Task-level Thermal Modeling for Temperature Management of Edge TPU

Edge TPU (Tensor Processing Unit) is being widely utilized in various edge computing applications as a high-efficiency, low-power accelerator for deep learning computations. However, temperature rise in Edge TPU can lead to performance degradation, reduced stability, and shortened lifespan, necessit...

Full description

Saved in:
Bibliographic Details
Published in2024 IEEE 30th International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA) pp. 138 - 139
Main Authors Han, Changhun, Oh, Sangeun
Format Conference Proceeding
LanguageEnglish
Published IEEE 21.08.2024
Subjects
Computational modeling
Deep learning
Edge TPU
Predictive models
Real-time systems
Stability analysis
Temperature
Temperature Management
Tensors
Thermal degradation
Thermal management
Thermal Modeling
Thermal stability
Online AccessGet full text

Cover

More Information
Summary:Edge TPU (Tensor Processing Unit) is being widely utilized in various edge computing applications as a high-efficiency, low-power accelerator for deep learning computations. However, temperature rise in Edge TPU can lead to performance degradation, reduced stability, and shortened lifespan, necessitating temperature management through thermal modeling. This paper proposes a task-level thermal modeling technique for predicting Edge TPU temperature. The proposed method estimates power consumption of CPU and Edge TPU based on workloads of various deep learning tasks and predicts the convergence temperature of Edge TPU using a steady temperature model. Through experiments, we confirmed that the proposed method accurately predicts Edge TPU temperature for various workloads. The average prediction error was 0.7°C. This study is expected to serve as a foundation for developing temperature management techniques by presenting an effective temperature prediction model that considers the thermal characteristics of Edge TPU.
ISSN:2325-1301
DOI:10.1109/RTCSA62462.2024.00032