Download presentation
Presentation is loading. Please wait.
Published byMoses Cannon Modified over 9 years ago
1
Advanced Computer Architecture Fundamental of Computer Design Instruction Set Principles and Examples Pipelining:Basic and Intermediate Concepts Memory Hierarchy Design Storage System Instruction-Level Parallelism:Concepts and Challenges Exploiting Instruction-Level Parallelism with Software Approaches Multiprocessors and Thread-Level Parallelism
2
Forces on Computer Architecture Computer Architecture Technology Programming Languages Operating Systems History Applications (A = F / M)
3
Fundamentals of Computer Design Introduction The Task of the Computer Designer Technology Trends Cost Price, and Their Trends Performance Quantitative Principles of Computer Design Putting It All Together: Performance and Price- Performance Power Consumption and Efficiency Fallacies and Pitfalls
4
Microprocessor Performance
5
Cost of Downtime
6
System Characteristics of the the Three Computing Classes
7
Technology Trends Clock Rate: ~30% per year Transistor Density: ~35% Chip Area: ~15% Transistors per chip: ~55% Total Performance Capability: ~100% by the time you graduate... –3x clock rate (3-4 GHz) –10x transistor count (1 Billion transistors) –30x raw capability plus 16x DRAM density, 32x disk density
8
The Most Important Functional Requirements and Architect Faces
9
1.4 Cost, Price, and Their Trends Prices of six generation of DRAMS
10
The Price of an Intel Pentium III over Time
11
What is “ Computer Architecture ” ? Coordination of many levels of abstraction Under a rapidly changing set of forces Design, Measurement, and Evaluation I/O systemInstr. Set Proc. Compiler Operating System Application Digital Design Circuit Design Instruction Set Architecture Firmware Datapath & Control Layout
12
Computer Architecture Topics Networks M Interconnection Network S PMPMPMP ° ° °° ° ° Topologies, Routing, Bandwidth, Latency, Reliability Processor-Memory-Switch Multiprocessors Networks and Interconnections Network Interfaces Shared Memory, Message Passing, Data Parallelism
13
Photograph of an Intel Pentium 4
14
This 8-inch Wafer Contains 564 MIPS64 20k Processors
16
Die yield
17
Estimated distribution of PC Costs
18
The components of price for a $1000 PC
19
1.5 Measuring and Reporting Performance: Execution Time
20
The programs in the SPEC CPU 2000 benchmark suites
21
The Embedded Benchmark EEMBC:The EDN Embedded Microprocessor Benchmarks Consortium
22
The machine, software, and baseline tuning parameters for the CINT2000
23
Comparing and Summarizing Performance
24
Weighted arithmetic mean execution for three machines
25
Execution times from Figure 1.15 normalized to each machine
26
1.6 Quantitative Principles of Computer Design Amdahl’s Law
27
Enhancement more, Improvement more
28
Amdahl’s Law (Page41)
29
Performance Comparison-Speedup Amdahl’s Law
30
The CPU Performance Equation(Page42)
31
CPU time Clock cycle time---Hardware technology and organization CPI---Organization and instruction set architecture Instruction count---Instruction set architecture and compiler technology
32
Overall CPI
33
Overall CPI Comparison (Page44)
34
CPI Com.
35
Speedup Pipeline(Operation manual,Regular design,…) Principle of locality-Temporal and Spatial Parallelism-Multiple Units, processors and Cluster Servers, Distributed Computing,… Clock Rate,(Circuits, Devices,…..) Optics,…..
36
1.7 Performance and Price-performance Seven different desktop systems
37
Performance and price-performance
39
Cluster Systems
40
The performance and the price-performance of cluster systems
41
Price-performance of cluster systems
42
Five different embedded processors
43
Relative performance of five different embedded processors for three of the five EEMBC benchmark suites EEMBC:The EDN Embedded Microprocessor Benchmarks Consortium
44
Relative price-performance of five different embedded processors for three of the five EEMBC benchmark suites
45
1.8 Power Consumption and Efficiency as the metric
46
1.9 Fallacies and Pitfalls Fallacies—misbelieves(F) Pitfalls---Easily made mistakes(P) –The relative performance of two processors with the same instruction set architecture(ISA) can be judged by clock rate or by the performance of a single benchmark suite. (F)(Fig.1.28) –Benchmarks remain valid indefinitely. (F)(Fig. 1.29) –Comparing hand-coded assembly and compiler- generated high-level language performance.(P) –Peak performance tracks observed performance. (F)
47
1.9 Fallacies and Pitfalls The Best design for a computer is the one that optimizes the primary objective without considering implementation.(F) Neglecting the cost of software in either evaluating a system or examining cost- performance. (P) Falling prey to Amdahl’s Law.(P) Synthetic benchmarks predict performance for real programs.
48
1.9 Fallacies and Pitfalls MIPS is an accurate measure for computing performance among computers.(F)
49
1.9 Fallacies and Pitfalls The problem with using MIPS as a measure for comparison –MIPS is dependent on the instruction set, making it difficult to compare MIPS of computer with different instruction sets. –MIPS varies between programs on the same computer. –Most importantly, MIPS can vary inversely to performance
50
P4 and P3 performance comparison-Relative performance
51
The tuning parameters for the SPEC CFP2000 report
52
The evolution of the SPEC benchmarks over time
53
The performance of three embedded processors
54
Measurements of peak performance and actual performance
55
1.10 Concluding Remarks Make the common case fast Chap. 2:The interaction between compiler and instruction set design. Part 3: Pipeline(Appendix A) Part 4: Memory Design(Chap.5) Part 5: Storage System (Chap. 7) (Page1-86),(page87-168),(page A-1~A- 87)…..
56
1.11 Historical Perspective and References The First General-purpose Electronic Computers Important special-purpose machines Commercial Developments Development of Quantitative Performance Measures:Successes and Failures
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.