CS305 Computer Architecture Fall 2009 Lecture 04 Bhaskaran Raman - - PowerPoint PPT Presentation

CS305 Computer Architecture Fall 2009 Lecture 04

Bhaskaran Raman Department of CSE, IIT Bombay

http://www.cse.iitb.ac.in/~br/ http://www.cse.iitb.ac.in/synerg/doku.php?id=public:courses:cs305-fall09:start

Today's Topics

• Performance metrics, CPI
• Performance comparison
• Benchmarks

Performance Comparison

• What performance metric to use?
• User cares about response time
• Performance is inversely proportional
• What is execution time?
• Response time
• CPU time: User time + System time
• System performance vs. CPU performance
• Throughput vs. response-time
• We will focus on CPU performance

Which Program's Execution Time?

• Real “workload” is ideal
• Practical options:
• Real programs: compilers, office-suite, scientific...
• Kernels: key pieces of programs

– Example: Livermore loops

• Toy benchmarks: small programs

– Examples: Quick-sort, tower of Hanoi...

• Synthetic benchmarks: try to capture “average”

frequency of instructions in real programs

– Example: Whetstone, Dhrystone

More on Performance Comparisons...

• Caveat of benchmarks
• They are needed
• But manufacturers tend to optimize for benchmarks
• Need to be updated periodically
• Benchmark suite: collection of programs
• E.g. SPEC2000
• Reporting performance
• Reproducibility: program version, compiler, flags
• SPEC specifies compiler flags for baseline comparison

Some Numerics...

• Total (or average) execution time is a possible

metric

• Weighted execution time is better

Computer A Computer B Computer C Program P1 (secs) 1 10 20 Program P2 (secs) 1000 100 20 Total (secs) 1001 110 40

W i×T i

Normalizing the Performance

• Normalize such that all programs take the same

time, on some machine

• Arithmetic mean predicts performance
• Geometric mean?

Norm(A)Norm(A)Norm(A)Norm(B)Norm(B)Norm(B)Norm(C)Norm(C)Norm(C) A B C A B C A B C P1 1 10 20 0.1 1 2 0.05 0.5 1 P2 1 0.1 0.02 10 1 0.2 50 5 1 AM 1 5.05 10.01 5.05 1 1.1 25.03 2.75 1 GM 1 1 0.63 1 1 0.63 1.58 1.58 1

Summary

• Performance inversely proportional to execution-

time

• We are concerned with CPU time of unloaded

machine

• Weighted execution time with weights from real

workload is ideal

• Else, normalize w.r.t one machine

Amdahl's Law

• Amdahl's law:
• Diminishing returns
• Limit on overall speedup
• Corollary: make the

common case fast

1-F F 1-F F/Speedup

Amdahl's Law

• Amdahl's law:
• Diminishing returns
• Limit on overall speedup

1-F F 1-F F/Speedup

• Corollary: make the

common case fast Overall speedup= 1−FF 1−F F Speedup

Illustrating Amdahl's Law

• Example: implement faster memory, or faster ALU?
• Proposed memory speedup: 10x
• Proposed ALU speedup: 3x
• Depends on fraction of instructions

– Suppose F mem=0.2,F alu=0.5,F other=0.3

Speedup with faster memory= 1 0.80.2/10=1.22 Speedup with faster ALU= 1 0.50.5/3=1.5

Example continued...

• Fixing for what value of is

going for a faster memory better?

F alu=0.5

F mem

1 1−F memF mem/101.5 ⇒F mem10 27=0.36

The CPU Performance Equation

CPU time=Num.clock cycles×Clock cycletime CPU time=Num.of clock cycles÷Clock rate

OR

CPU time=IC×CPI×Cycletime

Putting these together Num.of clock cycles

=InstructionCount×Cycles Per Instruction

=IC×CPI

For a program,

More on the Equation

• This form is convenient
• Involves many relevant parameters
• Remembering is easy

CPU time= Seconds Program = Seconds Clock cycle× Clock cycles Instruction ×Instructions Program

• With CPI as the independent variable

CPI= CPU time Clock cycletime×IC

Other Convenient Forms of the Equation

• Number of clock cycles can be counted as:

CPU clock cycles=∑

i=1 n

CPI i×ICi Hence ,CPU time=∑

i=1 n

CPI i×ICi×Clock cycletime

• Calculating in terms of

CPI CPI i

CPI= CPU time Clock cycletime×IC=∑

i=1 n

CPI i× ICi IC 

Usefulness of the Equation

• easier to measure than
• Equivalently, is measured through
• Equation includes relevant parameters such as the

cycle time

IC i

F i F i

IC i

Measuring the Parameters for the Equation

• Clock cycle time:
• Easy for existing architectures
• Needs to be estimated in the design process
• Instruction Count:
• Requires a compiler
• And, simulator/interpreter, or instrumentation code
• CPI for each instruction type:
• Easy for simple architectures
• Pipelines, caches introduce complications
• Need to simulate and measure average CPI

A Design Example

• A design choice for conditional branch

instructions:

• Choice 1: condition code is set by a compare

instruction, checked by the next (branch) instruction

– 20% instructions are branches, and another 20% are

compares

– 2 cycles per branch, 1 cycle for all others – Clock-rate is 25% faster

• Choice 2: single instruction for compare and branch
• Which choice is better?

Solution for Design Example

CPU time1= IC1×[0.8×10.2×2] 1.25×C = IC1 C × 1.2 1.25 CPU time2= IC1×[0.6×10.2×2] C = IC1 C