Abstract: The present disclosure discloses a soft-start protection circuit, comprising: a protection chip, a first switch tube, a second switch tube, a third switch tube, a fourth switch tube, a first capacitor, a first resistor, a second resistor, a third resistor, a fourth resistor, a fifth resistor, a sixth resistor and a seventh resistor. In the soft-start protection circuit of the present disclosure, the voltage of a voltage input end and the control end voltage of the first switch tube may be monitored; therefore, when there is a change in the two voltages, correct turn-on and turn-off actions may be performed in the soft-start protection circuit, increasing the accuracy of protection and at the same time reducing the turn-on speed of switch tubes to a certain extent, thus effectively reducing the impact of an impulse current on components and systems. Correspondingly, the present disclosure further discloses a circuit protection integrated chip.

Abstract: An application isolation method, system and device, and a computer-readable storage medium. The method includes: determining a target application to be isolated in Kubernetes; acquiring isolation polices of components in the target application, creating an initial network security policy corresponding to the target application; on the basis of the isolation policies, modifying a pushing rule, a popping nule and a matching label of the initial network security policy, so as to obtain a target network security policy; converting the target network security policy into an Iptables rule that matches the Kubernetes; and isolating the target application on the basis of the Iptables rule.

Abstract: An image processing method is provided. Since in the present invention, the first quantization threshold obtained by the saturated mapping method and the second quantization threshold obtained by the unsaturated mapping method are weighted, it is equivalent to fusing two quantization threshold values. The obtained optimal quantization threshold can be applied to most activations, to more effectively retain the effective information of the activations and use in subsequent image processing, thereby improving the accuracy of inference computations of the quantified deep neural network on low-bit-width hardware platforms. An image processing device and apparatus have the same beneficial effects as the above image processing method.

Abstract: A method and an apparatus for compatibility detection, a device and a non-transitory computer-readable storage medium. The method includes: a dependency form of an application system and dependency libraries of each of operating systems are acquired; compatibility matching detection is respectively performed on the dependency form and each of the dependency libraries, so as to obtain matching forms corresponding to each of the operating systems; by the matching forms, the operating systems are divided into first operating systems which are in dependency compatibility with the application system, and second operating systems which are not in dependency compatibility with the application system; it is determined that the application system is not compatible with the second operating system; and a compatibility white box test is performed on the first operating system, so as to obtain a compatibility result, which indicates whether the application system is compatible with the first operating system.

Abstract: A MON service migration method, apparatus, and device, and a readable storage medium, for use in any node in a distributed storage system. The method comprises: acquiring historical data of a MON service in a current node; in the node, determining a target magnetic disk for migrating the MON service, and migrating the historical data to the target magnetic disk; creating mount information of the MON service in a configuration file of the distributed storage system; and restarting the MON service according to the configuration file, such that the MON service migrates to the target magnetic disk. The present method does not need to remove nodes in the distributed storage system, and therefore the MON service migration process will not affect front-end services, improving the service capabilities and reliability of the distributed storage system.

Abstract: An information recording method, apparatus, and device, and a readable storage medium are provided. The method includes: when a server is started, determining a ring buffer in a Double Data Rate (DDR) of a Field-Programmable Gate Array (FPGA) acceleration card based on an OpenPower platform; determining a start address and an end address of the ring buffer and configuring the start address and the end address to the FPGA acceleration card; and during a running process of the server, recording preset debugging information to the ring buffer in real time, so as to perform fault location according to data in the ring buffer after a fault occurs in the server. According to the present application, during a running process of a server, preset debugging information is recorded using a DDR of an FPGA acceleration card.

Abstract: A protection system for abnormal power down of a child node in a rack-scale server includes a logic control module and a plugging/unplugging trigger module. The plugging/unplugging trigger module is configured to detect a contact state, at a second plugging depth, between a power supplying clamp of the child node and the power supplying copper bar through a metal probe, generate a detection signal by conversion, and send the detection signal to the logic control module; the logic control module is configured to successively turn off a main power supply and a standby power supply of the child node after determining that the power supplying clamp is disengaged from the second plugging depth of the power supplying copper bar and before the power supplying clamp is disengaged from the third plugging depth of the power supplying copper bar, so as to execute power down logic of normal power off.

