1

Overview

1. Motivation (Kevin)2. Thermal issues (Kevin)3. Power modeling (David)4. Thermal management (David)5. Optimal DTM (Lev)6. Clustering (Antonio)7. Power distribution (David)8. What current chips do (Lev)9. HotSpot (Kevin)

2

Power modeling

• Research Power Simulators– Wattch – Brooks and Martonosi ISCA2000– SimplePower – Vijaykrishnan et al (Penn

State) ISCA2000– TEMPEST – Dhodapkar et al (Intel/Wisconsin)– PowerAnalyzer – Umich/Colorado– AccuPower – SUNY Binghamton

• Industry Power Simulators– IBM PowerTimer – Brooks and Bose

PACS2000– Intel ALPS – Gunther, et al.

3

Power: The Basics

• Dynamic power vs. Static power– Dynamic: “switching” power– Static: “leakage” power– Dynamic power dominates, but static power

increasing in importance– Trends in each

• Static power: steady, per-cycle energy cost• Dynamic power: capacitive and short-circuit• Capacitive power: charging/discharging at

transitions from 01 and 10• Short-circuit power: power due to brief short-

circuit current during transitions.• Mostly focus on capacitive, but recent work on

others

4

Capacitive Power dissipation

Power ~ ½ CV2Af

Capacitance:Function of wire length, transistor size

Supply Voltage:Has been dropping with successive fab generations

Clock frequency:Increasing…Activity factor:

How often, on average, do wires switch?

Vin Vout

CL

Vdd

5

Short-Circuit Power Dissipation

• Short-Circuit Current caused by finite-slope input signals

• Direct Current Path between VDD and GND when both NMOS and PMOS transistors are conducting

VOUT

CL

ISC

VIN

6

Leakage Power

• Subthreshold currents grow exponentially with increases in temperature, decreases in threshold voltage

Tka

Vq

DSub a

T

ekI

VOUT

CL ISub

VIN

IGate

7

Modeling Hierarchy and Tool Flow

set of workloads

Early analytical performance modelsTrace/exec-driven, cycle-accurate simulation models

Microarchparms/specs

RTL MODEL (VHDL) RTLsim

Circuit-level (hierarchical) netlist model

gate-level model (if synthesized)

Layout-level physical design model

microarchlevel

RTLlevel

gate-level

ckt-level

layout-level

Energy Models

cktextract

Capextract,sim

Sim Test Cases

Bitvectortest cases

(Architectural)

edit/debug

edit/tune/debug

Design rules

design rulecheck,validate

PerformanceTest Cases

edit/debug

refine,update

sim,

8

Analysis Abstraction Levels

Abstraction Analysis Analysis Analysis Analysis Energy

Level Capacity Accuracy Speed Resources Savings

Most Worst Fastest Least Most

Application

Behavioral

Architectural (RTL)

Logic (Gate)

Transistor (Circuit)

Least Best Slowest Most Least

9

Power/Performance abstractions

• Low-level:– Hspice– PowerMill

• Medium-Level: – RTL Models

• Architecture-level:– PennState SimplePower– Intel Tempest– Princeton Wattch– IBM PowerTimer– Umich/Colorado PowerAnalyzer

10

Low-level models: Hspice

• Extracted netlists from circuit/layout descriptions– Diffusion, gate, and wiring capacitance is modeled

• Analog simulation performed– Detailed device models used– Large systems of equations are solved– Can estimate dynamic and leakage power dissipation

within a few percent– Slow, only practical for 10-100K transistors

• PowerMill (Synopsys) is similar but about 10x faster

11

Medium-level models: RTL

• Logic simulation obtains switching events for every signal

• Structural VHDL or verilog with zero or unit-delay timing models

• Capacitance estimates performed– Device Capacitance

• Gate sizing estimates performed, similar to synthesis

– Wiring Capacitance• Wire load estimates performed, similar to

placement and routing

• Switching event and capacitance estimates provide dynamic power estimates

12

Architecture level models

• Bottom-up Approach: – Estimate “CV2f” via analytical models– Tools: Wattch, PowerAnalyzer, Tempest (mixed-

mode)

• Top-Down Approach– Estimate “CV2f” via empirical measurements– Tools: PowerTimer, AccuPower, Most Industrial Tools

• Estimate “A” via statistics from architectural-performance simulators

Power ~ ½ CV2Af

13

Analytical Models: Capacitance

• Requires modeling wire length and estimating transistor sizes

• Related to RC Delay analysis for speed along critical path– But capacitance estimates require summing

up all wire lengths, rather than only an accurate estimate of the longest one.

