摘要：【中文】本发明公开了一种串行测试方法以及测试工具，该测试方法包括：保存机器的环境配置并重启所述机器；在重启所述机器之后，执行测试CPU的工具；在执行所述测试CPU的工具之后，执行测试硬盘的工具；在执行所述测试硬盘的工具之后，执行测试网卡的工具；重启所述机器并执行测试内存带宽的工具；重启所述机器，并执行测试内存延迟的工具。 【EN】The invention discloses a serial test method and a test tool, wherein the test method comprises the following steps: saving an environmental configuration of a machine and restarting the machine; executing a tool to test a CPU after restarting the machine; executing a tool for testing a hard disk after executing the tool for testing the CPU; after executing the tool for testing the hard disk, executing the tool for testing the network card; restarting the machine and executing a tool for testing the memory bandwidth; the machine is restarted and the tool to test memory latency is executed.

摘要：【中文】本发明公开了一种整机柜服务器的管理系统，该管理系统包括：机柜管理单元，机柜背板，包括控制芯片，控制芯片与机柜管理单元连接，至少一个节点，至少一个节点中的每个节点与控制芯片连接；风扇模块，通过机柜背板的控制芯片连接至机柜管理单元；其中，机柜管理单元通过控制芯片获取节点的基板管理控制器和风扇模块的状态信息，并且通过控制芯片向节点的基板管理控制器和风扇模块提供控制命令。本发明的上述技术方案，实现了机柜管理单元‑控制芯片‑基板管理控制器的整机柜管理拓扑，采用控制芯片作为中继，降低了软件复杂度风险，提高了可靠性，且控制芯片成本低。 【EN】The invention discloses a management system of a whole cabinet server, which comprises: the system comprises a cabinet management unit and a cabinet back plate, wherein the cabinet back plate comprises a control chip, the control chip is connected with the cabinet management unit, and at least one node, and each node in the at least one node is connected with the control chip; the fan module is connected to the cabinet management unit through a control chip of a cabinet back plate; the cabinet management unit acquires the state information of the baseboard management controller and the fan module of the node through the control chip, and provides a control command to the baseboard management controller and the fan module of the node through the control chip. According to the technical scheme, the whole cabinet management topology of the cabinet management unit, the control chip and the substrate management controller is realized, the control chip is used as a relay, the software complexity risk is reduced, the reliability is improved, and the cost of the control chip is low.

摘要：【中文】本申请公开了一种对快照系统中数据块引用进行快速定位的定位方法，包括：步骤1，对主卷创建卷属性，根据主卷创建快照系统，并对快照的卷属性和数据块的数据块属性进行赋值；步骤2，根据数据块的数据块属性，在快照系统中确定对应的快照，生成快照集合；步骤3，根据数据块逻辑块地址，加载快照集合中快照的元数据，确定快照集合中各快照在数据块逻辑地址处、映射到的物理块地址；步骤4，当判定物理块地址等于待迁移数据块源物理块地址时，根据待迁移数据块目标物理块地址，更新对应快照与被迁移数据块之间的映射关系。通过本申请中的技术方案，实现了对快照系统中数据块引用进行快速定位，提高数据块底层数据进行数据迁移的迁移效率。 【EN】The application discloses a positioning method for quickly positioning data block reference in a snapshot system, which comprises the following steps: step 1, creating a volume attribute for a main volume, creating a snapshot system according to the main volume, and assigning values to the volume attribute of a snapshot and the data block attribute of a data block; step 2, according to the data block attribute of the data block, determining a corresponding snapshot in a snapshot system, and generating a snapshot set; step 3, loading metadata of snapshots in the snapshot set according to the logical block addresses of the data blocks, and determining the physical block addresses mapped to the snapshots in the snapshot set at the logical addresses of the data blocks; and 4, when the physical block address is judged to be equal to the source physical block address of the data block to be migrated, updating the mapping relation between the corresponding snapshot and the migrated data block according to the target physical block address of the data block to be migrated. According to the technical scheme, the data block references in the snapshot system are quickly positioned, and the migration efficiency of data migration of data block bottom layer data is improved.

