1. 1  2004 Morgan Kaufmann Publishers Chapter 4

2. 2  2004 Morgan Kaufmann Publishers

3. 3 Outline 4.1Introduction 4.2CPU Performance and Its Factor 4.3Evaluating Performance 4.4Real Stuff: Two SPEC Benchmarks and the Performance of Recent Intel Processors 4.5Fallacies and Pitfalls 4.6Concluding Remarks 4.7Historical Perspective and Further Reading 4.8Exercises

4. 4  2004 Morgan Kaufmann Publishers 4.1Introduction

5. 5  2004 Morgan Kaufmann Publishers Keywords Response time Also called execution time. The total time required for the computer to complete a task, including disk accesses, memory accesses, I/O activities, operating system overhead, CPU execution time, and so on. CPU execution time Also called CPU time. The actual time the CPU spends computing for a special task. User CPU time The CPU time spent in a program itself. System CPU time The CPU time spent in the operating system performing tasks on behalf of the program. Clock cycle Also called tick, clock tick, clock period, clock, cycle. The time for one clock period, usually of the processor clock, which runs at a constant rate. Clock period The length of each clock cycle.

6. 6  2004 Morgan Kaufmann Publishers Measure, Report, and Summarize Make intelligent choices See through the marketing hype Key to understanding underlying organizational motivation Why is some hardware better than others for different programs? What factors of system performance are hardware related? (e.g., Do we need a new machine, or a new operating system?) How does the machine's instruction set affect performance? Performance

7. 7  2004 Morgan Kaufmann Publishers Which of these airplanes has the best performance? AirplanePassengersRange (mi)Speed (mph) Boeing 737-100101630598 Boeing 7474704150610 BAC/Sud Concorde13240001350 Douglas DC-8-501468720544 How much faster is the Concorde compared to the 747? How much bigger is the 747 than the Douglas DC-8?

8. 8  2004 Morgan Kaufmann Publishers Response Time (latency) — How long does it take for my job to run? — How long does it take to execute a job? — How long must I wait for the database query? Throughput — How many jobs can the machine run at once? — What is the average execution rate? — How much work is getting done? If we upgrade a machine with a new processor what do we increase? If we add a new machine to the lab what do we increase? Computer Performance: TIME, TIME, TIME

9. 9  2004 Morgan Kaufmann Publishers Elapsed Time –counts everything (disk and memory accesses, I/O, etc.) –a useful number, but often not good for comparison purposes CPU time –doesn't count I/O or time spent running other programs –can be broken up into system time, and user time Our focus: user CPU time –time spent executing the lines of code that are "in" our program Execution Time

10. 10  2004 Morgan Kaufmann Publishers 常用單位換算表 單位名稱十進位二進位縮寫 milli 10 -3 m micro 10 -6 μ nano 10 -9 n pico 10 -12 p Kilo 10 3 2 10 K Mega 10 6 2 20 M Giga 10 9 2 30 G Tera 10 12 2 40 T

11. 11  2004 Morgan Kaufmann Publishers For some program running on machine X, Performance X = 1 / Execution time X "X is n times faster than Y" Performance X / Performance Y = n Then the execution time on Y is n times longer than it is on X: Problem: –machine A runs a program in 20 seconds –machine B runs the same program in 25 seconds Book's Definition of Performance

12. 12  2004 Morgan Kaufmann Publishers Relative performance If computer A runs a program in 10 seconds and computer B runs the same program in 15 seconds, how much faster is A than B? We know that A is n times faster than B if Thus the performance ratio is 15/10 = 1.5 and A is therefore 1.5 times faster than B.

13. 13  2004 Morgan Kaufmann Publishers Clock Cycles Instead of reporting execution time in seconds, we often use cycles Clock “ticks” indicate when to start activities (one abstraction): cycle time = time between ticks = seconds per cycle clock rate (frequency) = cycles per second (1 Hz. = 1 cycle/sec) A 4 Ghz. clock has a cycle time (0.25 nanoseconds, 0.25 ns) time

14. 14  2004 Morgan Kaufmann Publishers So, to improve performance (everything else being equal) you can either (increase or decrease?) ________ the # of required cycles for a program, or ________ the clock cycle time or, said another way, ________ the clock rate. How to Improve Performance

15. 15  2004 Morgan Kaufmann Publishers Could assume that number of cycles equals number of instructions This assumption is incorrect, different instructions take different amounts of time on different machines. Why? hint: remember that these are machine instructions, not lines of C code time 1st instruction2nd instruction3rd instruction4th 5th6th... How many cycles are required for a program?