14

Register File: Capacitance Analysis

Pre-ChargeD

ecod

ers

Bitlines

(Data Width of Entries)

Wordlines

(Number of Entries)

Sense Amps

Cell

Bit

metal

gatecapNrdlineDrivediffcapWorwordline

CngthWordlinele

CinesNumberBitlCC

*

* 1

metal

diffcapNgdiffcapPchbitline

CgthBitlinelen

ClinesNumberWordCC

*

* 1

Cell Access Transistors (N1)

Number of Ports

Number of Ports

Bit

15

Register File Model: Accuracy

• Validated against a register file schematic used in internal Intel design

• Compared capacitance values with estimates from a layout-level Intel tool

• Interconnect capacitance had largest errors– Model neglects poly connections– Differences in wire lengths -- difficult to tell wire

distances of schematic nodes

Error Rates Gate Diff InterConn. Total Wordline(r) 1.11 0.79 15.06 8.02 Wordline(w) -6.37 0.79 -10.68 -7.99 Bitline(r) 2.82 -10.58 -19.59 -10.91 Bitline(w) -10.96 -10.60 7.98 -5.96 (Numbers in Percent)

16

Different Circuit Design Styles

• RTL and Architectural level power estimation requires the tool/user to perform circuit design style assumptions– Static vs. Dynamic logic– Single vs. Double-ended bitlines in register

files/caches– Sense Amp designs– Transistor and buffer sizings

• Generic solutions are difficult because many styles are popular

• Within individual companies, circuit design styles may be fixed

17

Clock Gating: What, why, when?

• Dynamic Power is dissipated on clock transitions

• Gating off clock lines when they are unneeded reduces activity factor

• But putting extra gate delays into clock lines increases clock skew

• End results:– Clock gating complicates design analysis but saves

power.

Clock

Gate

Gated Clock

18

Wattch: An Overview

Overview of Features• Parameterized models for different CPU units

– Can vary size or design style as needed• Abstract signal transition models for speed

– Can select different conditional clocking and input transition models as needed

• Based on SimpleScalar (has been ported to many simulators)

• Modular: Can add new models for new units studied

Wattch’s Design Goals• Flexibility• Planning-stage info• Speed• Modularity• Reasonable accuracy

19

Unit Modeling

Modeling Capacitance• Models depend on structure,

bitwidth, design style, etc.• E.g., may model capacitance of a

register file with bitwidth & number of ports as input parameters

Modeling Activity Factor• Use cycle-level simulator to

determine number and type of accesses– reads, writes, how many

ports• Abstract model of bitline activity

ParameterizedRegister File

PowerModel

PowerEstimate

Number of entries

Data width of entries

# Read Ports

# Write Ports

Bitline Activity

Number of Active Ports

20

Fetch Dispatch Issue/Execute Writeback/ Commit

Power (Units Accessed)

I-cache Bpred

Rename Table

Inst. Window Reg. File

Inst. Window Reg File ALU D-Cache Load/St Q

Result Bus Reg File Bpred

Performance Cache Hit? Bpred

Lookup?

Inst. Window Full?

Dependencies Satisfied?

Resources?

Commit Bandwidth?

One Cycle in Wattch

• On each cycle:– determine which units are accessed– model execution time issues– model per-unit energy/power based on which units used

and how many ports.

