摘要：【中文】本申请提供一种基于混合内存的数据处理方法及装置。该方法应用于分布式存储系统中的一个节点，分布式存储系统包括多个互相通信连接的节点，每个节点包括HFDD和外存储器，HFDD包括内存和固态硬盘SSD，内存包括随机存储器RAM和NVDIMM，所述方法包括：计算各数据的热度，其中，热度表示对应数据被访问的频繁程度；根据各数据的热度，以及内存、SSD和外存储器分别对应的存储容量将各数据进行存储。本申请实施例中，HFDD是基于RAM+NVM混合内存的容错分布式数据抽象，并且根据数据的热度对数据进行存储，一方面提高了内存的存储容量，另一方面，提高了数据访问的效率。 【EN】The application provides a data processing method and device based on a hybrid memory. The method is applied to a node in a distributed storage system, the distributed storage system comprises a plurality of nodes which are mutually communicated and connected, each node comprises an HFDD and an external storage, the HFDD comprises an internal storage and a solid-state disk SSD, and the internal storage comprises a random access memory RAM and an NVDIMM, and the method comprises the following steps: calculating the heat of each data, wherein the heat represents the frequency of accessing the corresponding data; and storing each data according to the heat of each data and the storage capacity corresponding to the internal memory, the SSD and the external memory respectively. In the embodiment of the application, the HFDD is a fault-tolerant distributed data abstraction based on the RAM + NVM hybrid memory, and the data is stored according to the heat of the data, so that on one hand, the storage capacity of the memory is improved, and on the other hand, the data access efficiency is improved.

摘要：【中文】本申请提供一种跨站点存储系统及数据信息访问方法。该系统包括多个站点，每个站点包括存储单元、存储集控制器、存储站点控制器和接入节点；站点之间通过存储站点控制器进行通信连接；各站点的存储单元、存储集控制器和存储站点控制器通信连接；各站点的接入节点分别与存储单元、存储集控制器和存储站点控制器通信连接；存储单元用于存储数据信息；存储集控制器用于控制存储单元对数据信息的存储；存储站点控制器用于接收客户端的访问请求，并从一个或多个站点中获取对应的数据信息；接入节点用于为客户端提供接入系统的接口。本申请提供的跨站点存储系统通过存储站点控制器将各站点进行通信连接，构成了能够存储并管理千亿级数据信息的存储系统。 【EN】The application provides a cross-site storage system and a data information access method. The system comprises a plurality of sites, wherein each site comprises a storage unit, a storage set controller, a storage site controller and an access node; the stations are in communication connection through a storage station controller; the storage unit, the storage set controller and the storage site controller of each site are in communication connection; the access node of each site is respectively in communication connection with the storage unit, the storage set controller and the storage site controller; the storage unit is used for storing data information; the storage set controller is used for controlling the storage of the data information by the storage unit; the storage site controller is used for receiving an access request of a client and acquiring corresponding data information from one or more sites; the access node is used for providing an interface for the client to access the system. The cross-site storage system provided by the application is used for performing communication connection on all sites through the storage site controller, so that the storage system capable of storing and managing billions of levels of data information is formed.

摘要：【中文】本申请涉及高性能计算技术领域，提供一种GPU资源使用方法、装置及存储介质。其中，GPU资源使用方法包括：容器管理系统根据客户端提交的计算请求创建调度任务；容器管理系统判断集群中空闲的GPU资源是否满足创建GPU容器的指令中指定的资源需求；若不满足需求，则容器管理系统先挂起调度任务直至满足需求时再执行；在调度任务被挂起时，容器管理系统不进行GPU容器的创建。在该方法中，容器管理系统会根据GPU资源的使用状况对调度任务进行排队处理，从而使得对于每个计算任务，都可以独占式使用GPU资源，被其使用的GPU资源不与其他计算任务共享，因此其执行进度和计算结果都可以按计划进行，不会受到其他计算任务的影响。 【EN】The application relates to the technical field of high-performance computing, and provides a GPU resource using method, a GPU resource using device and a storage medium. The GPU resource using method comprises the following steps: the container management system establishes a scheduling task according to a calculation request submitted by a client; the container management system judges whether idle GPU resources in the cluster meet the resource requirements specified in the instruction for creating the GPU container; if the requirements are not met, the container management system suspends the scheduling task first until the requirements are met and then executes the scheduling task; when the scheduling task is suspended, the container management system does not perform the creation of the GPU container. In the method, the container management system queues the scheduling tasks according to the use condition of the GPU resources, so that each computing task can exclusively use the GPU resources, and the used GPU resources are not shared with other computing tasks, so that the execution progress and the computing result can be performed according to a plan without being influenced by other computing tasks.

