• Eugenio Gianniti e Danilo Ardagna - Politecnico di Milano, Milan, Italy
  • Li Zhang - IBM T. J. Watson Research Center, NY, USA

Recent years saw an increasing success in the application of deep learning methods across various domains and for tackling different problems, ranging from image recognition and classification to text processing and speech recognition. In this paper we propose and validate an approach to model the execution time for training convolutional neural networks (CNNs) deployed on GPGPUs. We demonstrate that our approach is generally applicable to a variety of CNN models and different types of GPGPUs with high accuracy, aiming at the preliminary design phases for system sizing

Where: SBAC­-PAD 2018 - 30th International Symposium on Computing Architecture and High Performance Computing, September 2018, Lyon (France)