Kevin Skadron

1
The Laboratory for Computer Architecture at
Virginia (LAVA)

• Kevin Skadron
• University of Virginia
• Department of Computer Science

2
Why We Care About Thermal Management...
Source Toms Hardware Guidehttp//www6.tomshardw
are.com/cpu/01q3/010917/heatvideo-01.html
3
Dynamic Thermal Management

• Dynamically adjust execution to control
  temperature
• Avoid catastrophic failure (heat sink, fan)
• Permit the use of a less expensive thermal
  package
• Design for less than the worst case
• Package costs 1 / W above 40 W
• Peak power as high as 130 W in 1-2 generations
  (SIA roadmap)
• Temperatures over 100C

4
Dynamic Thermal Management

• Deal with hot spots
• Localized heating occurs much faster than
  chip-wide
• Chip-wide treatment is too conservative
• Prove temperature will be safely bounded

5
Thermal Modeling

• Want a fine-grained model of temperature
• Power dissipation too indirect, not easy to
  measure in HW

6
Ohms Law for Temperature

• V ? temp
• I ? power
• R ? thermal resistance
• C ? thermal capacitance
• RC ? time constant
• I ?t V ?t
• ?V ------- --------
• C RC
• Lets us compute stepwise changes in temperature
  for any granularity at which we can get P, T, R,
  C
• steady-state V IR (T PR)

7
Thermal Modeling

• Use thermal resistance and capacitance of Si
• Develop computationally efficient model based on
  lumped values
• Pi
  ?t Ti ?t
• ?Ti -------- ---------
• Ci RiCi
• Integrate in Wattch (power/performance
  simulator)
• Time evolution of temperature is driven by unit
  activities and power dissipations on a
  per-cycle basis
• Detect hot spots and activate thermal response
• Typical time constant 10-100 ?s

8
Fetch Toggling

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

IF
ID
EX
MEM
WB
9
Fetch Toggling

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

IF
ID
EX
MEM
WB
IF
ID
EX
MEM
WB
10
Fetch Toggling

• Fetch toggling
• disable fetch every N cycles
• 4/5, 2/3, 1/2, 1/3, 1/5,
• How to set the fetch rate?

IF
ID
EX
MEM
WB
IF
ID
EX
MEM
WB
11
Feedback-Control of Fetch Toggling

• Formal feedback control
• PID m KC (e KI?e Kdde/dt)
• easy to compute
• toggling f(m)

e
m
setpoint
P
T
ActuatorI-fetch toggling
Thermaldynamics
Controller
Temp. sensor
measured T
12
Other Thermal-Management Techniques

• Fetch toggling
• Fetch throttling
• Decode throttling
• Speculation control
• Frequency/voltage scaling

13
Per-Structure Response

• Hot spots
• Branch predictor (probed every cycle)
• Load-store queue
• L1 D-cache (for high-BW apps)
• most major structures are a hot spot for at
  least one SPEC2k app
• Modified Wattch
• Sampling rate 1000 cycles (RC of hot spots is
  10-100 ?s)
• Base temp. of 100?C (SIA roadmap)
• Emergency threshold of 108? (Yuan/Hong SEMI-THERM
  01)
• Set point of 107.9?

14
Thermal Modeling Where to go from here?(i.e.,
lots of research questions)

• Floor-planning issues and granularity of lumped
  R/C values
• Thermal coupling among blocks
• Response lag in temperature sensors
• Validation techniques
• Visualization
• How to deal with large time scales?

15
Thermal Management Where to go from here?
(i.e., lots more research questions)

• New mechanisms
• Characterize benchmarks
• When to use frequency/voltage scaling
• Faster HW techniques for sensing temperature
  changes
• Robust response despite sensor lag
• Hot spots
• Temperature effects on leakage current
• Joint control of temp., power, and performance

16
Thermal Management Where to go from here?
(i.e., lots more research questions)

• New mechanisms
• When to use clock scaling
• Robust response despite sensor lag
• Temperature effects on leakage current
• Joint control of temperature, power, and
  performance

17
Summary

• New tools for thermal management
• Models
• Mechanisms

Source Toms Hardware Guidehttp//www6.tomshardw
are.com/cpu/01q3/010917/heatvideo-01.html
18
Backup slides
19
Performance Loss
Performance loss reduced by 65