发明名称 |
DATA PARALLEL PROCESSING METHOD AND APPARATUS BASED ON MULTIPLE GRAPHIC PROCESSING UNITS |
摘要 |
A parallel data processing method based on multiple graphic processing units (GPUs) is provided, including: creating, in a central processing unit (CPU), a plurality of worker threads for controlling a plurality of worker groups respectively, the worker groups including one or more GPUs; binding each worker thread to a corresponding GPU; loading a plurality of batches of training data from a nonvolatile memory to GPU video memories in the plurality of worker groups; and controlling the plurality of GPUs to perform data processing in parallel through the worker threads. The method can enhance efficiency of multi-GPU parallel data processing. In addition, a parallel data processing apparatus is further provided. |
申请公布号 |
US2016321777(A1) |
申请公布日期 |
2016.11.03 |
申请号 |
US201615210278 |
申请日期 |
2016.07.14 |
申请人 |
TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED |
发明人 |
Jin Xing;Li Yi;Zou Yongqiang;Guo Zhimao;Wang Eryu;Xue Wei;Chen Bo;Li Yong;Bao Chunjian;Xiao Lei |
分类号 |
G06T1/20;G06F9/52;G06F9/50;G06T1/60 |
主分类号 |
G06T1/20 |
代理机构 |
|
代理人 |
|
主权项 |
1. A parallel data processing method based on multiple graphic processing units (GPUs), comprising:
creating, in a central processing unit (CPU), a plurality of worker threads for controlling a plurality of worker groups respectively, the worker groups comprising one or more GPUs; binding each worker thread to a corresponding GPU; loading a plurality of batches of training data from a nonvolatile memory to GPU video memories in the plurality of worker groups; and controlling the plurality of GPUs to perform data processing in parallel through the worker threads. |
地址 |
Shenzhen CN |