21

Units Modeled by Wattch

Array Structures Caches, Reg Files,

Map/Bpred tables Content-

Addressable Memories (CAMs) TLBs, Issue Queue,

Reorder Buffer Complex combinational blocks ALUs, Dependency

Check Clocking network Global Clock

Drivers, Local Buffers

22

PowerTimer

• IBM Tool First Develop During Summer of 2000– Continued Development: 2001 => Today– Methodology Applied to Research and

Product Power-Performance Simulators with IBM

– Currently in Beta-Release– Working towards Full Academic Release

23

PowerTimer: Empirical Power

IFU

6%

ISU

10%

LSU

19%

GX

1%

L2

23%

Clock Tree

10%

RAS

5%

FBC

3%

CIU

4%

ZIO

4%

FPU

5%Core

Buffer

1%

FXU

4%IDU

3%L3 Tags

2%

Issue Queues

32%

Completion Table9%

Dispatch6%

Other10%

Map Tables43%

Pre-silicon, POWER4-like superscalar design

24

Processor Power Density

Pre-silicon, POWER4-like superscalar designOriginally presented at PACS2002

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

IFU IDU ISU FXU LSU FPU L2 L3 Tag BHT Icache FX-IssueQ

25

PowerTimer

Circuit PowerData (Macros)

Tech Parms

uArch Parms

Program Executable or Trace

SubUnit Power = f(SF, uArch, Tech)

ComputeSub-Unit

Power

Architectural Performance

Simulator

Power

CPI

AF/SF Data

26

PowerTimer: Energy Models

• Energy models for uArch structures formed by summation of circuit-level macro data

Power=C1*SF+HoldPower

Power=C2*SF+HoldPower

Power=Cn*SF+HoldPower

Macro1

Macro2

MacroN

Sub-Units (uArch-level Structures)

Energy Models

SF Data

Power Estimate

27

Empirical Estimates with CPAM

• Estimate power under “Input Hold” and “Input Switching” Modes

• Input Hold: All Macro Inputs (Except Clocks) Held– Can also collect data for Clock Gate Signals

• Input Switching: Apply Random Switching Patterns with 50% Switching on Input Pins

MacroMacro Inputs

• 0% Switching (Hold Power)• 50% Switching Power

28

Example Unit

• Made up of 5 macros

0

100

200

300

400

500

600

700

800

0 10 20 30 40 50SF

mW

macro1

macro2

macro3

macro4

macro5

total

29

PowerTimer: Models f(SF)

0

200

400

600

800

1000

1200

1400

0 10 20 30 40 50SF

mW

Unit1

Unit2

Unit3

Unit4

Unit5

Assumption: Power linearly dependent on Switching FactorThis separates Clock Power and Switching Power

At 0% SF, Power = Clock Power (significant without clock gating)

ClockPower

SwitchingPower

30

Key Activity Data

• SF => Moves along the Switching Power Curve– Estimated on a per-unit basis from RTL Analysis

• AF => Moves along the Clock Power Curve– Extracted from Microarchitectural Statistics (Turandot)

0

200

400

600

800

1000

1200

1400

0 10 20 30 40 50SF

mW

Unit1

Unit2

Unit3

Unit4

Unit5

Changes in SF

Changes in AF

31

Microarchitectural Statistics

• Stats are very similar to tracking used in Wattch, etc• Differences:

– Clock Gating Modes (3 modes)– Customized Scaling Based on Circuit Style (4 styles)

• Clock Gating Modes:– P_constrained = P_unconstrained (not clock-gateable)– P_constrained_1 = AF * (Pclock + Plogic) (common)– P_constrained_2 = AF * Pclock + Plogic (rare)– P_constrained_3 = Pclock + AF * Plogic (very rare)

• Scaling Based on Circuit Styles– AF_1 = #valid (Latch-and-Mux, No Stall

Gating)– AF_2 = #valid - #stalls (Latch-and-Mux, With Stall

Gating)– AF_3 = #writes (Arrays that only gate

updates)– AF_4 = #writes + #reads (Arrays, RAM Macros)

32

Clock Gating: Valid-Bit Gating

V V V V V V

Clock

• Latch-Based Structures: Execute Pipelines, Issue Queues

33

Clock Gating Modes

• P_constrained_1 = AF * (Pclock + Plogic)

Plogic

clockvalid

• P_constrained_2 = AF * Pclock + Plogic

SelectionLogic

Pclock

Pclock Plogic

clockvalid

34

Valid-bit Gating, Stalls?

