A New Golden Age for
Computer Architecture:
Domain-Specific Hardware/Software Co-Design,
Enhanced Security, Open Instruction Sets,
and Agile Chip Development
John Hennessy and David Patterson
Stanford and UC Berkeley
13 June 2018
https://www.youtube.com/watch?v=3LVeEjsn8Ts
1
Outline
Part II: Current
Architecture Challenges -
Ending of Dennard Scaling
and Moore’s Law, Security
Part I: History of
Architecture -
Mainframes,
Minicomputers,
Microprocessors,
RISC vs CISC, VLIW
Part III: Future Architecture Opportunities -
Domain Specific Languages and Architecture,
Open Architectures, Agile Hardware Development
2
IBM Compatibility Problem in Early 1960s
7094
7074
7080
7010
By early 1960’s, IBM had 4 incompatible lines of computers!
➡
701
➡
650
➡
702
1401 ➡
Each system had its own:
Instruction set architecture (ISA)
▪
I/O system and Secondary Storage:
▪
magnetic tapes, drums and disks
▪ Assemblers, compilers, libraries,...
▪ Market niche: business, scientific, real time, ...
IBM System/360 – one ISA to rule them all
3
Control versus Datapath
▪ Processor designs split between datapath, where numbers are stored and
arithmetic operations computed, and control, which sequences operations on
datapath
▪ Biggest challenge for computer designers was getting control correct
Control
h
t
a
p
a
t
a
D
C
P
.
g
e
R
.
t
s
n
I
s
r
e
t
s
i
g
e
R
Instruction
Condition?
Control Lines
▪ Maurice Wilkes invented the
idea of microprogramming to
design the control unit of a
processor*
▪ Logic expensive vs. ROM or RAM
▪ ROM cheaper than RAM
▪ ROM much faster than RAM
* "Micro-programming and the design of the control circuits in an electronic digital computer,"
M. Wilkes, and J. Stringer. Mathematical Proc. of the Cambridge Philosophical Society, Vol. 49, 1953.
Main Memory
U
L
A
Data
Busy?
Address
4
Microprogramming in IBM 360
Model
Datapath width
Microcode size
Clock cycle time (ROM)
Main memory cycle time
Price (1964 $)
Price (2018 $)
M30
8 bits
4k x 50
750 ns
1500 ns
$192,000
$1,560,000
M40
16 bits
4k x 52
625 ns
2500 ns
$216,000
$1,760,000
M50
32 bits
2.75k x 85
500 ns
2000 ns
$460,000
$3,720,000
M65
64 bits
2.75k x 87
200 ns
750 ns
$1,080,000
$8,720,000
Fred Brooks, Jr.
5
IC Technology, Microcode, and CISC
▪ Logic, RAM, ROM all implemented using same transistors
▪ Semiconductor RAM ≈ same speed as ROM
▪ With Moore’s Law, memory for control store could grow
▪ Since RAM, easier to fix microcode bugs
▪Allowed more complicated ISAs (CISC)
▪ Minicomputer (TTL server) example:
-Digital Equipment Corp. (DEC)
-VAX ISA in 1977
▪ 5K x 96b microcode
6
Writable Control Store
▪ If Control Store is RAM, then could tailor “firmware”
to application: “Writable Control Store”
▪ Microprogramming became popular in academia
- Patterson PhD thesis*
- SIGMICRO was for microprogramming**
▪ Xerox Alto (Bit Slice TTL) in 1973
-1st computer with Graphical User Interface & Ethernet
-BitBlt and Ethernet controller in microcode
* Verification of microprograms, David Patterson, UCLA, 1976
** “The design of a system for the synthesis of correct microprograms,”
David Patterson, Proc. 8th Annual Workshop of Microprogramming, 1975
Chuck Thacker
7
Microprocessor Evolution
▪ Rapid progress in 1970s, fueled by advances in MOS technology,
▪
▪
▪
imitated minicomputers and mainframe ISAs
“Microprocessor Wars”: compete by adding instructions (easy for microcode),
justified given assembly language programming
Intel iAPX 432: Most ambitious 1970s micro, started in 1975
▪ 32-bit capability-based object-oriented architecture, custom OS written in Ada
▪ Severe performance, complexity (multiple chips), and usability problems; announced 1981
Intel 8086 (1978, 8MHz, 29,000 transistors)
“Stopgap” 16-bit processor, 52 weeks to new chip
▪
ISA architected in 3 weeks (10 person weeks) assembly-compatible with 8 bit 8080
▪
IBM PC 1981 picks Intel 8088 for 8-bit bus (and Motorola 68000 was late)
▪
▪ Estimated PC sales: 250,000
▪ Actual PC sales: 100,000,000 ⇒ 8086 “overnight” success
▪ Binary compatibility of PC software ⇒ bright future for 8086
8