Abstract
Tensor Processing Units (TPUs) are specialized hardware accelerators for deep learning developed by Google. This poster aims to explore TPUs in cloud and edge computing focusing on its applications in AI. An in-depth overview of TPUs is presented, highlighting their unique architectural design tailored to neural network computations. The architecture is dissected to reveal how it differentiates TPUs from traditional chip architectures, particularly in the context of matrix operations which are central to neural network processing.
Furthermore, key aspects of TPU functionality, such as compilation techniques and the integration with supporting frameworks like TensorFlow, PyTorch, and JAX, are examined. This analysis underpins the understanding of how TPUs optimize the execution of deep learning models, thus providing a clear insight into their operational efficiency on a programmatic level. The poster also delves into a comprehensive comparative analysis, evaluating the performance of Cloud and Edge TPUs against other established chip architectures like CPUs and GPUs. This comparison is crucial in demonstrating the substantial performance improvements TPUs offer in both cloud and edge computing scenarios.
Additionally, the poster underscores the imperative need for further research in optimization techniques for efficient deployment of AI architectures on the Edge TPU and benchmarking standards for a more robust comparative analysis in edge computing scenarios. The primary motivation behind this push for research is that efficient AI acceleration, facilitated by TPUs, can lead to substantial savings in terms of time, money, and environmental resources.
You have full access to this open access chapter, Download conference paper PDF
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Sanmartín Carrión, D., Prohaska, V., Diez, O. (2024). Exploration of TPUs for AI Applications. In: Daimi, K., Al Sadoon, A. (eds) Proceedings of the Second International Conference on Advances in Computing Research (ACR’24). ACR 2024. Lecture Notes in Networks and Systems, vol 956. Springer, Cham. https://doi.org/10.1007/978-3-031-56950-0_47
Download citation
DOI: https://doi.org/10.1007/978-3-031-56950-0_47
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-56949-4
Online ISBN: 978-3-031-56950-0
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)