TY - JOUR
T1 - Code coverage-based power estimation techniques for microprocessors
AU - Qu, Gang
AU - Kawabe, Naoyuki
AU - Usami, Kimiyoshi
AU - Potkonjak, Miodrag
PY - 2002/10
Y1 - 2002/10
N2 - We have developed a function-level power estimation methodology for predicting the power dissipation of embedded software. For a given microprocessor core, we empirically build the "power data bank", which stores the power information of the built-in library functions and basic instructions. To estimate the average power of an embedded software on this core, we first decompose the machine code into library functions and user-defined functions. We then use program profiling/tracing tools to get the execution information of the target software. Next, we evaluate the total energy consumption and execution time based on the "power data bank", and their ratio is taken as the average power. High efficiency is achieved because no power simulator is used once the "power data bank" is built. We apply this method to a commercial microprocessor core and get power estimates with an average error of 3%. Using this method, microprocessor vendors can provide users the "power data bank" without releasing details of the core to help users get early power estimates and eventually guide power optimization.
AB - We have developed a function-level power estimation methodology for predicting the power dissipation of embedded software. For a given microprocessor core, we empirically build the "power data bank", which stores the power information of the built-in library functions and basic instructions. To estimate the average power of an embedded software on this core, we first decompose the machine code into library functions and user-defined functions. We then use program profiling/tracing tools to get the execution information of the target software. Next, we evaluate the total energy consumption and execution time based on the "power data bank", and their ratio is taken as the average power. High efficiency is achieved because no power simulator is used once the "power data bank" is built. We apply this method to a commercial microprocessor core and get power estimates with an average error of 3%. Using this method, microprocessor vendors can provide users the "power data bank" without releasing details of the core to help users get early power estimates and eventually guide power optimization.
KW - Code decomposition
KW - Function-level power model
KW - Microprocessor cores
KW - Power estimation
KW - Program profiling and tracing
UR - http://www.scopus.com/inward/record.url?scp=0347934777&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0347934777&partnerID=8YFLogxK
U2 - 10.1142/S0218126602000616
DO - 10.1142/S0218126602000616
M3 - Article
AN - SCOPUS:0347934777
VL - 11
SP - 557
EP - 574
JO - Journal of Circuits, Systems and Computers
JF - Journal of Circuits, Systems and Computers
SN - 0218-1266
IS - 5
ER -