16. 16  2004 Morgan Kaufmann Publishers Multiplication takes more time than addition Floating point operations take longer than integer ones Accessing memory takes more time than accessing registers Important point: changing the cycle time often changes the number of cycles required for various instructions (more later) time Different numbers of cycles for different instructions

17. 17  2004 Morgan Kaufmann Publishers Our favorite program runs in 10 seconds on computer A, which has a 4 GHz. clock. We are trying to help a computer designer build a new machine B, that will run this program in 6 seconds. The designer can use new (or perhaps more expensive) technology to substantially increase the clock rate, but has informed us that this increase will affect the rest of the CPU design, causing machine B to require 1.2 times as many clock cycles as machine A for the same program. What clock rate should we tell the designer to target?" Don't Panic, can easily work this out from basic principles Example

18. 18  2004 Morgan Kaufmann Publishers A given program will require –some number of instructions (machine instructions) –some number of cycles –some number of seconds We have a vocabulary that relates these quantities: –cycle time (seconds per cycle) –clock rate (cycles per second) –CPI (cycles per instruction) a floating point intensive application might have a higher CPI –MIPS (millions of instructions per second) this would be higher for a program using simple instructions Now that we understand cycles

19. 19  2004 Morgan Kaufmann Publishers 4.2CPU performance and Its Factors

20. 20  2004 Morgan Kaufmann Publishers Keywords Clock cycle per instruction (CPI) Average number of clock cycles per instruction for a program or program fragment. Instruction mix A measure of the dynamic frequency of instructions across one or many programs.

21. 21  2004 Morgan Kaufmann Publishers Performance Performance is determined by execution time Do any of the other variables equal performance? –# of cycles to execute program? –# of instructions in program? –# of cycles per second? –average # of cycles per instruction? –average # of instructions per second? Common pitfall: thinking one of the variables is indicative of performance when it really isn’t.

22. 22  2004 Morgan Kaufmann Publishers A simple formula relates the most basic metrics (clock cycles and clock cycle time) to CPU time: Alternatively, because clock rate and clock cycle time are inverses,

23. 23  2004 Morgan Kaufmann Publishers Improving Performance Our favorite program runs in 10 seconds on compute A, which has 4 GHz clock. We are trying to help a computer designer build a computer, B, that will run this program in 6 seconds. The designer has determined that a substantial increase in the clock rate is possible, but this increase will affect the rest of the CPU design, causing computer B to require 1.2 times as many clock cycles as computer A for this program. What clock rate should we tell the designer to target?

24. 24  2004 Morgan Kaufmann Publishers Solution: (1) Let’s first find the number of clock cycles required for the program on A:

25. 25  2004 Morgan Kaufmann Publishers (2)CPU time for B can be found using this equation: computer B must therefore have twice the clock rate of A to run the program in 6 seconds.

26. 26  2004 Morgan Kaufmann Publishers Suppose we have two implementations of the same instruction set architecture (ISA). For some program, Machine A has a clock cycle time of 250 ps and a CPI of 2.0 Machine B has a clock cycle time of 500 ps and a CPI of 1.2 What machine is faster for this program, and by how much? If two machines have the same ISA which of our quantities (e.g., clock rate, CPI, execution time, # of instructions, MIPS) will always be identical? CPI Example

27. 27  2004 Morgan Kaufmann Publishers Solution: Now we can compute the CPU time for each computer: Clearly, computer A is faster. The amount faster is given by the ratio of the execution times: We can conclude that computer A is 1.2 times as fast as computer B for this program.

28. 28  2004 Morgan Kaufmann Publishers Figure 4.2 The basic components of performance and how each is measured. Components of performanceUnits of measure CPU execution time for a programSeconds for the program Instruction countInstructions executed for the program Clock cycles per instruction (CPI)Average number of clock cycles per instruction Clock cycle timeSeconds per clock cycle

29. 29  2004 Morgan Kaufmann Publishers A compiler designer is trying to decide between two code sequences for a particular machine. Based on the hardware implementation, there are three different classes of instructions: Class A, Class B, and Class C, and they require one, two, and three cycles (respectively). The first code sequence has 5 instructions: 2 of A, 1 of B, and 2 of C The second sequence has 6 instructions: 4 of A, 1 of B, and 1 of C. Which sequence will be faster? How much? What is the CPI for each sequence? # of Instructions Example