摘要：【中文】本申请提供一种信息选择方法、装置、电子设备及可读存储介质，包括：用于从多个用户中为目标用户选出可推荐用户，方法包括：计算多个用户中每个用户分别在多个话题的活跃度；根据多个用户中每个用户分别在多个话题的活跃度，计算多个话题中两两话题之间的相似度；根据预先设置的关联路径类型、用户在话题的活跃度以及两两话题之间的相似度，获取多个用户中的每个待筛选用户与目标用户的全部关联路径；对于每个待筛选用户，利用随机游走算法计算全部关联路径中每条关联路径的关联概率，获取全部关联路径中每条关联路径的关联概率的加和；根据加和计算待筛选用户与目标用户的可推荐概率。与现有技术相比，本申请提高了链接预测的准确性。 【EN】The application provides an information selection method, an information selection device, an electronic device and a readable storage medium, wherein the information selection method comprises the following steps: for selecting a recommendable user for a target user from a plurality of users, a method comprising: calculating the activity of each user in the plurality of users on a plurality of topics respectively; calculating the similarity between every two topics in the multiple topics according to the activity of each user in the multiple users on the multiple topics; acquiring all association paths of each user to be screened and a target user in a plurality of users according to preset association path types, the activeness of the users on topics and the similarity between every two topics; for each user to be screened, calculating the association probability of each association path in all association paths by using a random walk algorithm, and acquiring the sum of the association probabilities of each association path in all association paths; and calculating the recommendable probability of the user to be screened and the target user according to the sum. Compared with the prior art, the method and the device improve the accuracy of the link prediction.

摘要：【中文】本申请实施例提供一种频繁项集的挖掘方法、装置、存储介质和电子设备，该挖掘方法包括：获取原始数据集；将原始数据集划分为多个子数据集；计算多个子数据集中每个子数据集的至少一个频繁项集；将所有子数据集的所有频繁项集进行汇总，得到原始数据集的至少一个候选频繁项集；从至少一个候选频繁项集中筛选出原始数据集的频繁项集。本申请实施例通过上述方法，能够减少频繁项集的挖掘时间。 【EN】The embodiment of the application provides a frequent item set mining method, a frequent item set mining device, a storage medium and electronic equipment, wherein the mining method comprises the following steps: acquiring an original data set; dividing an original data set into a plurality of subdata sets; calculating at least one frequent item set of each sub data set in a plurality of sub data sets; summarizing all frequent item sets of all the sub data sets to obtain at least one candidate frequent item set of the original data set; and screening out the frequent item set of the original data set from the at least one candidate frequent item set. By the method, the mining time of the frequent item set can be reduced.

摘要：【中文】本申请提供一种数据的查询方法、装置及存储介质。方法包括：接收用户发送的携带有数据索引的第一数据查询请求；将数据索引的查询范围缩小到用户预设的数据查询范围以内，获得第二数据查询请求；利用第二数据查询请求，在数据系统中查询对应的数据。通过个性化的设置，使得用户拥有个性化的数据查询范围。在进行数据查询时，则将用户请求查询的范围限缩到自己个性化的数据查询范围以内，以便用户只能在自己个性化的数据查询范围内查询自己想要数据，即确保了社区版的数据的安全，又确保了用户可以个性化使用。 【EN】The application provides a data query method, a data query device and a storage medium. The method comprises the following steps: receiving a first data query request which is sent by a user and carries a data index; narrowing the query range of the data index to be within a data query range preset by a user to obtain a second data query request; and querying corresponding data in the data system by using the second data query request. Through personalized setting, the user can have a personalized data query range. When data query is carried out, the query range requested by the user is limited to the personalized data query range, so that the user can query the data desired by the user only in the personalized data query range, the safety of the community version data is ensured, and the personalized use of the user is also ensured.

摘要：【中文】本发明公开了一种基于全卷积神经网络的图像裂缝分割方法，大致包括数据集的构建、模型训练和结果预测三个阶段，利用训练好的全卷积深度学习模型对待检测的图像进行语义分割，完成对图像中裂缝区域的特征提取，识别图像中的裂缝。本发明方法中全卷积深度学习模型的设计构思为：以VGG16网络为基础，将其最后的全连接层替换为卷积层，使用上采样操作将低分辨率的特征图恢复为高分辨率的特征图，并利用跳级结构融合多层深度特征,以提高裂缝分割精度。本发明方法节省人力成本，并排除人为主观因素的干扰，具有效率高，准确率高，实用性强的优点。 【EN】The invention discloses an image crack segmentation method based on a full convolution neural network, which roughly comprises three stages of data set construction, model training and result prediction, and the trained full convolution deep learning model is utilized to carry out semantic segmentation on an image to be detected, so that the feature extraction of a crack region in the image is completed, and cracks in the image are identified. The design concept of the full convolution deep learning model in the method is as follows: based on the VGG16 network, the last full connection layer is replaced by a convolutional layer, the low-resolution feature map is restored to be a high-resolution feature map by using an up-sampling operation, and a multi-layer depth feature is fused by using a skip level structure so as to improve the fracture segmentation precision. The method saves labor cost, eliminates interference of artificial subjective factors, and has the advantages of high efficiency, high accuracy and strong practicability.