Abstract: An optical circuit building method, an optical circuit, and an optical signal processing method and apparatus, the method comprising: constructing a convolution weight matrix corresponding to multiple groups of convolution weights (S11); performing singular value decomposition on the convolution weight matrix to obtain a first unitary matrix, a diagonal matrix and a second unitary matrix (S12); separately determining a first MZI structure corresponding to the first unitary matrix, a second MZI structure corresponding to the diagonal matrix and a third MZI structure corresponding to the second unitary matrix (S13); and connecting the first MZI structure, the second MZI structure and the third MZI structure to obtain an optical circuit (S14). In an optical circuit obtained by means of the foregoing, convolution calculation processing corresponding to multiple groups of convolution weights may be carried out at the same time, operation depth may be reduced, and operation efficiency may be improved.

Abstract: A data recovery method, apparatus and device, and a readable storage medium are provided. The method comprises: receiving and analyzing a data recovery request, and obtaining a total cluster data recovery amount and a total local data recovery amount; determining a basic recovery speed by means of the total cluster data recovery amount and the total local data recovery amount; performing data recovery, and compiling statistics on a proportion of recovered small files within a specified number of recovery periods; and adjusting the basic recovery speed by means of the proportion of the small files, and performing data recovery according to the adjusted basic recovery speed. In the method, a basic recovery speed can be adjusted in accordance with the small file recovery proportion, such that the overall efficiency of data recovery can be improved, and the influence on a front-end service of a user can also be reduced.

Abstract: A method, apparatus and device for deleting a distributed system file, and a storage medium, comprising: querying whether an incomplete file deletion operation exists under a sub-tree root corresponding to a certain MDS (S102); if there is an incomplete file deletion operation, continuing to delete data under the sub-tree root corresponding to the MDS, and determining whether the sub-tree root is a copy (S103); and if the sub-tree root is a copy, deleting the sub-tree root copy in a memory of the MDS (S104). By means of the steps, the number of copies of the sub-tree root in the distributed file system can be reduced, thereby reducing the number of interaction times of master copy locking between different MDSs, improving the file deletion efficiency, and improving user friendliness and differentiated competitiveness of a product.

Abstract: An out-of-band updating method and system of an Expander, applied to a controller connected to the expander by means of a communication bus. The method comprises: receiving firmware updating data sent by a user (S101); fragmenting the firmware updating data (S102); and only when determining that the expander is in an idle state, sending the fragmented data until each piece of data is sent to the Expander, such that the expander completely receiving the firmware updating data completes updating by restarting (S103). The data transmission speed is facilitated to be improved, the normal operation of a service is ensured, and the conditions of influencing state monitoring and log loss are avoided.

Abstract: A method and device for determining a cluster network card, and a computer readable storage medium. The method comprises: reading a network segment of a Cluster IP from a configuration file of a current node; determining the type of the network segment of the Cluster IP by means of an address string in the network segment of the Cluster IP; obtaining an IP and a network card corresponding to the type of the network segment of the Cluster IP on the current node; and calculating the network segment of each IP, determining the IP corresponding to the network segment with the same address string as a Cluster IP, and determining the network card where the IP is located as a cluster network card. According to the disclosure, the cluster network card can be automatically and accurately found according to the configuration file.

Abstract: The disclosure provides an airflow regulation and control apparatus. A ventilation hole is formed in a windshield in a penetrating manner; a rectangular plate is mounted on the windshield; the rectangular plate is lifted by means of power provided by a lifting wing disposed thereon, and the height of the rectangular plate when moving upwards exceed the height of an upper edge of a main body portion of the windshield; when wind strength is low, the rectangular plate is located at a lower portion under the action of gravity and shields the ventilation hole; when the wind strength is increased, the lifting wing is blown by an airflow to generate upwards lifting force, and drives the rectangular plate to moved upwardly so as to shield an opening in an upper portion of the windshield and block the airflow.

