Abstract: A method for creating a virtual machine includes receiving a virtual machine creation request comprising parameter information of a virtual network interface card occupied by a to-be-created virtual machine, obtaining current resource usage information of the network interface card resource pools of at least one computing node, wherein the at least one computing node is deployed on a cloud platform, each computing node comprises a network interface card resource pool comprising physical network interface cards, determining a target computing node, in the at least one computing node based on the parameter information and the current resource usage information, and invoking the target computing node to create the virtual machine.

Abstract: A local apparatus is configured to perform resolution on a conflict field generated by code files of a plurality of versions, and send a conflict resolution result to the service apparatus. The conflict field includes at least one conflict block, and the conflict resolution result includes at least one of a resolution result of a local resolvable conflict block and an identifier of a local irresolvable conflict block. The remote apparatus is configured to obtain the conflict resolution result from the service apparatus, generate a collaborative processing window based on the conflict resolution result, and receive a result of processing the conflict resolution result by a remote user based on the collaborative processing window.

Abstract: A deep learning job scheduling method includes obtaining a job request of a deep learning job, determining a target job description file template from a plurality of pre-stored job description file templates based on the job request, determining an identifier of a target job basic image from identifiers of a plurality of pre-stored job basic images based on the job request, generating a target job description file based on the target job description file template and the identifier of the target job basic image, sending the target job description file to a container scheduler, and selecting the target job basic image from the pre-stored job base images based on the target job description file, and creating at least one container for executing the job request.

Abstract: In a code processing method, a computing device determines a source branch and a target branch of a software development project, where the source branch includes a plurality of commit codes, and the plurality of commit codes include a first commit code and a second commit code. The computing device presents the first commit code and the second commit code associated with the first commit code to a user, and performs code merging on the first commit code, some or all of second commit codes, and a commit code in the target branch according to user input.

Abstract: In a fault image generation method, a processing device obtains a non-fault image and a first fault image, where the non-fault image records a first object that is not faulty, the first fault image records a second object that is faulty, and a type of the first object is different from a type of the second object. The processing device then migrates a fault pattern of the second object in the first fault image to the first object in the non-fault image, to obtain a second fault image, where the second fault image presents the first object in a faulty state.

Abstract: A node provides a service to a client node in a network. The node is configured to execute a code for providing the service to the client node in an enclave of a trusted execution environment (TEE) and to execute a code library in the enclave to attest to the client node the identity of the service provided. The service provided to the client node may be a distributed service including a result of a cooperation of a plurality of neighbor nodes, which are connected to the node either directly or through other intermediate nodes. The code library is configured to attest to the client node the identity of the distributed service.

Abstract: After obtaining a virtual machine migration message, a control node performs first network configuration on a virtual switch of a new virtual machine. After performing the first network configuration, the control node creates the new virtual machine, and performs second network configuration on a second virtual switch in a virtual network in which a to-be-migrated virtual machine is located. In the present application, before the new virtual machine is created, a network connection of the to-be-migrated virtual machine is maintained, and a part of network configuration for the new virtual machine is completed.

Abstract: A bandwidth scheduling method includes receiving a bandwidth request sent by a data center, where the bandwidth request includes bandwidth required to transmit non-real-time traffic; allocating, based on historical bandwidth information, the bandwidth required by the data center to transmit the non-real-time traffic in a future time period, where the historical bandwidth information is used to predict occupation of total bandwidth by the data center in a region in which the data center is located at each moment in the future time period; and sending a bandwidth response to the data center, where the bandwidth response includes an allocation result.

Abstract: An apparatus is disclosed. The apparatus may be implemented in a database node or a storage node and includes one or more processors and memory storing instructions for causing the processor to perform a number of operations. Responsive to a page write request, the apparatus determines an identifier corresponding to the requested page, sends component blocks corresponding to the page to a storage node for appending to an append-only storage log of the storage node, receives the physical location of the stored component blocks, and associates the physical storage location of the stored component blocks with the logical identifier corresponding to the page.

Abstract: A data migration method and an apparatus are provided. The method is as follows: sending, by a first storage system, a location update request to a location server, where the location update request is used to indicate the location server to update location information of a first bucket from being located in a second storage system to being located in the first storage system; migrating data in a first bucket from the second storage system; receiving a data access request, where the data access request is used to access the data in the first bucket; and determining based on a type of the data access request and a migration status of the data, that the first storage system or the second storage system processes the data access request.

Abstract: This application provides a data backup method. The method includes: obtaining, by a first node, an identifier of a backup execution node from a storage device; and backing up data of the first node in the storage device responsive to determining that an ID of the first node is the same as the ID of the backup execution node. The embodiments of this application can improve reliability of the data stored by the first node, and prevent a plurality of nodes from redundantly backing up duplicate data in the storage device. The techniques disclosed herein reduce resource consumption.

