Abstract: Disclosed in the present application are a battery charging method and apparatus, and a device and a medium. The method includes; collecting, at a preset time interval, battery charging state data of a rechargeable battery within each of the preset time intervals in real time; when the battery charging state data within any one of the preset time intervals is collected, using the battery charging state data within the preset time interval as a training data set, inputting the training data set to an initial neural network model for training, and during the process of training, updating a preset network parameter on the basis of the difference value between a model output value corresponding to each piece of battery charging state data and a preset threshold until difference value between model output value corresponding to present battery charging state data and the preset threshold meets a preset error condition.

Abstract: The present application discloses a key updating method and apparatus, a file sharing method and apparatus, a device, and a computer storage medium. The key updating method includes: acquiring a target root key and an auxiliary target root key; acquiring the identity information of a data owner of target data; acquiring a target timestamp when the target data is encrypted; acquiring a key generation parameter and an auxiliary key that are generated at a timestamp previous to the target timestamp; generating a target private key of the data owner on based on the key generation parameter and auxiliary key of the timestamp previous to the target timestamp, the identity information and the target timestamp, the key generation parameter generated at the timestamp previous to a starting timestamp being the target root key, the auxiliary key generated at the timestamp previous to the starting timestamp being the auxiliary target root key.

Abstract: Disclosed are a Power Factor Correction (PFC) control method, apparatus, and device, and a computer-readable storage medium. The method includes: acquiring an input voltage value, input current value, and output voltage value from a PFC circuit (S101); acquiring a current reference value for current loop control by using a Prony's method according to the input voltage value, the output voltage value, and a preset voltage reference value (S102); and performing current loop control according to the current reference value and the input current value, and outputting a corresponding Pulse Width Modulation (PWM) signal, so as to control a switch tube in the PFC circuit to be correspondingly switched on or switched off (S103).

Abstract: A method, apparatus, and device for dynamically modifying a PCH PCIE root port where an onboard VGA is located, and a readable storage medium. The current item may be determined at a PEI phase by reading a GPIO interface, and a target PCH PCIE root port used by the item is further determined; then configuration information is acquired, and a target onboard VGA is found and initialized; because a video bridge configuration file defined in an SDL file is saved in a structure of a temporary file when code is compiled, in the method, the value of a target member of the structure is modified as the port number of the target PCH PCIE root port at a DXE phase, so that a video bridge is switched to the target PCH PCIE root port.

Abstract: The present disclosure provides a method and device for configuring cold redundancy of a power source. The method includes determining an efficient range and an inefficient range of an efficiency-load curve of a power supply unit (PSU), wherein the efficiency-load curve is acquired by testing the PSU in advance. If the PSU operates in the efficient range when voltage of the PSU is before a preconfigured power source configuration point, or if the PSU operates in the inefficient range when the voltage of the PSU is after the preconfigured power source configuration point, the method includes recalculating the preconfigured power source configuration point. If neither, then the method includes determining that the preconfigured power source configuration point is a power source configuration point, wherein the power source configuration point is a voltage value at which a standby PSU is switched to a power supply PSU when cold redundancy is configured.

Abstract: A high-reliability multiphase power supply system and method. A second processing unit is configured with a first field-effect transistor, a drain electrode of the first field-effect transistor is connected to a power supply, a source electrode of the first field-effect transistor is connected to the drain electrode of a second field-effect transistor, the source electrode of the second field-effect transistor is connected to ground, and the gate electrodes of the first field-effect transistor and the second field-effect transistor are connected to a first processing unit; the second processing unit is configured with a first current detection module and a second current detection module, the first current detection module and the second current detection module are electrically connected to a bus unit, and the bus unit is electrically connected to a substrate management controller.

Abstract: A node mode adjustment method for when a storage cluster BBU fails and related component. The storage cluster comprises multiple IO GROUPs; each IO GROUP comprises two nodes, and each node is connected to a corresponding BBU of the node. The method comprises: monitoring the status of each BBU in the storage cluster; after detecting that any one BBU has failed, when it is determined that the BBU connected to the peer node of the node to which the failed BBU is connected has not failed, and it is determined that the storage cluster is not downgraded, maintaining in a write-back mode the node connected to the BBU detected to have failed and the peer node of said node. Applying the solution of the present application will help to ensure that the storage cluster has the advantages of high availability and high IO performance.

