Transcript Lec0-intro - ECE Users Pages - Georgia Institute of Technology

ECE 4100/6100
Advanced Computer Architecture
Lecture 0 Introduction
Prof. Hsien-Hsin Sean Lee
School of Electrical and Computer Engineering
Georgia Institute of Technology
Course Information
• Instructor: Hsien-Hsin Sean Lee ([email protected])
• Office: TBD
• Online resources:
– www.ece.gatech.edu/~leehs/ECE6100
– Constantly updated, check it out regularly
– www.tsquare.gatech.edu
• Prerequisite: ECE3055 Computer Organization and
Operating System or the equivalent.
Pre-requisite
• Textbook
– Hennessy and Patterson, Computer Architecture: A Quantitative
Approach (4th edition), Morgan Kaufmann, 2006.
– (Recommended) Jean-Loup Baer, Microprocessor Architecture -- From Simple Pipelines to Chip Multiprocessors. Cambridge
Univ. Press, 2009.
• Other teaching materials
– Key papers available later in class meetings and course web
– Slides & Lectures
Textbook
2
Course Scope  To Learn
•
•
•
•
•
•
•
•
•
•
•
Core concept of modern microprocessor architecture
ISA, performance
Instruction-Level parallelism
Branch prediction and Front-end fetch
Dynamic HW Scheduling Techniques
Memory Hierarchy
Multiprocessors, SMT, Multi-core, Many-core
Cache Coherence and Memory Consistency Models
Case studies of Commercial Microprocessors
VLIW, EPIC, Static Scheduling
Physical design, emerging trend, technology integration (if
time allowed)
3
This course
requires a lot of time commitment
4
BY THE WAY
This is not really an “Advanced”
Computer Architecture class per se
5
Grading
• 4 Programming Assignments: 60%
– Individual work, no collaboration
– TA’s check-off required
– Could cause panicking, heartburn, dizziness, etc.
• Exams
– One in-class exams: 15%
– One final exam: 25%
6
Stack of A Computing Problem
Problems
Apps Trend
Algorithms
Programming Languages
Compilers
ISA
System Architecture
Implementation
MicroArchitecture
Architects’
Territory
Logic and Circuits
Transistors
Manufacturing
Technology
Trend
7
Moore’s Law (a.k.a. Intel’s Roadmap)
90 nm
596 mm2
65 nm
698 mm2
1.7 billions
Montecito
2 billions
Tukwila
3 billions
Nvidia Fermi
@40nm
10 μm
13.5mm2
42millions
Exponential growth
3.9 billions
Altera Stratix
FPGA @28nm
2,250
Source: Intel Corp.
8
Feature Size
Feature size shrinks by 70% per 18 to 24 months
9
Feature Size (More Recent)
10
Transistor Cost
11
Power Density Trend
P  CVdd f  Vdd I st  Vdd Ileak
2
Source: Intel Corp.
12
Google Server Farms (Oregon)
13
Heat Dissipation
Pure copper
Cooler
jet
Cooligy’s channel
3D Cooler Pro
Source: K. Skadron
Cooking-Aware (or Colwell’s Charcoal-aware) Computing
PS3 Grill
(http://www.therealps2grill.com
14
Watch this
Click the chip
15
Focus on Computer Architecture
software
instruction set
hardware
Technology
Programming
Languages
Applications
Computer
Architecture
Operating
Systems
Virtualization
History (A = F / M)
16
Instruction Sets
High
Performance
Servers
X86, Itanium,
Sparc, Alpha
x86
Desktop,
Notebook
PC
x86
PPC
Embedded
Processors
Market Segments
Atom
ARM
MIPS
SH
Present
17
Constantly Changing Definition
• 50s to 60s: Computer Architecture ~ Computer Arithmetic
• 70s to mid 80s: Instruction Set Design, especially ISA
appropriate for compilers
• 90s: Speculation: Predict this, predict that; memory system;
I/O system; Multiprocessors; Networks
• 2000s: Power efficiency , Communication, On-die
Interconnection Network, Multi-this, Multi-that. (We are here)
• 2015 and beyond: Thousand-core processors, Self adapting
systems? Self organizing structures? DNA Systems/Quantum
Computing?
18
Job Description of a Computer Architect
• Used to be “Performance, performance, performance”
• Make trade-off of performance, complexity effectiveness,
power, technology, cost, etc.
New Fads
• Availability
– Where you store your photos, emails and shared docs today?
– Cloud computing
• Reliability
– Toyota blamed soft errors for the sudden acceleration problem
• Security
– Intel acquired McAfee
• Power management
– It is about money !
19
Job Description of a Computer Architect
Understand application requirements
–
–
–
–
General purpose Desktop (Intel, AMD)
Game and multimedia (STI’s Cell, Nvidia, Wii, Xbox 360)
Embedded and real-time (ARM, Atom, MIPS)
Online transactional processing (OLTP), data warehouse
servers (Sun Fire T2000 (UltraSparc T1), IBM POWER
(p690), Google Cluster)
– Scientific (finite element analysis, protein folding, weather
forecast, defense related (DE Shaw Anton, IBM BlueGene,
Cray T3D/T3E, IBM SP2)
– Sometimes, there is no boundary …
20