Overview Computer architecture Scaling performance and CMOS 1 - - PowerPoint PPT Presentation

1 • Trends in Microprocessor Architecture

R05 Chip Multiprocessors (ACS MPhil) Robert Mullins

Chip Multiprocessors (ACS MPhil) 2

Overview

• Computer architecture
• Scaling performance and CMOS

– Where have performance gains come from? – Modern superscalar processors – The limits of superscalar processors

• Going parallel
• This course

Chip Multiprocessors (ACS MPhil) 3

Computer architecture

“Computer architecture is the interface between what technology can provide and what the marketplace demands” “Computer architecture is a science of trade-offs” Yale Patt

Chip Multiprocessors (ACS MPhil) 4

Computer architecture

“Computer architecture is the science and art of selecting and interconnecting hardware components to create computers that meet functional, performance and cost goals” Mark Hill “Computer architecture forms the bridge between application need and the capabilities of the underlying technology” Tilak Agerwala and Siddhartha Chatterjee

Chip Multiprocessors (ACS MPhil) 5

Computer architecture

• We cannot architect a new computer without defining

performance, power and cost goals. The design process is all about understanding and making trade-

• ffs
• What is our target market and what applications will

we be running?

• The “best” architecture is a moving target

– The needs of the marketplace change – Shifting fabrication technology characteristics – New technologies

• memory, packaging, compiler, languages, ...

Chip Multiprocessors (ACS MPhil) 6

Computer architecture

“Computer architect's often err by preparing for yesterday's computations” Bill Dally (Easy to make the same error during a PhD!) Tomorrow's applications and technologies are not easy to predict!

Chip Multiprocessors (ACS MPhil) 7

Historic performance gains

Reproduced from “Computer architecture: A quantitative approach”, Hennessy/Patterson

Chip Multiprocessors (ACS MPhil) 8

Historic performance gains

Burger's “end of the road paper” suggested performance would be limited to 12.5%/annum Predicted: 1997-2014 7.4x Actual: ~36x If at historical rate: 1720x

Chip Multiprocessors (ACS MPhil) 9

Microprocessor trends

https://github.com/karlrupp/microprocessor-trend-data Chip Multiprocessors (ACS MPhil) 10

Historic performance gains

• Microprocessor performance increased at a

rate of ~52%/year between 1986-2002

– ~800X improvement over 16 years – How was such an improvement in performance achieved? – Is this a reasonable rate of performance growth given the advances in fabrication technology?

• Exe. time = Instr. count x CPI x Clock Period

Chip Multiprocessors (ACS MPhil) 11

Historic performance gains

• Technology scaling

– 7 process generations – Scaling provides ~1.4x transistor performance improvment per generation – 10.5X – (careful, this doesn't automatically translate directly into performance gains)

Reproduced with kind permission

• f Mark Horowitz

Chip Multiprocessors (ACS MPhil) 12

Historic performance gains

• Gates per clock

– Less logic between pipeline registers – Reduction from ~100 to 10 gate delays – 10X

• How?

– Pipelining

• 5 to 20 stages (~4X)

– Circuit-level advances

• e.g. new logic families
• ~2.5X

Reproduced with kind permission

• f Mark Horowitz

Chip Multiprocessors (ACS MPhil) 13

Historic performance gains

~105X

Reproduced from “CMOS VLSI Design” Weste/Harris (2005)

Chip Multiprocessors (ACS MPhil) 14

Historic performance gains

• IPC & instr. count

– ~5-8X improvement in SPECint/MHz – This is despite clock frequency improvements – Includes advances in compiler technology and impact of increased bus widths

Improvement in SPECint95/Mhz over time Reproduced with kind permission

• f Mark Horowitz

Chip Multiprocessors (ACS MPhil) 15

Historic performance gains

• How was it possible to maintain and even

decrease CPI (improve IPC)

– Moore's law! – How were the additional transistors exploited?

• Intel 386 to Pentium 4

– 386: 275K transistors (die size = 43mm2) – P4: 42M transistors (die size = 217mm2)

• 5X from increased die size
• 27X from technology scaling
• Today's (2017) largest chips contain > 10

billion transistors

Chip Multiprocessors (ACS MPhil) 16

Historic performance gains

Reproduced from CMOS VLSI Design, Weste and Harris (2005)

Chip Multiprocessors (ACS MPhil)

Moore’s Law

Chip Multiprocessors (ACS MPhil) 18

The future of Moore's Law: 2D to 3D

• Beyond 2021 it won't be

economically desirable to shrink transistor dimensions

• Recently introduced

vertical transistors (e.g. dual-gate and tri-gate)

• Monolithic 3D predicted by

2024

• Roadmap to consider

applications in future (more of an end-to-end view vs. bottom-up)

The latest ITRS Roadmap (2015) predicts that physical gate length will not shrink beyond 2021. Earlier predictions (2013) were more

• ptimistic.

Chip Multiprocessors (ACS MPhil) 19

Modern superscalar processors

• Revision (See Hennessy/Patterson)

– Significant hardware support for Instruction Level Parallelism (ILP) in most commercial microprocessors

• Multiple-issue architectures
• Deep pipelines, branch prediction, speculative execution
• Large on-chip caches (L1/L2/L3)
• Out-of-order execution, register renaming
• Dynamic memory address disambiguation
• SIMD instructions
• ...