Abstract: The present application discloses a person re-identification method and apparatus based on a deep learning network, a device, and a medium. The method includes: obtaining an initial person re-identification network; creating a homogeneous training network corresponding to the initial person re-identification network, where the homogeneous training network comprises a plurality of homogeneous branches with a same network structure; training the homogeneous training network by using a target loss function, and determining a final weight parameter of each network layer in the homogeneous training network; and loading the final weight parameter by using the initial person re-identification network to obtain a final person re-identification network, to perform a person re-identification task by using the final person re-identification network.

Abstract: A method and apparatus for BMC-parameter configuring and taking-effect, a device and a medium.

Abstract: A communication method, system, device and computer-readable storage medium are applicable to a Windows system. The method involves: determining a media file used when communicating through a message queue; acquiring identification information of a communication sender; and adding the identification information to an access control list of the media file, such that the communication sender can access the media file, so that the communication sender can communicate through the message queue. Identification information of the communication sender is added to the access control list of the media file, increasing the probability that the communication sender successfully accesses the media file, and thereby improving the success rate of the communication sender through the message queue, and improving the communication stability of the Windows system.

Abstract: The present application discloses a file transmission method and apparatus, a device, and a storage medium. The steps of the method include: counting the number of target files transmitted in a transmission channel; obtaining a preset peak transmission rate of the transmission channel, the preset peak transmission rate being less than or equal to the rated maximum transmission rate of the transmission channel; calculating an average rate on the basis of the number of files and the preset peak transmission rate; adjusting the transmission rate of each of the target files in the transmission channel to be the average rate; and transmitting the target files in the transmission channel. The present method relatively ensures the stability of the operation of a server during file transmission. In addition, the present application also provides a file transmission apparatus, a device, and a storage medium, which have the same beneficial effects as those described above.

Abstract: A parallel bus phase correction method and a device are provided. The method comprises: correcting a data bus, respectively performing phase correction tests on a clock line; determining a first optimal window of the clock line according to the clock test results; correcting the clock line by using a median value of the first optimal window, respectively performing phase correction tests on the data bus according to the multiple second phase adjustment values, and recording corresponding data test results; determining a second optimal window of the data bus according to the data test results; and performing phase correction on normal data transmission on the basis of the median value of the first optimal window and the median value of the second optimal window. The method achieves phase correction and ensures the correctness and accuracy of data transmission, even if a small clock offset is present.

Abstract: A link detection method includes: a server-side program acquiring connection relationships between a service interface of a TOR switch and service interfaces of various servers; the server-side program periodically sending a query message to each client-side program by means of a network management switch; in response to the query message, each client-side program returning, to the server-side program and by means of the network management switch, interface information of a service interface, which is included in a TOR switch or a server at which each client-side program is located and has connection relationships with service interfaces of other devices; and the server-side program reporting, to an upper-layer program, interface information of a service interfaces of the other devices, and the received interface information, such that the upper-layer program analyzes, on the basis of the interface information, connection situation of links for realizing corresponding connection relationships.

Abstract: An FPGA cloud platform acceleration resource allocates and coordinates accelerator card resources according to delays between a host of a user and FPGA accelerator cards deployed at various network segments. Upon an FPGA usage request of the user, allocating an FPGA accelerator card in an FPGA resource pool that has a minimum delay to the host. A cloud monitoring management platform obtains_transmission delays to a virtual machine network according to different geographic locations of various FPGA cards in the FPGA resource pool, and allocating a card having a minimum delay to each user. The cloud monitoring management platform prevents unauthorized users from accessing acceleration resources in the resource pool. The invention protects FPGA accelerator cards that are not authorized for users, and ensures that the card allocated to a user has a minimum network delay, thereby optimizing acceleration performance, and improving user experience.

