Computer Architecture

A Quantitative Approach, Sixth Edition

Chapter 1
Fundamentals of Quantitative
Design and Analysis

Copyright © 2019, Elsevier Inc. All rights reserved. 1

 Introduction
Computer Technology
 Performance improvements:
 Improvements in semiconductor technology
 Feature size, clock speed
 Improvements in computer architectures
 Enabled by HLL compilers, UNIX
 Lead to RISC architectures

 Together have enabled:

 Lightweight computers
 Productivity-based managed/interpreted
programming languages

Copyright © 2019, Elsevier Inc. All rights reserved. 2

 Introduction
Single Processor Performance

Copyright © 2019, Elsevier Inc. All rights reserved. 3

 Introduction
Current Trends in Architecture
 Cannot continue to leverage Instruction-Level
parallelism (ILP)
 Single processor performance improvement ended in
2003

 New models for performance:

 Data-level parallelism (DLP)
 Thread-level parallelism (TLP)
 Request-level parallelism (RLP)

 These require explicit restructuring of the

application

Copyright © 2019, Elsevier Inc. All rights reserved. 4

 Classes of Computers
Classes of Computers
 Personal Mobile Device (PMD)
 e.g. start phones, tablet computers
 Emphasis on energy efficiency and real-time
 Desktop Computing
 Emphasis on price-performance
 Servers
 Emphasis on availability, scalability, throughput
 Clusters / Warehouse Scale Computers
 Used for “Software as a Service (SaaS)”
 Emphasis on availability and price-performance
 Sub-class: Supercomputers, emphasis: floating-point
performance and fast internal networks
 Internet of Things/Embedded Computers
 Emphasis: price

Copyright © 2019, Elsevier Inc. All rights reserved. 5

 Classes of Computers
Parallelism
 Classes of parallelism in applications:
 Data-Level Parallelism (DLP)
 Task-Level Parallelism (TLP)

 Classes of architectural parallelism:

 Instruction-Level Parallelism (ILP)
 Vector architectures/Graphic Processor Units (GPUs)
 Thread-Level Parallelism
 Request-Level Parallelism

Copyright © 2019, Elsevier Inc. All rights reserved. 6

 Classes of Computers
Flynn’s Taxonomy
 Single instruction stream, single data stream (SISD)

 Single instruction stream, multiple data streams (SIMD)

 Vector architectures
 Multimedia extensions
 Graphics processor units

 Multiple instruction streams, single data stream (MISD)

 No commercial implementation

 Multiple instruction streams, multiple data streams

(MIMD)
 Tightly-coupled MIMD
 Loosely-coupled MIMD

Copyright © 2019, Elsevier Inc. All rights reserved. 7

 Defining Computer Architecture
Defining Computer Architecture
 “Old” view of computer architecture:
 Instruction Set Architecture (ISA) design
 i.e. decisions regarding:
 registers, memory addressing, addressing modes,
instruction operands, available operations, control flow
instructions, instruction encoding

 “Real” computer architecture:

 Specific requirements of the target machine
 Design to maximize performance within constraints:
cost, power, and availability
 Includes ISA, microarchitecture, hardware

Copyright © 2019, Elsevier Inc. All rights reserved. 8

 Defining Computer Architecture
Instruction Set Architecture
 Class of ISA
 General-purpose registers
 Register-memory vs load-store
 RISC-V registers Register Name Use Saver
 32 g.p., 32 f.p. x9 s1 saved callee
Register Name Use Saver x10-x17 a0-a7 arguments caller
x0 zero constant 0 n/a x18-x27 s2-s11 saved callee
x1 ra return addr caller x28-x31 t3-t6 temporaries caller
x2 sp stack ptr callee f0-f7 ft0-ft7 FP temps caller
x3 gp gbl ptr f8-f9 fs0-fs1 FP saved callee
x4 tp thread ptr f10-f17 fa0-fa7 FP callee
x5-x7 t0-t2 temporaries caller arguments

x8 s0/fp saved/ callee f18-f27 fs2-fs21 FP saved callee

frame ptr f28-f31 ft8-ft11 FP temps caller

Copyright © 2019, Elsevier Inc. All rights reserved. 9

 Defining Computer Architecture
Instruction Set Architecture
 Memory addressing
 RISC-V: byte addressed, aligned accesses faster
 Addressing modes
 RISC-V: Register, immediate, displacement
(base+offset)
 Other examples: autoincrement, indexed, PC-relative
 Types and size of operands
 RISC-V: 8-bit, 32-bit, 64-bit