• Option 1: Stalls cannot be gated

clkvalid

Data From Previous Pipestage

Data For Next Pipestage

Stall FromPrevious Pipestage

clkvalid

Data From Previous Pipestage

Data For Next Pipestage

Stall FromPrevious Pipestage

• Option 2: Stalls can be gated

35

Scaling: Array Structures

• Option 1: Reads and Writes Eligible to Gate for Power

Cell

Write Bitline

write_wordline_activewrite_gate

Read Bitline

write_gate write_data

read_wordline_activeread_gate

read_data

36

Scaling: Array Structures

• Option 2: Only Writes Eligible to Gate for Power

Cell

Write Bitline

write_wordline_activewrite_gate

write_gate write_data

read_data

read_entry_0

read_entry_1

read_entry_2

read_entry_n

Gating

Mode

Valid Valid+

Stalls

Writes Writes+

Reads

Gate

Both

Gate

Clock

Gate

Logic

Examples

0 No No No No No No No Control Logic, Buffers, Small Macros

1 Yes No No No Yes No No Issue Queues, Execute Pipelines2 No Yes No No Yes No No

3 No No Yes No Yes No No Caches

4 No No No Yes Yes No No Some Queues

5 Yes No No No No Yes No CAMs, Selection Logic

6 No Yes No No No Yes No

7 No No Yes No No Yes No No Known macros

8 No No No Yes No Yes No No Known macros

9 Yes No No No No No Yes No Known macros

10 No Yes No No No No Yes No Known macros

11 No No Yes No No No Yes No Known macros

12 No No No Yes No No Yes No Known macros

12 Clock Gating Modes

38

PowerTimer Observations

• PowerTimer works well for POWER4-like estimates and derivatives– Scale base microarchitecture quite well– E.g. optimal power-performance pipelining

study– Lack of run-time, bit-level SF not seen as a

problem within IBM (seen as noise)• Chip bit-level SFs are quite low (5-15%)• Most (60-70%) power is dissipated while

maintaining state (arrays, latches, clocks)• Much state is not available in early-stage timers

39

Comparing models: Flexibility

• Flexibility necessary for certain studies– Resource tradeoff analysis– Modeling different architectures

• Purely analytical tools provides fully-parameterizable power models– Within this methodology, circuit design styles could

also be studied

• PowerTimer scales power models in a user-defined manner for individual sub-units– Constrained to structures and circuit-styles

currently in the library

• Perhaps Mixed Mode tools could be very useful

40

Comparing models: Accuracy

• PowerTimer -- Based on validation of individual pieces– Extensive validation of the performance model (AFs)– Power estimates from circuits are accurate – Circuit designers must vouch for clock gating

scenarios– Certain assumptions will limit accuracy or require

more in-depth analysis

• Analytical Tools– Inherent Issues

• Analytical estimates cannot be as accurate as SPICE analysis (“C” estimates, CV2 approximation)

– Practical Issues• Without industrial data, must estimate transistor sizing,

bits per structure, circuit choices

41

Comparing models: Speed

• Performance simulation is slow enough!• Post-Processing vs. Run-Time Estimates• Wattch’s per-cycle power estimates: roughly

30% overhead– Post-processing (per-program power estimates)

would be much faster (minimal overhead)

• PowerTimer allows both no overhead post-processing and run-time analysis for certain studies (di/dt, thermal)– Some clock gating modes may require run-time

analysis

• Third Option: Bit Vector Dumps– Flexible Post-Processing Huge Output Files

42

Static Power Dissipation

• Static power: dissipation due to leakage current

• Growing worse because as Vdd scales, Vth must also scale to maintain switching speeds– “Off” transistors are no longer fully cut off

43

Leakage Power

• The fraction of leakage power is increasing exponentially with each generation

• Also exponentially dependent on temperature

Static power/ Dynamic Power

0

10

20

30

40

50

60

70

298

303

308

313

318

323

328

333

338

343

348

353

358

363

368

373

Temperature(K)