30. 30  2004 Morgan Kaufmann Publishers Solution: We can use the equation for CPU clock cycles based on instruction count and CPI to find the total number of clock cycles for each sequence: This yields The CPI values can be computed by

31. 31  2004 Morgan Kaufmann Publishers Two different compilers are being tested for a 4 GHz. machine with three different classes of instructions: Class A, Class B, and Class C, which require one, two, and three cycles (respectively). Both compilers are used to produce code for a large piece of software. The first compiler's code uses 5 million Class A instructions, 1 million Class B instructions, and 1 million Class C instructions. The second compiler's code uses 10 million Class A instructions, 1 million Class B instructions, and 1 million Class C instructions. Which sequence will be faster according to MIPS? Which sequence will be faster according to execution time? MIPS example

32. 32  2004 Morgan Kaufmann Publishers 4.3Evaluating Performance

33. 33  2004 Morgan Kaufmann Publishers Keywords Workload A set of programs run on a computer that is either the actual collection of applications run by a user or is constructed from real programs to approximate such a mix. A typical workload specifies both the programs as well as the relative frequencies. Arithmetic mean The average of the execution times that is directly proportional to total execution time. Weight arithmetic mean An average of the execution time of a workload with weighting factors designed to reflect the presence of the programs in a workload; computed as the sum of the products of weighting factors and execution times.

34. 34  2004 Morgan Kaufmann Publishers Figure4.3 System description of a desktop system using the fastest Pentium 4 available in 2003. Hardware Hardware vendorDell Model numberPrecision Workstation 360 (3.2 GHz Pentium 4 Extreme Edition) CPUIntel Pentium 4 (800 MHz system bus) CPU MHz3200 FPUIntegrated CPU(s) exabled1 CPU(s) orderable1 ParallelNo Primary cache12K(I) micro-ops + 8KB(D) on chip Secondary cache512KB (I+D) on chip L3 cache2048KB(I+D) on chip Other cacheN/A Memory4*512MB ECC DDR400 SDRAM CL3 Disk subsystem1*80GB ATA/100 7200 RPM Other hardware

35. 35  2004 Morgan Kaufmann Publishers Software Operating systemWindows XP Professional SP1 CompilerIntel C++ Compiler 7.1 (20030402Z) Microsoft Visual Studio.NET (7.0.9466) MicroQuill SmartHeap Library 6.01 File system typeNTFS System stateDefault

36. 36  2004 Morgan Kaufmann Publishers Figure4.4 Execution times of two programs on two different computers. Computer AComputer B Program 1 (seconds)110 Program 2 (seconds)1000100 Total time (seconds)1001110

37. 37  2004 Morgan Kaufmann Publishers 4.4 Real Stuff: Two SPEC Benchmarks and the Performance of Recent Intel Processors

38. 38  2004 Morgan Kaufmann Publishers Keywords System performance evaluation cooperative (SPEC) benchmark A set of standard CPU-intensive, integer and floating point benchmarks based on real programs.

39. 39  2004 Morgan Kaufmann Publishers Performance best determined by running a real application –Use programs typical of expected workload –Or, typical of expected class of applications e.g., compilers/editors, scientific applications, graphics, etc. Small benchmarks –nice for architects and designers –easy to standardize –can be abused SPEC (System Performance Evaluation Cooperative) –companies have agreed on a set of real program and inputs –valuable indicator of performance (and compiler technology) –can still be abused Benchmarks

40. 40  2004 Morgan Kaufmann Publishers Benchmark Games An embarrassed Intel Corp. acknowledged Friday that a bug in a software program known as a compiler had led the company to overstate the speed of its microprocessor chips on an industry benchmark by 10 percent. However, industry analysts said the coding error…was a sad commentary on a common industry practice of “cheating” on standardized performance tests…The error was pointed out to Intel two days ago by a competitor, Motorola …came in a test known as SPECint92…Intel acknowledged that it had “optimized” its compiler to improve its test scores. The company had also said that it did not like the practice but felt to compelled to make the optimizations because its competitors were doing the same thing…At the heart of Intel’s problem is the practice of “tuning” compiler programs to recognize certain computing problems in the test and then substituting special handwritten pieces of code… Saturday, January 6, 1996 New York Times

41. 41  2004 Morgan Kaufmann Publishers SPEC ‘89 Compiler “enhancements” and performance

42. 42  2004 Morgan Kaufmann Publishers SPEC CPU2000

43. 43  2004 Morgan Kaufmann Publishers SPEC 2000 Does doubling the clock rate double the performance? Can a machine with a slower clock rate have better performance?

