TY - GEN
T1 - Performance optimal processor throttling under thermal constraints
AU - Rao, Ravishankar
AU - Vrudhula, Sarma
PY - 2007/12/1
Y1 - 2007/12/1
N2 - We derive analytically, the performance optimal throttling curve for a processor under thermal constraints for a given task sequence. We found that keeping the chip temperature constant requires an exponential speed curve. Earlier works that propose constant throttling only keep the package/case temperature constant, and are hence suboptimal. We develop high-level thermal and power models that are simple enough for analysis, yet account for important effects like the power-density variation across a chip (hotspots), leakage dependence on temperature (LDT), and differing thermal characteristics of the silicon die and the thermal solution. We use a piecewise-linear approximation for the exponential leakage dependence on temperature, and devise a method to remove the circular dependency between leakage power and temperature. To solve the multi-task speed control problem, we first solve analytically, the single task problem with a constraint on the final package temperature using optimal control theory. We then find the optimum final package temperature of each task by dynamic programming. We compared the total execution time of several randomly generated task sequences using the optimal control policy against a constant speed throttling policy, and found significantly smaller total execution times. We compared the thermal profiles predicted by the proposed high-level thermal model to that of the Hotspot thermal model, and found them to be in good agreement.
AB - We derive analytically, the performance optimal throttling curve for a processor under thermal constraints for a given task sequence. We found that keeping the chip temperature constant requires an exponential speed curve. Earlier works that propose constant throttling only keep the package/case temperature constant, and are hence suboptimal. We develop high-level thermal and power models that are simple enough for analysis, yet account for important effects like the power-density variation across a chip (hotspots), leakage dependence on temperature (LDT), and differing thermal characteristics of the silicon die and the thermal solution. We use a piecewise-linear approximation for the exponential leakage dependence on temperature, and devise a method to remove the circular dependency between leakage power and temperature. To solve the multi-task speed control problem, we first solve analytically, the single task problem with a constraint on the final package temperature using optimal control theory. We then find the optimum final package temperature of each task by dynamic programming. We compared the total execution time of several randomly generated task sequences using the optimal control policy against a constant speed throttling policy, and found significantly smaller total execution times. We compared the thermal profiles predicted by the proposed high-level thermal model to that of the Hotspot thermal model, and found them to be in good agreement.
KW - Leakage dependence on temperature
KW - Power
KW - Thermal management
KW - Thermal model
KW - Throttling
UR - http://www.scopus.com/inward/record.url?scp=38849163446&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=38849163446&partnerID=8YFLogxK
U2 - 10.1145/1289881.1289925
DO - 10.1145/1289881.1289925
M3 - Conference contribution
AN - SCOPUS:38849163446
SN - 9781595938268
T3 - CASES'07: Proceedings of the 2007 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems
SP - 257
EP - 266
BT - CASES'07
T2 - CASES'07: 2007 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems
Y2 - 30 September 2007 through 3 October 2007
ER -