Chip Multiprocessors (ACS MPhil) 20

Modern superscalar processors

Chip Multiprocessors (ACS MPhil) 21

Limits of superscalar processors

• Cost and complexity of extracting ILP

– Diminishing returns – Increased complexity limits ability to optimise design

• The underlying fabrication technology characteristics

are becoming more challenging too

– Increases verification complexity and time – Increases time-to-market

Chip Multiprocessors (ACS MPhil) 22

Limits of superscalar processors

• Pipeline depth limits

– Interruptions to the pipeline (branches) – Performance of the memory system – Clocking overheads (registers/clock skew) – Need to balance stages and maintain the atomicity

• f some operations

– Limited ILP – Power cost

(See also “Optimal Pipeline Depth” link on Seminar 1 wiki page)

Chip Multiprocessors (ACS MPhil) 23

Limits of superscalar processors

"Coming challenges in microarchitecture and architecture", Ronen et al, 2001

Chip Multiprocessors (ACS MPhil) 24

Limits of superscalar processors

• Interconnect versus transistor scaling

– Smaller transistors = faster/lower power – Wires don't scale in the same way ☹ – Centralised structures don't scale well – Pressure to decentralise – Consider bypass network between FUs

• Clustered implementations

Chip Multiprocessors (ACS MPhil) 25

Limits of superscalar processors

• Voltage scaling and power limits

– Voltage scaling has slowed

• 5V to 1V - gave us 25X power savings
• 1V to 0.7V (limit at end of CMOS around 2020)
• Only 2X power savings left from voltage scaling!

– Sensible power limits already reached – Pressure to reduce power consumption

• Process variation complications

– Fault tolerance requirements in the longer term

Chip Multiprocessors (ACS MPhil) 26

Going parallel

• Accept we can make little progress with

single-thread performance

• Look towards thread-level parallelism

– Achieve our performance gains in a new way: – Rapidly increase the number of cores

• 2X-3X per generation

– Don't scale the clock frequency

• Create simpler more power efficient cores instead

Chip Multiprocessors (ACS MPhil) 27

Going parallel

Pawlowski (Intel) 2007 It is now 2018..... Numbers of cores has scaled less agressively than this. In 2017 @ 14nm, High-end server part: 28 Core, Xeon (Skylake) 56 threads Clock frequency 2.5GHz (max turbo freq. 3.8GHz) TDP (power) = 205 W

Chip Multiprocessors (ACS MPhil) 28

Going parallel

• Going parallel is simple?

– Replicate existing processor designs to ease design process – Many applications already exist where thread-level parallelism is plentiful – We've had 30+ years of experience writing parallel programs

Chip Multiprocessors (ACS MPhil) 29

Going parallel

• Many new challenges:

– On-chip and off-chip communication – Simpler cores and Amdahl's law – Power constrained design – Support for the shared-memory paradigm? – Synchronization and thread-scheduling support? – Everyone must now write scalable and correct parallel programs!

Chip Multiprocessors (ACS MPhil) 30

Going parallel

• Power is a first order design constraint

– Power consumption is already at a sensible limit (for many applications we would like to reduce it) – We are going to increase the number of cores by 2-3X per generation

• Power savings?

– Core shrink (<1.4X) – Simpler cores (1.4-2X?) – Some VDD savings – Need to add “uncore” logic too! – Techniques for adaptive EPI?

Chip Multiprocessors (ACS MPhil) 31

Going parallel

• Beyond homogenous multicore

– Power consumption is a limiting factor in the design of multicore processors – For many designs this has prompted the integration of many specialized accelerators

• An ASIC implementation of an algorithm may be 10-

1000X more energy efficient that a software implementation

• e.g. Apple A8 SoC:

– ~50% custom accelerators – ~25% CPUs (2) – ~25% GPU

Chip Multiprocessors (ACS MPhil)

Future of multicore?

• “NAVIGO”, [Hempstead,

Wei and Brooks, 2011]

• Examined throughput
• rientated workload
• Suggest gains limited to

35% per year due to power constraints

Chip Multiprocessors (ACS MPhil)

Future gains

Need for applications to be approximation/fault tolerant

Node: “2nm/1.5nm” Vertical Gate-All-Around-Device (GAA) Monolithic-3D (stacking of devices) VDD = 0.4V

IRDS Roadmap (2016) Chip Multiprocessors (ACS MPhil) 34

This course

• Introduction to the challenges of building and

programming chip multiprocessors

– Lots to learn from traditional parallel computers, but many problems and trade-offs are new

• New applications
• The trade-offs on-chip are very different to those when

designing physically larger parallel machines

• Power and energy constraints
• Parallel programming for the masses

Chip Multiprocessors (ACS MPhil) 35

This course

• 1. Trends in microprocessor architecture
• 2. Introduction to parallel computing
• 3. Parallel algorithms
• 4. Chip Multiprocessors (I)
• 5. Chip Multiprocessors (II)
• 6. Transactional memory
• 7. On-chip interconnection networks
• 8. Manycore research issues
• 2017: Rune Holm ARM (Machine Learning Group)
• 2016: Gavin Stark, Netronome (CTO)
• 2014: David Moloney, Movidius (CTO)
• 2012: Matt Horsnell, ARM
• 2011: Eben Upton, Broadcom
• ...