Pe

rcen

tag

e

180nm 130nm 100nm 90nm 80nm 70nm

Increasingratioacrossgenerations

Source: Sankaranarayanan et al, University of Virginia, based on ITRS 2001 data

44

Static Power - HotLeakage• Main mechanisms for leakage current

– Subthreshold (Berkely predictive model):

– Gate• Igate = Igate0 * exp(a*(tox-tox0)) * exp(b*(vdd-vdd0))

• There is also a weak temp dependence, ignored here

• Gate leakage has essentially been ignored – New gate insulation materials may reduce urgency of

problem, eg recent Intel announcement• R. Chau, Technology@intel Magazine. www.intel.com

• Gate-induced drain leakage (GIDL) occurs at negative gate voltages and high Vdd or high values of reverse body bias– New leakage path opens!

t

offthv

V

tVddVddba

OXleakage vn

VVeve

L

WCI t

dd

02)*(0 exp10

45

Effects of Parameter Variations

• Ioff depends exponentially on Vth

• There is a large fluctuation of Ioff from die to die and from gate to gate

• Controlling Vth is difficult in nanometer scale– Drain-induced barrier lowering

• Channel length is not constant• Exacerbated in sub-100nm devices

– Discrete dopant effects• In a very small channel, small number of dopants• Presence of these dopants and random fluctuation of their

number, lead to changes in Vth from device to device

• Process variation affects– Gate length (Ldrawn)– Gate oxide thickness (Tox)– Channel dose (Nsub)

Srivastava et al, ISLPED 2002

46

Static Power

Modeling Leakage– Butts and Sohi (MICRO-33)

• Pstatic = Vcc · N · kdesign · Îleak

• Îleak determined by circuit simulation, kdesign empirically• Key contribution: separate technology from design

– HotLeakage (UVA TR CS-2003-05, DATE’04)• Extension of Butts & Sohi approach: scalable with Vdd, Vth,

Temp, and technology node; adds gate leakage• Îleak determined by BSIM3 subthreshold equation and BSIM4

gate-leakage equations, giving an analytical expression that accounts for dependence on factors that may change at runtime, namely Vdd, Vth, and Temp

• kdesign replaced by separate factors for N- and P-type transistors

• kdesign also exponentially dependent on Vdd and Tox, linearly dependent on Temp

• Currently integrated with SimpleScalar/Wattch for caches• http://lava.cs.virginia.edu/HotLeakage

47

Static Power

• Modeling Leakage (cont.)– Su et al, IBM (ISLPED’03)

• Similar approach to HotLeakage – but they observe that modeling the change in leakage allows linearization of the equations

– Many, many other papers on various aspects of modeling different aspects of leakage

• Most focus on subthreshold• Few suggest how to model leakage in

microarchitecture simulations

48

Power modeling summary

• Wattch provides excellent relative accuracy– Underestimates full chip power (some units

not modeled, etc)

• PowerTimer models based on circuit-level power analysis– Inaccuracy is introduced in SF/AF and

scaling assumptions

• Modeling leakage is essential for thermal research– “Thermal runaway” is even possible

49

Overview

1. Motivation (Kevin)2. Thermal issues (Kevin)3. Power modeling (David)4. Thermal management (David)5. Optimal DTM (Lev)6. Clustering (Antonio)7. Power distribution (David)8. What current chips do (Lev)9. HotSpot (Kevin)

50

Existing Work

• Research Ideas– DEETM – Huang and Torrellas MICRO2000– DTM – Brooks and Martonosi HPCA2001– Control-Theoretic DTM – Skadron, Abdelzaher,

Stan HPCA2002– Thermal Scheduling – Cai, Lim, Daasch

WCED2002

• Commercial Products– PowerPC G3 Microprocessor– Pentium III– Pentium 4

51

Overview

• Hard to optimize power-performance at design time for all cases

• Forces conservative choices for issues like cooling, current delivery, resource sizes