Abstract: An out-of-band updating method and system of an Expander, applied to a controller connected to the expander by means of a communication bus. The method comprises: receiving firmware updating data sent by a user (S101); fragmenting the firmware updating data (S102); and only when determining that the expander is in an idle state, sending the fragmented data until each piece of data is sent to the Expander, such that the expander completely receiving the firmware updating data completes updating by restarting (S103). The data transmission speed is facilitated to be improved, the normal operation of a service is ensured, and the conditions of influencing state monitoring and log loss are avoided.

Abstract: The present invention discloses a filtering method and system of parallel computing results, through simultaneously generating the input value of the first valid position fvp of each fragment, and simultaneously computing to obtain the output result corresponding to the input value of each first valid position fvp with the respective first valid position fvp of each fragment, and according to the output result of the first valid position fvp of the first fragment, the parallel computing results are filtered through the manner of selecting the output results of the second to the S-th fragments in sequence, to finally obtain correct parallel computing results. In the present invention, through adopting the manner of parallel filtering, the original serial filtering computation is changed to parallel computation of S fragments, the computing time is only one S-th of the original time, thereby improving the computing efficiency and satisfying the timing requirements of parallel computation.

Abstract: A load balancing method, apparatus and device for a parallel model training task, and a computer-readable storage medium, includes: acquiring data traffic and a theoretical computational amount of each network layer in a target model; determining a theoretical computing capability of each computing device, and obtaining an initial computational amount corresponding to each computing device according to the theoretical computing capability and the theoretical computational amount; performing a load balancing operation according to the initial computational amount by using multiple device critical layer position division rule, so as to obtain a plurality of initial balancing schemes; compiling statistics on time performance parameters corresponding to the initial balancing schemes, and determining an intermediate balancing scheme from the initial balancing schemes according to the time performance parameters; and adjusting the intermediate balancing scheme according to the data traffic, so as to obtain a final

Abstract: The present invention provides a method and apparatus for data caching. The method comprises: output matrixes are acquired one by one, a plurality of acquired output matrixes are written alternately into two queue sets of a first cache unit according to a sequence in which the output matrixes are acquired, and the output matrixes stored line by line in a first cache unit are written into a second cache unit one by one, according to the sequence in which the output matrixes are written into the second cache unit, valid data of each output matrix of the second cache unit is determined one by one according to preset parameters, and the valid data of each output matrix is written into a third cache unit, and the valid data of the output matrixes stored in the third cache unit are configured to be sequentially written into a memory according to a sequence in which the valid data are written into the third cache unit.

Abstract: A nonvolatile memory scheduling method, system and device, and a computer-readable storage medium. The method includes: obtaining a utilization rate of a current memory space (S101); when the utilization rate is less than a first threshold, determining a first space to be converted of the memory space, converting the first space into a high-speed storage space, and storing data in a conventional storage device into the high-speed storage space (S102); and when the utilization rate is greater than a second threshold, determining a second space to be converted of the high-speed storage space, storing data stored in the second space into the conventional storage device, and converting the second space into the memory space (S103).

Abstract: Provided is a safety control method for an AI server, which is applied in an FPGA. The method includes: obtaining a current electrical current and a current power of a GPU in the AI server according to a preset frequency; determining whether the GPU satisfies a first control privilege transfer requirement; when the GPU satisfies the first control privilege transfer requirement, taking over control privilege of a heat dissipation system from a BMC; and controlling the heat dissipation system according to the current electrical current and the current power of the GPU; wherein the first control privilege transfer requirement includes: the current electrical current of the GPU exceeds a preset electrical current, or a rate of change of the current electrical current of the GPU exceeds a preset rate of change of electrical current, or the current power of the GPU exceeds a first preset power.

Abstract: Disclosed are a method for realizing an nGraph framework supporting an FPGA backend device, and a related apparatus. The method includes: integrating an OpenCL standard API library into an nGraph framework; creating, in the nGraph framework, an FPGA backend device creation module for registering an FPGA rear-end device, initializing an OpenCL environment and acquiring the FPGA backend device; creating, in the nGraph framework, an FPGA buffer space processing module for opening up an FPGA buffer space and for reading and writing an FPGA cache; creating, in the nGraph framework, an OP kernel implementation module for creating an OP kernel and compiling the OP kernel; and creating, in the nGraph framework, an FPGA compiling execution module for registering, scheduling and executing the OP kernel.