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】本发明公开了一种计算机USB端口的电源切换电路，包括：状态判断单元；主电源开关单元，连接于主电源、状态判断单元和计算机的USB端口；辅助电源开关单元，连接于辅助电源、状态判断单元和USB端口；其中，状态判断单元根据计算机系统的上电顺序判断计算机系统的状态并根据状态控制主电源开关单元和辅助电源开关单元。本发明的目的至少在于，能够提供一种USB端口电路，能够在S0状态时满足大电流供电，且S3状态时提供可限制的电流能力，以保证系统维持S3状态时，系统供给CPU、内存、CPLD等关键器件的电流能力不被影响。 【EN】The invention discloses a power supply switching circuit of a computer USB port, which comprises: a state judgment unit; the main power switch unit is connected with the main power, the state judgment unit and the USB port of the computer; the auxiliary power supply switch unit is connected with the auxiliary power supply, the state judgment unit and the USB port; the state judging unit judges the state of the computer system according to the power-on sequence of the computer system and controls the main power switch unit and the auxiliary power switch unit according to the state. At least one object of the present invention is to provide a USB port circuit that can satisfy the requirement of high current power supply in S0 state and provide a current capability that can be limited in S3 state to ensure that the current capability of the system to supply critical devices such as CPU, memory, CPLD, etc. is not affected when the system maintains S3 state.

摘要：【中文】本发明公开了一种集群并行程序部署方法以及装置，该集群并行程序部署方法包括：依次对所述并行程序的运行环境的多个程序库构建镜像，以生成运行环境的镜像；将所述运行环境的镜像作为基础镜像对所述并行程序构建镜像，以生成并行程序的镜像；将生成的所述并行程序的镜像部署到所需节点上。本发明有上述技术方案，提高了HPC应用分发和部署效率，解决了传统部署方式遇到的问题，简化了部署流程和难度。 【EN】The invention discloses a cluster parallel program deployment method and a device, wherein the cluster parallel program deployment method comprises the following steps: sequentially constructing mirror images for a plurality of program libraries of the running environment of the parallel program to generate the mirror images of the running environment; taking the mirror image of the running environment as a basic mirror image to construct a mirror image for the parallel program so as to generate the mirror image of the parallel program; and deploying the generated image of the parallel program to a required node. The invention has the technical scheme, improves the distribution and deployment efficiency of the HPC application, solves the problems encountered by the traditional deployment mode, and simplifies the deployment process and difficulty.

摘要：【中文】本发明公开了一种RAID卡检测方法以及装置，该方法包括：S1，使用storcli工具第一次获取RAID卡与SAS链路之间产生的phy error的第一项计数、第二项计数和第三项计数；S2，判断第一项计数、第二项计数和第三项计数中的每一个是否都小于基准值，S3，如果判断的结果为是，则执行带宽压力测试并返回至步骤S1再次获取第一项计数、第二项计数和第三项计数；S4，根据再次获取的第一项计数、第二项计数和第三项计数判断第一次获取的第一项计数、第二项计数和第三项计数中的每一个是否都无增长，如果判断结果为都无增长，则RAID卡SAS链路的测试结果为通过。通过上述技术方案，至少能够检测RAID卡SAS链路信号质量，确保出厂的RAID卡处于健康状态。 【EN】The invention discloses a RAID card detection method and a device, wherein the method comprises the following steps: s1, using a storcli tool to obtain a first item count, a second item count and a third item count of a phy error generated between the RAID card and the SAS link for the first time; s2, judging whether each of the first item count, the second item count and the third item count is smaller than a reference value, S3, if yes, executing a bandwidth pressure test and returning to the step S1 to obtain the first item count, the second item count and the third item count again; and S4, judging whether each of the first item count, the second item count and the third item count obtained for the first time is not increased according to the first item count, the second item count and the third item count obtained again, and if the judgment result shows that the item count, the second item count and the third item count are not increased, the test result of the SAS link of the RAID card is passed. By the technical scheme, the SAS link signal quality of the RAID card can be at least detected, and the RAID card leaving factory is ensured to be in a healthy state.