• Want to explore dynamic power optimizations for run-time power management– Dynamic Voltage/Frequency Scaling [Burd, 2000]– Dynamic Hardware Resizing [Albonesi, 1999]– Fetch Throttling [Sanchez, 1997]– Global Clock Gating [Gunther, 2001]– Speculation Control [Manne, 1998]– Dynamic Thermal Management [Brooks, 2001]

[Huang, 2000]

52

Important to optimize P & T early

00.5

11.5

22.5

33.5

44.5

5

0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6Relative 1/Performance

Rel

ativ

e Po

wer

tradeoff via changing Vdd and HItradeoff via changing frequencytradeoff via changing pipeline depth

18FO4 23FO4

14FO4

12FO4

Maximum Power Budget

53

Dynamic Thermal Management

• Goal: – Provide dynamic techniques to cool chip

when needed– Exploit natural variations due to different

applications, phase behavior, …– Allow designers to target average, rather

than worst-case behavior

• Design Decisions:– Mechanism & policy for triggering response?– What should response be?– How to select DTM trigger levels?

54

Power consumption impacts cost

• System costs associated with power dissipation:– Thermal control cost

• Heatsinks, fans– Power delivery

• Power supply• Decoupling caps…

CPU

From: Gunther, et al. “Managing the Impact of Increasing Microprocessor Power Consumption,” Intel Technology Journal, Q1, 2001

55

Average and Worst Case Power

• System costs are constrained by worst case power dissipation

• Average case power dissipation can often be much lower– Aggressive Clock Gating– Applications variations– Underutilized resources

• Not enough ILP• Floating point units during integer

code execution

• Currently about a 30% difference

• Likely to further diverge…

0

20

40

60

80

100

Alpha 21264 Intel PPro

Max

Avg

56

Dynamic Thermal Management

Time

Tem

pera

ture

DTM Disabled DTM/Response Engaged

Designed for Cooling Capacity w/out DTM

DTM TriggerLevel

Designed for Cooling Capacity w/ DTM

SystemCost Savings

57

DTM: Definitions

TriggerReached

Response Delay – Invocation time (e.g. adjust clock)

ResponseDelay

Policy Delay – Number of cycles engaged

PolicyDelay

CheckTemp

CheckTemp

TurnResponse

On

InitiationDelay

Initiation Delay – OS interrupt/handler

ShutoffDelay

TurnResponse

Off

Shutoff Delay – Disabling time (e.g. re-adjust clock)

58

DTM: When, How, and What

Initiation Mechanism:

How do we enabletechnique?

Response Mechanism: What technique do we enable?

Trigger Mechanism:

When do we enable

DTM techniques?

59

DTM: Trigger Mechanisms

• Mechanism: How to deduce temperature?

• Direct approach: Temperature sensors providing feedback– Implemented in some

PowerPC chips (G3, G4) [Sanchez, 1997]

– Sensor quantity, placement, and precision will be discussed later

• Other indirect approaches possible

Policy: When to begin responding? Trigger level set too high:

Packaging cost will be high

Little advantage Trigger level set too low

Frequent triggering causes performance to suffer

Choose trigger level to exploit difference between average and worst-case power.

60

DTM: Initiation Mechanisms

• Operating system or microarchitectural control?– Hardware support can significantly reduce

performance penalty

• Policy Delay Settings– For Volt/Freq scaling, much of the performance

penalty can be attributed to enabling/disabling – Increasing policy delay reduces overhead;

smarter initiation techniques would help as well

61

DTM: Response Mechanisms

• Scaling Techniques– Clock Frequency Scaling [Intel Pentium 4]– Voltage and Frequency Scaling– Temperature-tracking frequency scaling[Skadron03]

• Adjusts frequency to account for T-dep. of switching speed• Microarchitectural Techniques

– Speculation Control [Manne98]– Low-Power Cache Techniques [Huang00]

• Hierarchical Responses– Decode Throttling [Sanchez97]– Fetch Toggling [Brooks01]– Feedback controlled Fetch Gating [Skadron02]– Migrating Computation [Skadron03]– Dual Pipelines [Lim02]– Hybrid DTM [Skadron04]