Abstract: An image recognition method and apparatus based on a systolic array, and a medium are disclosed. The method includes: converting obtained image feature information into a one-dimensional feature vector; converting an obtained weight matrix into a one-dimensional weight vector, and allocating a corresponding weight group to each node in a trained three-dimensional systolic array model; performing multiply-accumulate of the feature vector and a weight value on the one-dimensional feature vector in parallel by using the three-dimensional systolic array model, to obtain a feature value corresponding to each node, the feature values with different values reflecting article categories contained in an image; and determining an article category contained in the image according to the feature value corresponding to each node and a pre-established corresponding relationship between the feature value and the article category.

Abstract: A method for obtaining a cluster feature code includes: determining a plurality of key nodes from respective nodes in a cluster; obtaining plaintexts of feature codes of the respective key nodes; according to the plaintexts of the feature codes of the respective key nodes, obtaining ciphertexts of the feature codes of the respective key nodes, by utilizing a first-level public key; calculating a check code according to the ciphertexts of the feature codes of the respective key nodes; and according to the check code, obtaining the cluster feature code, by utilizing a second-level public key. By means of the present application, the scope of influence on the entire system when system nodes change is reduced.

Abstract: A quantum data erasure method, system and device, and a readable storage medium. The method includes: acquiring an equal-probability quantum state system; measuring the equal-probability quantum state system to collapse the equal-probability quantum state system into a binary random number sequence; generating a corresponding random angle value according to the binary random number sequence; and performing a bitwise rotation operation on quantum data in a quantum device according to the random angle value to complete this quantum data erasure. In the present application, the introduction of a quantum true random number can ensure that erased data will not be recovered and reversely cracked, and is of great value in protecting data assets; moreover, randomly processed data still has the characteristics such as quantum coherence and quantum entanglement, and can be used in subsequent operations, whereby a time-consuming labor-intensive process of preparing a quantum system is not required every time.

Abstract: Provided is a method for auditing an abnormality of a block device in a cloud platform. By automatically auditing abnormal data of a path layer and a multipath layer, the accuracy of multipath and path information on a host and whether a link has a fault may be quickly analyzed, operation and maintenance personnel may discover problems as soon as possible, the problem troubleshooting steps are simplified, the operation and maintenance personnel are liberated from complex work of manual troubleshooting, the troubleshooting capability of a cloud computing platform is improved, the work of the operation and maintenance personnel is simplified, and the problems are rapidly positioned.

Abstract: An nGraph-based graphics processing unit (GPU) backend distributed training method and system, a computer-readable storage medium, and an electronic device. The method includes: receiving a training request, and obtaining corresponding training data; obtaining a Nvidia® Collective multi-GPU Communication Library (NCCL) file by means of a system path of the NCCL file linked to an nGraph framework; invoking an NCCL communication interface configuration according to the training request to obtain a training model, the NCCL communication interface is an NCCL file-based communication operation interface located at a GPU backend of the nGraph framework; and performing GPU backend training on the training data using the training model. The present application can satisfy an urgent need of a user for performing neural network distributed training on the basis of an nGraph GPU backend, thus further improving the performance of deep learning network training.

Abstract: A system for accelerating an RNN network including: a first cache, which is used for outputting Wx1 to WxN or Wh1 to WhN in parallel in N paths in a cyclic switching manner, and the degree of parallelism is k; a second cache, which is used for outputting xt or ht-1 in the cyclic switching manner; a vector multiplication circuit, which is used for, by using N groups of multiplication arrays, respectively calculating Wx1xt to WxNxt, or respectively calculating Wh1ht-1 to WhNht-1; an addition circuit, which is used for calculating Wx1xt+Wh1ht-1+b1 to WxNxt+WhNht-1+bN; an activation circuit, which is used for performing an activation operation according to an output of the addition circuit; a state updating circuit, which is used for acquiring ct-1, calculating ct and ht, updating ct-1, and sending ht to the second cache; a bias data cache; a vector cache; and a cell state cache.