摘要：【中文】本发明公开了一种基于LINUX平台的批量执行框架的实现方法及装置，该方法包括S1.基于CS架构建立框架结构；S2.框架基于IPLIST的个数，创建对应个数的线程；S3.将每个对ip的请求操作放入队列中，通过异步方式对每个队列中的操作进行状态管理；本发明方法大大降低了运维的难度，对于批量机器的操作变得轻松又有效；试验验证，传统技术中手动操作一台机器需要5min，则本发明方法操作百台机器也只需要5min，高效的节省的人力成本。 【EN】The invention discloses a method and a device for realizing a batch execution framework based on a LINUX platform, wherein the method comprises the following steps of S1, establishing a framework structure based on a CS framework; s2, the framework creates threads with corresponding number based on the number of IPLIST; s3, putting each request operation for the ip into a queue, and performing state management on the operation in each queue in an asynchronous mode; the method greatly reduces the difficulty of operation and maintenance, and is easy and effective to operate batch machines; experiments prove that in the traditional technology, 5min is needed for manually operating one machine, and the method only needs 5min for operating hundreds of machines, so that the labor cost is saved efficiently.

摘要：【中文】本发明公开了一种RAID卡老化测试方法和RAID卡老化测试方法系统，该方法包括加载测试环境，接入待测试RAID卡，并在测试环境下，进行RAID卡状态信息检查；和/或PCIe链路检测；和/或SAS链路检测；和/或RAID卡IO性能测试；和/或RAID卡日志检查；以在出厂前完成这些RAID卡检查，有效拦截硬件损伤、信号差的RAID卡，减少终端用户RAID掉盘事件的发生；在一定程度上提高产品质量，本发明还提供了用于实施上述本发明方法的系统，以完成对RAID卡出厂前的自动化检测，以确保出厂的RAID卡处于健康状态，能稳定工作；实现测试的自动化，减轻产线员工的压力；并在检测过程中对RAID卡问题原因快速诊断，有效提高Raid卡老化测试效率，提高测试结果准确性。 【EN】The invention discloses a RAID card aging test method and a system thereof, wherein the method comprises loading a test environment, accessing an RAID card to be tested, and checking the state information of the RAID card in the test environment; and/or PCIe link detection; and/or SAS link detection; and/or IO performance test of the RAID card; and/or RAID card log checking; the RAID card inspection is completed before leaving the factory, RAID cards with hardware damage and poor signals are effectively intercepted, and RAID disk failure events of terminal users are reduced; the invention also provides a system for implementing the method of the invention to complete the automatic detection of the RAID card before leaving the factory so as to ensure that the leaving RAID card is in a healthy state and can stably work; the automation of the test is realized, and the pressure of production line staff is reduced; and the reason of the RAID card problem is quickly diagnosed in the detection process, so that the aging test efficiency of the RAID card is effectively improved, and the accuracy of the test result is improved.

摘要：【中文】本发明提出了一种远程交互的方法和装置，该远程交互的方法用于集群管理系统，集群管理系统包括集群客户端和服务端，集群客户端和服务端之间建立有Websocket长连接，该方法包括：集群客户端通过Websocket长连接接收服务端发送的控制指令；集群客户端执行控制指令，得到反馈数据；集群客户端向服务端发送反馈数据。本发明通过集群客户端和服务端之间的Websocket长连接，从而服务端可直接向客户端发送操作指令、文件等数据，最终达到对集群实时管理的目的。 【EN】The invention provides a remote interaction method and a device, wherein the remote interaction method is used for a cluster management system, the cluster management system comprises a cluster client and a server, a Websocket long connection is established between the cluster client and the server, and the method comprises the following steps: the cluster client receives a control instruction sent by the server through the Websocket long connection; the cluster client executes the control instruction to obtain feedback data; and the cluster client sends feedback data to the server. According to the invention, through the Websocket long connection between the cluster client and the server, the server can directly send data such as operation instructions and files to the client, and finally the purpose of real-time cluster management is achieved.