TensorFlow: A system for large-scale machine learning

TLDR

TensorFlow is a large‑scale machine‑learning system designed for heterogeneous computing environments. The paper describes TensorFlow’s dataflow model and demonstrates its performance on real‑world applications. TensorFlow represents computation as dataflow graphs, mapping nodes across machines and devices—including CPUs, GPUs, and TPUs—to enable distributed execution. Its architecture gives developers flexibility to experiment with optimizations, supports deep neural networks, and is widely adopted in Google services and research.

Abstract

TensorFlow is a machine learning system that operates at large scale and in heterogeneous environments. TensorFlow uses dataflow graphs to represent computation, shared state, and the operations that mutate that state. It maps the nodes of a dataflow graph across many machines in a cluster, and within a machine across multiple computational devices, including multicore CPUs, general-purpose GPUs, and custom designed ASICs known as Tensor Processing Units (TPUs). This architecture gives flexibility to the application developer: whereas in previous "parameter server" designs the management of shared state is built into the system, TensorFlow enables developers to experiment with novel optimizations and training algorithms. TensorFlow supports a variety of applications, with particularly strong support for training and inference on deep neural networks. Several Google services use TensorFlow in production, we have released it as an open-source project, and it has become widely used for machine learning research. In this paper, we describe the TensorFlow dataflow model in contrast to existing systems, and demonstrate the compelling performance that TensorFlow achieves for several real-world applications.