摘要：【中文】本申请涉及大数据技术领域，提供一种大数据应用优化方法、装置及存储介质。其中，大数据应用优化方法应用于Hadoop集群中的主节点，包括：获取Hadoop集群中的至少一个服务节点发送的预设信息，根据预设信息对每个服务节点上的NUMA节点进行打分，其中，得分最高的至少一个NUMA节点为目标NUMA节点；向目标NUMA节点所在的目标服务节点发送绑定信息，绑定信息用于指示目标服务节点执行目标NUMA节点与大数据应用之间的绑定操作。绑定后，大数据应用能够固定在目标NUMA节点上执行，从而有利于提高大数据应用的吞吐量，实现大数据应用针对NUMA架构的性能优化。 【EN】The application relates to the technical field of big data, and provides a big data application optimization method, a big data application optimization device and a storage medium. The big data application optimization method is applied to a main node in a Hadoop cluster and comprises the following steps: obtaining preset information sent by at least one service node in a Hadoop cluster, and scoring NUMA nodes on each service node according to the preset information, wherein at least one NUMA node with the highest score is a target NUMA node; and sending binding information to a target service node where the target NUMA node is located, wherein the binding information is used for indicating the target service node to execute the binding operation between the target NUMA node and the big data application. After binding, the big data application can be fixed on a target NUMA node to be executed, so that the throughput of the big data application is improved, and the performance optimization of the big data application for the NUMA architecture is realized.

摘要：【中文】本发明公开了一种富含GABA和花青素的芽菜的培育方法及其发酵乳饮料的制备方法，特点是培育方法包括步骤将紫花菜种子置于次氯酸钠溶液中浸泡用纯水冲洗，于含有2.7‑10.8 g/L葡萄糖和0.5‑1.5 g/L费菜总黄酮的混合溶液中浸泡沥干，然后于12/12h光照暗处理后，每天更换浸泡液，清洗芽菜沥干冷冻干燥粉碎，即得富含GABA和花青素的芽菜；发酵乳饮料制备方法包括以下步骤：将富含GABA和花青素的芽菜粉末与由小麦麦粒和玉米碎粒制备的含糖乳饮料等体积混合煮沸，超声波处理后过滤，取滤液加入植物乳杆菌菌悬液，装罐保温发酵后熟得富含GABA和花青素的发酵乳饮料，优点是γ‑氨基丁酸和花青素含量高。 【EN】The invention discloses a method for cultivating sprouts rich in GABA and anthocyanin and a method for preparing fermented milk beverage thereof, which are characterized in that the cultivating method comprises the steps of soaking Chinese violet seeds in sodium hypochlorite solution, washing the Chinese violet seeds with pure water, soaking and draining the Chinese violet seeds in a mixed solution containing 2.7-10.8g/L glucose and 0.5-1.5 g/L sedum aizoon total flavonoids, then replacing soaking liquid every day after 12/12h dark light treatment, cleaning the sprouts, draining, freeze drying and crushing the sprouts to obtain the sprouts rich in GABA and anthocyanin; the preparation method of the fermented milk beverage comprises the following steps: the bud vegetable powder rich in GABA and anthocyanin and the sugar-containing milk beverage prepared from wheat kernels and corn nibs are mixed in equal volume and boiled, filtered after ultrasonic treatment, the filtrate is added with lactobacillus plantarum suspension, and the mixture is canned, kept warm, fermented and ripened to obtain the fermented milk beverage rich in GABA and anthocyanin.

摘要：【中文】本申请提供一种日志收集方法、系统、节点、电子设备及可读存储介质，涉及计算机技术领域。该方法包括：通过日志收集组件采集所述节点中运行的至少一个pod的日志存储路径下存储的日志文件，其中，在所述日志存储路径下存储有每个pod中的每个容器产生的日志文件；通过所述日志收集组件将采集的所述日志文件传输至日志存储系统。该方案中，通过在节点中部署日志收集组件，使得可以通过日志收集组件在日志存储路径下收集其存储的节点内的所有容器产生的日志文件，然后将日志文件传输至日志存储系统，由此可轻松实现对节点内的每个pod内的容器产生的日志文件的收集。 【EN】The application provides a log collection method, a log collection system, a node, electronic equipment and a readable storage medium, and relates to the technical field of computers. The method comprises the following steps: collecting log files stored under a log storage path of at least one pod running in the node through a log collection component, wherein the log files generated by each container in each pod are stored under the log storage path; and transmitting the collected log file to a log storage system through the log collection component. In the scheme, by deploying the log collection component in the node, the log files generated by all containers in the node stored by the log collection component can be collected under the log storage path through the log collection component, and then the log files are transmitted to the log storage system, so that the collection of the log files generated by the containers in each pod in the node can be easily realized.