TY - GEN
T1 - Simulated Annealing for Timeliness and Energy aware Deep Learning Job Assignment
AU - Kang, Dong Ki
AU - Youn, Chan Hyun
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/10
Y1 - 2019/10
N2 - In this paper, we propose a simulated annealing method for timeliness and energy aware deep learning (DL) job assignment. In the proposed method, we make a decision of server position for DL training jobs in order to consider both the timeliness of deep neural network (DNN) model updating and the associated energy consumption. For timeliness management, we design three penalty functions; step, linear, and exponential functions, so as to practically penalize the service quality degradation of DL inference due to delay of DL training. For energy management, we formulate the operation cost optimization problem considering energy consumption for DL training. By using these management schemes, our method is able to find the trade-off of the timeliness and energy consumption of DL training. Especially, in order to overcome the non-smoothness of the defined problem, we design a simulated annealing (SA) based metaheuristic which finds approximated optimal solution for DL training job assignment. For the performance evaluation, we show the preliminary experimental results of DL training jobs with AlexNet, Inception-V3, and ResNet on NVIDIA GPU devices.
AB - In this paper, we propose a simulated annealing method for timeliness and energy aware deep learning (DL) job assignment. In the proposed method, we make a decision of server position for DL training jobs in order to consider both the timeliness of deep neural network (DNN) model updating and the associated energy consumption. For timeliness management, we design three penalty functions; step, linear, and exponential functions, so as to practically penalize the service quality degradation of DL inference due to delay of DL training. For energy management, we formulate the operation cost optimization problem considering energy consumption for DL training. By using these management schemes, our method is able to find the trade-off of the timeliness and energy consumption of DL training. Especially, in order to overcome the non-smoothness of the defined problem, we design a simulated annealing (SA) based metaheuristic which finds approximated optimal solution for DL training job assignment. For the performance evaluation, we show the preliminary experimental results of DL training jobs with AlexNet, Inception-V3, and ResNet on NVIDIA GPU devices.
KW - deep learning inference
KW - deep learning training
KW - deep neural network
KW - energy consumption
KW - power usage
KW - timeliness
UR - https://www.scopus.com/pages/publications/85078260522
U2 - 10.1109/ICTC46691.2019.8939901
DO - 10.1109/ICTC46691.2019.8939901
M3 - Conference paper
AN - SCOPUS:85078260522
T3 - ICTC 2019 - 10th International Conference on ICT Convergence: ICT Convergence Leading the Autonomous Future
SP - 571
EP - 575
BT - ICTC 2019 - 10th International Conference on ICT Convergence
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 10th International Conference on Information and Communication Technology Convergence, ICTC 2019
Y2 - 16 October 2019 through 18 October 2019
ER -