44. 44  2004 Morgan Kaufmann Publishers Figure4.7 SPEC web9999 performance for a variety of Dell PowerEdge systems using the Xeon versions of the Pentium III and Pentium 4 microprocessors. SystemProcessor Number of disk drives Number of CPUs Number of networks Clock rate (GHz) Result 1550/1000Pentium III22212765 1650Pentium III3211.41810 2500Pentium III8241.133435 2550Pentium III1211.261454 2650Pentium 4 Xeon5243.065698 4600Pentium 4 Xeon10242.24615 6400/700Pentium III Xeon5440.74200 6600Pentium 4 Xeon XP84826700 8450/700Pentium III Xeon7880.78001

45. 45  2004 Morgan Kaufmann Publishers

46. 46  2004 Morgan Kaufmann Publishers Experiment Phone a major computer retailer and tell them you are having trouble deciding between two different computers, specifically you are confused about the processors strengths and weaknesses (e.g., Pentium 4 at 2Ghz vs. Celeron M at 1.4 Ghz ) What kind of response are you likely to get? What kind of response could you give a friend with the same question?

47. 47  2004 Morgan Kaufmann Publishers 4.5Fallacies and Pitfalls

48. 48  2004 Morgan Kaufmann Publishers Keywords Amdahl’s law A rule stating that the performance enhancement possible with a given improvement is limited by the amount that the improved feature is used. Million instruction per second (MIPS) A measurement of program execution speed based on the number of millions of instructions. MIPS is computed as the instruction count divided by the product of the execution time and.

49. 49  2004 Morgan Kaufmann Publishers Execution Time After Improvement = Execution Time Unaffected +( Execution Time Affected / Amount of Improvement ) Example: "Suppose a program runs in 100 seconds on a machine, with multiply responsible for 80 seconds of this time. How much do we have to improve the speed of multiplication if we want the program to run 4 times faster?" How about making it 5 times faster? Principle: Make the common case fast Amdahl's Law

50. 50  2004 Morgan Kaufmann Publishers Solution: Since we want the performance to be five times faster, the new execution time should be 20 seconds, giving

51. 51  2004 Morgan Kaufmann Publishers MIPS as a Performance Measure Consider the computer with three instruction classes and CPI measurements from the last example on page 252. Now suppose we measure the code for the same program from two different compilers and obtain the following data: Assume that the computer’s clock rate is 4 GHz. Which code sequence will execute faster according to MUPS? According to execution time? Code from Instruction counts (in billions) for each instruction class ABC Compiler 1511 Compiler 21011

52. 52  2004 Morgan Kaufmann Publishers Solution: Here we use three equations: Then =>

53. 53  2004 Morgan Kaufmann Publishers Now Let’s compute the MIPS rate for each version of the program: So, the code from compiler 2 has a higher MIPS rating, but the code from compiler 1 runs faster!

54. 54  2004 Morgan Kaufmann Publishers Suppose we enhance a machine making all floating-point instructions run five times faster. If the execution time of some benchmark before the floating-point enhancement is 10 seconds, what will the speedup be if half of the 10 seconds is spent executing floating-point instructions? We are looking for a benchmark to show off the new floating-point unit described above, and want the overall benchmark to show a speedup of 3. One benchmark we are considering runs for 100 seconds with the old floating-point hardware. How much of the execution time would floating- point instructions have to account for in this program in order to yield our desired speedup on this benchmark? Example

55. 55  2004 Morgan Kaufmann Publishers Performance is specific to a particular program/s –Total execution time is a consistent summary of performance For a given architecture performance increases come from: –increases in clock rate (without adverse CPI affects) –improvements in processor organization that lower CPI –compiler enhancements that lower CPI and/or instruction count –Algorithm/Language choices that affect instruction count Pitfall: expecting improvement in one aspect of a machine’s performance to affect the total performance Remember

56. 56  2004 Morgan Kaufmann Publishers 4.6Concluding Remarks

57. 57  2004 Morgan Kaufmann Publishers The execution time is related to other important measurements we can make by the following equation: