniprocessor Performance Constrained by power, instruction-level parallelism, memory latency

43 43 Multiprocessors Multicore microprocessors More than one processor per chip Covered in detail in TDT4260 Computer Architecture Requires explicitly parallel programming Compare with Instruction Level Parallelism (ILP) Hardware executes multiple instructions at once Hidden from the programmer Hard to do Programming for performance Load balancing Optimizing communication and synchronization

44 44 Chapter 1.7 Manufacturing and Benchmarking the AMD Opteron X4

45 45 Manufacturing ICs Yield: proportion of working dies per wafer

46 46 AMD Opteron X2 Wafer X2: 300mm wafer, 117 chips, 90nm technology X4: 45nm technology

47 47 Integrated Circuit Cost Cost per die Cost per wafer Dies per wafer Yield Dies per wafer Wafer area Die area Yield 1 (1 (Defects per area Die area/2)) 2 Nonlinear relation to area and defect rate Wafer cost and area are fixed Defect rate determined by manufacturing process Die area determined by architecture and circuit design

48 48 SPEC CPU Benchmark Programs used to measure performance Supposedly typical of actual workload Standard Performance Evaluation Corp (SPEC) Develops benchmarks for CPU, I/O, Web, SPEC CPU2006 Elapsed time to execute a selection of programs Negligible I/O, so focuses on CPU performance Normalize relative to reference machine Summarize as geometric mean of performance ratios CINT2006 (integer) and CFP2006 (floating-point) Again: Covered in detail in TDT4260

49 49 CINT2006 for Opteron X Name Description IC 10 9 CPI Tc (ns) Exec time Ref time SPECratio perl Interpreted string processing 2, , bzip2 Block-sorting compression 2, , gcc GNU C Compiler 1, , mcf Combinatorial optimization ,345 9, go Go game (AI) 1, , hmmer Search gene sequence 2, , sjeng Chess game (AI) 2, , libquantum Quantum computer simulation 1, ,047 20, h264avc Video compression 3, , omnetpp Discrete event ent simulation , astar Games/path finding 1, , xalancbmk XML parsing 1, ,143 6, Geometric mean 11.7 High cache miss rates

50 50 Chapter 1.8 Fallacies and Pitfalls

51 51 Pitfall: Amdahl s Law Improving one aspect of a computer and expecting a proportional improvement in overall performance T improved Taffected improvement factor T unaffected Example: multiply accounts for 80s/100s How much improvement in multiply performance to get 5 overall? Can t be done! n Corollary: make the common case fast