Using GPI-2 for Distributed Memory Paralleliziation of the Caffe Toolbox to Speed up Deep Neural Network Training
원문 초록
Deep Neural Network (DNN) are currently of great inter- est in research and application. The training of these net- works is a compute intensive and time consuming task. To reduce training times to a bearable amount at reasonable cost we extend the popular Caffe toolbox for DNN with an efficient distributed memory communication pattern. To achieve good scalability we emphasize the overlap of computation and communication and prefer fine granu- lar synchronization patterns over global barriers. To im- plement these communication patterns we rely on the the Global address space Programming Interface version 2 (GPI-2) communication library. This interface provides a light-weight set of asynchronous one-sided communica- tion primitives supplemented by non-blocking fine gran- ular data synchronization mechanisms. Therefore, Caf- feGPI is the name of our parallel version of Caffe. First benchmarks demonstrate better scaling behavior com- pared with other extensions, e.g., the Intel TM Caffe. Even within a single symmetric multiprocessing machine with four graphics processing units, the CaffeGPI scales bet- ter than the standard Caffe toolbox. These first results demonstrate that the use of standard High Performance Computing (HPC) hardware is a valid cost saving ap- proach to train large DDNs. I/O is an other bottleneck to work with DDNs in a standard parallel HPC setting, which we will consider in more detail in a forthcoming paper.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Identification of Minimally Restrictive Assembly Sequences using Supervisory Control Theory
The kernel of graph indices for vector search
Model-Based Real-Time Pose and Sag Estimation of Overhead Power Lines Using LiDAR for Drone Inspection
PocketVina Enables Scalable and Highly Accurate Physically Valid Docking through Multi-Pocket Conditioning
Pushing the Limits of Extreme Weather: Constructing Extreme Heatwave Storylines with Differentiable Climate Models
NoLoCo: No-all-reduce Low Communication Training Method for Large Models
토론
이 논문과 연결된 포럼 글이 아직 없습니다.