Copyright © 2019, Elsevier Inc. All rights reserved. 10

 Defining Computer Architecture
Instruction Set Architecture
 Operations
 RISC-V: data transfer, arithmetic, logical, control,
floating point
 See Fig. 1.5 in text
 Control flow instructions
 Use content of registers (RISC-V) vs. status bits (x86,
ARMv7, ARMv8)
 Return address in register (RISC-V, ARMv7, ARMv8)
vs. on stack (x86)
 Encoding
 Fixed (RISC-V, ARMv7/v8 except compact instruction
set) vs. variable length (x86)
Copyright © 2019, Elsevier Inc. All rights reserved. 11
 Trends in Technology
Trends in Technology
 Integrated circuit technology (Moore’s Law)
 Transistor density: 35%/year
 Die size: 10-20%/year
 Integration overall: 40-55%/year

 DRAM capacity: 25-40%/year (slowing)

 8 Gb (2014), 16 Gb (2019), possibly no 32 Gb

 Flash capacity: 50-60%/year

 8-10X cheaper/bit than DRAM

 Magnetic disk capacity: recently slowed to 5%/year

 Density increases may no longer be possible, maybe increase from 7 to
9 platters
 8-10X cheaper/bit then Flash
 200-300X cheaper/bit than DRAM

Copyright © 2019, Elsevier Inc. All rights reserved. 12

 Trends in Technology
Bandwidth and Latency
 Bandwidth or throughput
 Total work done in a given time
 32,000-40,000X improvement for processors
 300-1200X improvement for memory and disks

 Latency or response time

 Time between start and completion of an event
 50-90X improvement for processors
 6-8X improvement for memory and disks

Copyright © 2019, Elsevier Inc. All rights reserved. 13

 Trends in Technology
Bandwidth and Latency

Log-log plot of bandwidth and latency milestones

Copyright © 2019, Elsevier Inc. All rights reserved. 14

 Trends in Technology
Transistors and Wires
 Feature size
 Minimum size of transistor or wire in x or y
dimension
 10 microns in 1971 to .011 microns in 2017
 Transistor performance scales linearly
 Wire delay does not improve with feature size!
 Integration density scales quadratically

Copyright © 2019, Elsevier Inc. All rights reserved. 15

 Trends in Power and Energy
Power and Energy
 Problem: Get power in, get power out

 Thermal Design Power (TDP)

 Characterizes sustained power consumption
 Used as target for power supply and cooling system
 Lower than peak power (1.5X higher), higher than
average power consumption

 Clock rate can be reduced dynamically to limit

power consumption

 Energy per task is often a better measurement

Copyright © 2019, Elsevier Inc. All rights reserved. 16
 Trends in Power and Energy
Dynamic Energy and Power
 Dynamic energy
 Transistor switch from 0 -> 1 or 1 -> 0
 ½ x Capacitive load x Voltage2

 Dynamic power
 ½ x Capacitive load x Voltage2 x Frequency switched

 Reducing clock rate reduces power, not energy

Copyright © 2019, Elsevier Inc. All rights reserved. 17

 Trends in Power and Energy
Power
 Intel 80386
consumed ~ 2 W
 3.3 GHz Intel
Core i7 consumes
130 W
 Heat must be
dissipated from
1.5 x 1.5 cm chip
 This is the limit of
what can be
cooled by air

