Patents Assigned to Guangdong Inspur Smart Computing Technology Co., Ltd.
-
Patent number: 11977401Abstract: A power supply soft-start control method, control apparatus and control device, and a storage medium. Frequency adjustment is added to a power supply soft-start process. A voltage step value is utilized to control a voltage adjusting loop to perform voltage feedback adjustment, and a frequency step value is utilized to control a frequency adjusting loop to perform frequency feedback adjustment. In an adjusting process, according to a voltage set value of the voltage adjusting loop and a frequency set value of the frequency adjusting loop, one of the voltage adjusting loop and the frequency adjusting loop is selected to control a soft-start output voltage, whereby voltage adjustment and frequency adjustment are mutually restricted.Type: GrantFiled: January 25, 2021Date of Patent: May 7, 2024Assignee: GUANGDONG INSPUR SMART COMPUTING TECHNOLOGY CO., LTD.Inventor: Wenchao Ma
-
Patent number: 11829690Abstract: A radiation risk assessment method and device, an electronic device and a storage medium are provided. The method comprises: determining a risk signal to be assessed based on a product design model, and judging whether the risk signal to be assessed is periodic; if the risk signal to be assessed is periodic, determining the spectrum amplitude corresponding to the risk signal to be assessed; acquiring the current intensity of a radiation source and the distance between the risk signal to be assessed and the radiation source; calculating to obtain the radiation intensity of the risk signal to be assessed using the spectrum amplitude, the current intensity and the distance; and determining the corresponding radiation risk assessment result according to the radiation intensity.Type: GrantFiled: October 31, 2019Date of Patent: November 28, 2023Assignee: Guangdong Inspur Smart Computing Technology Co., Ltd.Inventor: Jie Dang
-
Patent number: 11822969Abstract: The present application discloses a task allocation method and system based on a resource management platform. The method comprises: receiving an artificial intelligence model training and/or testing task and a name of data set required for processing the task; acquiring data set distribution information of a plurality of nodes; judging if the node has the required data sets according to names of the data sets in the node; and selecting a node with the size of the required data set meeting preset requirements for task allocation according to the size of the required data set in the node if the node has the required data set.Type: GrantFiled: December 30, 2019Date of Patent: November 21, 2023Assignee: GUANGDONG INSPUR SMART COMPUTING TECHNOLOGY CO., LTD.Inventor: Dekui Wang
-
Publication number: 20230333898Abstract: A working method and device for a deep learning training task. GPUs are allocated to multiple deep learning training tasks according to the remaining resources of the GPUs in a single server node or multiple server nodes to achieve the effect of considering multiple deep learning training tasks while ensuring the utilization rate of the GPUs. The method comprises : obtaining a deep learning training task parameter input by a user, determining the type of the deep learning training task from the task parameter, the type of the deep learning training task type comprising : single model and multi-model; selecting GPUs by different policies according to different deep learning training task types; and selecting, according to the position of the GPU, a CPU having a shortest communication distance from the GPU for working.Type: ApplicationFiled: December 30, 2019Publication date: October 19, 2023Applicant: Guangdong Inspur Smart Computing Technology Co., Ltd.Inventors: Renming Zhao, Pei Chen
-
Patent number: 11681623Abstract: A pre-read data caching method and apparatus, a device, and a storage medium, the method including: receiving a read command for a target file; if determining that there is target pre-read data of the target file in a pre-read queue, then moving the pre-read data from the pre-read queue into a secondary cache queue; reading the target pre-read data in the secondary cache queue; and, after reading is complete, moving the target pre-read data from the secondary cache queue into a reset queue, the invalidation priority level of the pre-read queue being the lowest.Type: GrantFiled: January 23, 2021Date of Patent: June 20, 2023Assignee: GUANGDONG INSPUR SMART COMPUTING TECHNOLOGY CO., LTD.Inventors: Shuaiyang Wang, Wenpeng Li, Duan Zhang
-
Publication number: 20220351044Abstract: Disclosed are a method, apparatus and device for dynamically adjusting a neural network channel, and a computer-readable storage medium. The method comprises : selecting, on the basis of a preset pruning rule, initial pruned channels and initial unpruned channels of an initial deep neural network; after the initial deep neural network is trained, correcting a channel that is incorrectly pruned, and dynamically adjusting the number of network channels; cyclically executing the steps of training a network, correcting channel pruning, and determining whether the ratio of the number of current pruned channels to the total number of channels reaches a preset proportion threshold value, until the ratio of the number of current pruned channels to the total number of channels reaches the preset proportion threshold value; and obtaining a deep neural network after network channel pruning is completed, and training the deep neural network to obtain a target deep neural network.Type: ApplicationFiled: January 20, 2020Publication date: November 3, 2022Applicant: Guangdong Inspur Smart Computing Technology Co., Ltd.Inventor: Shaoyan Guo
-
Publication number: 20220334882Abstract: The present application discloses a task allocation method and system based on a resource management platform. The method comprises: receiving an artificial intelligence model training and/or testing task and a name of data set required for processing the task; acquiring data set distribution information of a plurality of nodes; judging if the node has the required data sets according to names of the data sets in the node; and selecting a node with the size of the required data set meeting preset requirements for task allocation according to the size of the required data set in the node if the node has the required data set.Type: ApplicationFiled: December 30, 2019Publication date: October 20, 2022Applicant: Guangdong Inspur Smart Computing Technology Co., Ltd.Inventor: Dekui Wang