62

Dynamic Voltage/Frequency Scale

• Voltage Scheduler predicts workload requirements

• Set frequency/voltage to near-optimal, energy savings

• Burd, et al., ISSCC2000– 5MHz @ 1.2V: 6 MIPS, 2.8mW– 80MHz @ 3.8V: 85 MIPS, 460mW– 70us 1.2V <-> 3.8V

• Transmeta Crusoe– Commercial implementation (500-700MHz,

1.2-1.6V)

63

Temperature-Tracking Frequency

Temperature affects :

• Transistor threshold and mobility

• Ion, Ioff, Igate, delay

• ITRS: 85°C for high-performance, 110°C for embedded!

• So adjust frequency as f(T) -- TTDFS

IonNMOS

Ioff

64

Speculation Control

• Manne et al. (ISCA ’98)– Branch confidence estimator used to determine

whether to speculate– Pipeline gating based on confidence estimation– 38% reduction in wrong-path instructions with ~1%

performance loss

• But Parikh et al. (HPCA ’02) found much smaller savings; ED product is zero or negative– Significant energy savings only come with significant

loss of performance– This is because many instructions are squashed early

in the pipeline, so reduction in wrong-path instructions is not a useful metric

– Benefit is actually a function of prediction accuracy• Only for very badly predicted programs do you get

benefit • Well-predicted programs suffer

65

Dynamic Hardware Resizing

• Complexity Adaptive Processors• Based on application characteristics

– Underutilized structures may be reduced with minimal performance impact

– Resize Caches, Issue Queues, etc.– Resize => Reduce Capacitance => Reduce

Energy– Of course, this only helps manage heat if it

reduces heat dissipation within hot spots• And does so for a sufficiently long duration

66

DEETM

• Dynamic Energy Efficiency and Temperature Management

• Slack algorithm detects if slowdown can be tolerated– If so, invoke techniques to reduce energy

• Temperature algorithm– If temperature limit is reached, invokes

techniques

• Techniques considered– Filter Cache, Voltage Scaling, etc.

67

Control-theoretic DTM

• Fetch toggling– disable fetch every N cycles– 4/5, 2/3, 1/2, 1/3, 1/5, …

– How to set the fetch rate?• (Assume idealized temperature sensing)

IF ID EX MEM WB

IF ID EX MEM WB

68

Feedback-Control of Fetch Toggling

• Formal feedback control

Integral ctrl: u[k] = u[k-1] + Kc · e[k-1]

• easy to compute• toggling = f(m)• optimality: will discuss shortly

Controller Actuator:I-fetch toggling

Thermaldynamics

Temp. sensor

e u[k]setpoint P T

measured T

Original work used PID

69

Formal Feedback Control

• Regulatory control problem: hold value to a specified setpoint– Example: temperature

– Proved that PID controller will not allow temperature to exceed setpoint by more than 0.02°

• Max power dissipation, thermal dynamics,sampling rate max overshoot

• This precision is excessive but illustrates the value of formal feedback control theory

70

Performance Loss

• Performance loss reduced by 65%

0%

5%

10%

15%

20%

25%

30%gc

c

mes

a art

equa

ke

craf

ty

face

rec

fma3

d

pars

er

eon

perlb

mk

vorte

x

bzip

ME

AN

Per

cen

t Lo

ss in

Per

form

ance

toggle1

PID

71

Migrating Computation

• When one unit overheats, migrate its functionality to a distant, spare unit (MC)– Spare register file (Skadron et al. 2003)– Separate core (CMP) (Heo et al. 2003)– Microarchitectural clusters– etc.

• Raises many interesting issues– Cost-benefit tradeoff for that area– Use both resources (scheduling)– Extra power for long-distance communication– Floorplanning

• This is probably one of the richest opportunities for DTM

72

Migrating Computation – Reg File

73

Thermal Scheduling (Cai 2002)

• Primary pipeline: maximal performance, complex pipeline structure

• Second pipeline: Minimum power and energy consumption, very simple in order structure and target mobile anywhere-anytime applications.