Abstract: There is provided a method and apparatus for vectorized resource scheduling using multi-dimensional hierarchical arrays with a uniform data type. According to embodiments, the hierarchy of resources may be represented as a multi-dimensional array of a uniform data type in a scheduling context. This data structure is known as a tensor. In various embodiments, a tensor may be used to model each of resource usage, requested resources, resource capabilities, tenant resource usage, etc. There is provided tensor scheduling operations (e.g. tensor scheduling algorithms) to make use of this resource model. Some embodiments may exploit specialized hardware, such as GPUs and tensor-processing units, in order to accelerate the tensor scheduling operations.

Abstract: In a knowledge obtaining method, a computing device obtains pieces of first knowledge from a model knowledge base based on a parameter, where the first knowledge is regarding a first machine-learning model. The model knowledge base stores data regarding attributes of multiple machine-learning models, and the data are organized in a searchable manner based the parameter, which includes one or a combination of: knowledge in a machine learning task, an attribute of the machine learning task, and knowledge between a plurality of machine learning tasks. The computing device then provides the pieces of the first knowledge to a user.

Abstract: A stripe reassembling technique includes a stripe server that selects stripes; uses data chunks including valid data in the stripes S1, S2, and S3 as data chunks in a new stripe S4, and generates data of a parity chunk for data of the data chunks in S4 according to an erasure coding (EC) algorithm the same as that of S1, S2, and S3 ; and stores the parity data of the parity chunk on a parity storage node.

Abstract: A method for starting a vehicle and a related device are provided. The method includes: a server sends first security information and second security information to a mobile terminal and a telematics box respectively through an encrypted secure transmission channel or an encrypted secure transmission protocol, so that the telematics box can verify, by using the first security information and the second security information, a vehicle start instruction sent by the mobile terminal. When the verification succeeds, the telematics box controls the vehicle to start. According to the application, plaintext transmission of the security information between the mobile terminal and the telematics box is avoided, thereby avoiding leakage of the security information, and reducing a risk that the vehicle is maliciously started. In addition, a transmission process of the security information may not be restricted by a distance between the vehicle and the mobile terminal.

Abstract: In a service providing method, a service providing apparatus receives a first service request from a tenant, and the first service request includes an identifier of a tenant. The service providing apparatus adds a first service instance to a target serving cell associated with the tenant, where the first service instance is for processing the first service request.

Abstract: An edge system includes a plurality of first-layer circles and a second-layer circle, and a geographical area represented by the second-layer circle includes a geographical area represented by each first-layer circle, that is, the first-layer circles are deployed in the second-layer circle. Each first-layer circle includes a plurality of sites, the sites include an initial site and are also included in the second-layer circle, and the initial site processes a data operation request sent by a client in a first-layer circle including the initial site, and in the second-layer circle.

Abstract: In a message processing method, a message stream platform receives a plurality of log messages from a production platform, and the plurality of log messages are used to record information about a first service executed by the production platform. The message stream platform stores the plurality of log messages in a log file based on time segments. Then, the message stream platform may send messages in a same time segment in the plurality of log files to a consumption platform based on the time segment mark. According to the message processing method, the messages in the same time segment in the plurality of log files may be sent to the consumption platform to ensure that the consumption platform obtains the log messages generated in the same time segment.

Abstract: A vehicle-to-everything terminal provides a vehicle-to-everything server with a security credential that can prove an identity of the vehicle-to-everything terminal, and requests the vehicle-to-everything server to apply for a certificate for the vehicle-to-everything terminal. The security credential may be a token preconfigured in the vehicle-to-everything terminal, or may be a digital signature of the vehicle. The vehicle-to-everything server performs identity verification on the vehicle-to-everything terminal based on the security credential. After the verification succeeds, the vehicle-to-everything server selects a proper certificate server to apply for a certificate for the vehicle-to-everything terminal.

Abstract: An ontology management method includes receiving, by a machine-to-machine (M2M) platform, an ontology mapping request sent by a first application entity, where the ontology mapping request carries a first ontology identifier, a second ontology identifier, and mapping method description information, determining, by the M2M platform, a first ontology and a second ontology based on the first ontology identifier and the second ontology identifier, determining, by the M2M platform, a mapping manner based on the mapping method description information, and creating mapping between the first ontology and the second ontology according to the determined mapping manner, and storing, by the M2M platform, the mapping between the first ontology and the second ontology in an ontology mapping resource, and returning an identifier of the ontology mapping resource to the first application entity.