Copyright © 2019, Elsevier Inc. All rights reserved. 18

 Trends in Power and Energy
Reducing Power
 Techniques for reducing power:
 Do nothing well
 Dynamic Voltage-Frequency Scaling

 Low power state for DRAM, disks

 Overclocking, turning off cores
Copyright © 2019, Elsevier Inc. All rights reserved. 19
 Trends in Power and Energy
Static Power
 Static power consumption
 25-50% of total power
 Currentstatic x Voltage
 Scales with number of transistors
 To reduce: power gating

Copyright © 2019, Elsevier Inc. All rights reserved. 20

 Trends in Cost
Trends in Cost
 Cost driven down by learning curve
 Yield

 DRAM: price closely tracks cost

 Microprocessors: price depends on

volume
 10% less for each doubling of volume

Copyright © 2019, Elsevier Inc. All rights reserved. 21

 Trends in Cost
Integrated Circuit Cost
 Integrated circuit

 Bose-Einstein formula:

 Defects per unit area = 0.016-0.057 defects per square cm (2010)

 N = process-complexity factor = 11.5-15.5 (40 nm, 2010)
Copyright © 2019, Elsevier Inc. All rights reserved. 22
 Dependability
Dependability
 Module reliability
 Mean time to failure (MTTF)
 Mean time to repair (MTTR)
 Mean time between failures (MTBF) = MTTF + MTTR
 Availability = MTTF / MTBF

Copyright © 2019, Elsevier Inc. All rights reserved. 23

 Measuring Performance
Measuring Performance
 Typical performance metrics:
 Response time
 Throughput

 Speedup of X relative to Y
 Execution timeY / Execution timeX

 Execution time
 Wall clock time: includes all system overheads
 CPU time: only computation time

 Benchmarks
 Kernels (e.g. matrix multiply)
 Toy programs (e.g. sorting)
 Synthetic benchmarks (e.g. Dhrystone)
 Benchmark suites (e.g. SPEC06fp, TPC-C)

Copyright © 2019, Elsevier Inc. All rights reserved. 24

 Principles
Principles of Computer Design
 Take Advantage of Parallelism
 e.g. multiple processors, disks, memory banks,
pipelining, multiple functional units

 Principle of Locality
 Reuse of data and instructions

 Focus on the Common Case

 Amdahl’s Law

Copyright © 2019, Elsevier Inc. All rights reserved. 25

 Principles
Principles of Computer Design
 The Processor Performance Equation

Copyright © 2019, Elsevier Inc. All rights reserved. 26

 Principles
Principles of Computer Design
 Different instruction types having different
CPIs

Copyright © 2019, Elsevier Inc. All rights reserved. 27

 Principles
Principles of Computer Design
 Different instruction types having different
CPIs

Copyright © 2019, Elsevier Inc. All rights reserved. 28

Fallacies and Pitfalls
 All exponential laws must come to an end
 Dennard scaling (constant power density)
 Stopped by threshold voltage
 Disk capacity
 30-100% per year to 5% per year
 Moore’s Law
 Most visible with DRAM capacity
 ITRS disbanded
 Only four foundries left producing state-of-the-art
logic chips
 11 nm, 3 nm might be the limit
Copyright © 2019, Elsevier Inc. All rights reserved. 29
Fallacies and Pitfalls
 Microprocessors are a silver bullet
 Performance is now a programmer’s burden
 Falling prey to Amdahl’s Law
 A single point of failure
 Hardware enhancements that increase
performance also improve energy
efficiency, or are at worst energy neutral
 Benchmarks remain valid indefinitely
 Compiler optimizations target benchmarks

Copyright © 2019, Elsevier Inc. All rights reserved. 30

Fallacies and Pitfalls
 The rated mean time to failure of disks is
1,200,000 hours or almost 140 years, so
disks practically never fail
 MTTF value from manufacturers assume
regular replacement
 Peak performance tracks observed
performance
 Fault detection can lower availability
 Not all operations are needed for correct
execution
Copyright © 2019, Elsevier Inc. All rights reserved. 31