• Transparent to OS and applications• Maximal utilizing on die clock/power gating for

energy saving

FE DE EX

RF

DE IOP

OOP

Primary

Secondary

Majority mobile apps with performance requirements

Text email, caller-id, reminder and other none high performance w/ anywhere-anytime requested apps

74

Scheduling Algorithm (Cai 2002)

S1: Normal Operation (Primary Pipeline)S2: Stall Fetch & Clear PipelineS3: Alternate Operation (Secondary Pipeline)S4: Disable Clock or Scale F-V

S1

S2

S3

T1 < TH

S4

T1 TH

T1 TH

T1 TL

T1 TL

T1 > TL

& T2 < TH

T1 > TL

& T2 TH

T1 > TL

|| T2 > TL

T1 TL

& T2 TL

TS2

TS1

Temperature (C)Ta

Tmax

tcycle

TH

tcool theat

Time (s)

TL

75

Hybrid DTM

• DVS is attractive because of its cubic advantage– P V2f– This factor dominates when DTM must be aggressive– But changing DVS setting can be costly

• Resynchronize PLL• Sensitive to sensor noise spurious changes

• “ILP techniques” are attractive because they can use instruction level parallelism to hide/reduce impact of DTM

– Only effective when DTM is mild

• So use both!– Need to find “crossover point”

76

Hybrid DTM, cont.

• Combine fetch gating with DVS– When DVS is better, use it– Otherwise use fetch gating– Determined by magnitude of temperature overshoot– Crossover at FG duty cycle of 3– FG has low overhead: helps reduce cost of sensor

noise

1.0

1.1

1.2

1.3

20 5 2Duty Cycle

Slo

wd

ow

n

1.0

1.1

1.2

1.3

1.4

05101520Duty Cycle

Slo

wd

ow

n

FG

DVSHyb

77

Hybrid DTM, cont.

• DVS doesn’t need more than two settings for thermal control

– Lower voltage cools chip faster

• FG by itself does need multiple duty cycles and hence requires PI control

• But in a hybrid configuration, FG does not require PI control

– FG is only used at mild DTM settings– Can pick one fixed duty cycle

• This is beneficial because feedback control is vulnerable to noise

78

Simulation Details

• 85°C maximum temperature– Guard band requires a trigger threshold of 81.8°

• Ambient temperature (inside computer case): 45°C

• Rpackage = 0.8 K/W (old package model)– 0.7 K/W necessary if DTM not available

• Die thickness: 0.5mm• Currently neglecting interface material

• 9 SPEC2000 benchmarks, both integer and FP– 4 hover near 81.8°C, rest are above

• SimpleScalar/Wattch, modified to model pipeline and power of an Alpha 21364 as closely as possible

• Scaled to 130nm, 1.3V, 3.0 GHz

79

Performance Comparison

• TT-DFS is best but can’t prevent excess temperature– Suitable for use with aggressive clock rates at low temp.

• Hybrid technique reduces DTM cost by 25% vs. DVS (DVS overhead important)

• A substantial portion of MC’s benefit comes from the altered floorplan, which separates hot units

1.045

1.270

1.359

1.231

1.112

1.00

1.10

1.20

1.30

1.40

TTDFS DVS FG Hyb MC

Slo

wd

ow

n F

acto

r

80

Floorplan Impact2 cycle penalty

LSQ moves:no penalty

81

Conclusions so far

• DTM can be used to reduce cooling costs• Proper modeling is required

– HotSpot is publicly available athttp://lava.cs.virginia.edu/HotSpot

• ILP matters• Hybrid techniques beneficial

– Merge advantages of different schemes– Simplify control

• Architectural techniques important in thermal design

• Growing use of clusters and redundant units opens an incredibly rich design space

82

DTM: Summary and Key Issues

• Dynamic optimizations translate max-power problem to average-power problem

• Heightens importance of average-power techniques like clock gating

• Key Issues:– Initiation interval– Collection of possible response mechanisms– Sensor accuracy and data fusion– Hot spots