WO2021120843A1 - 云主机内存分配方法及云主机、设备及存储介质 - Google Patents

云主机内存分配方法及云主机、设备及存储介质 Download PDF

Info

Publication number
WO2021120843A1
WO2021120843A1 PCT/CN2020/123237 CN2020123237W WO2021120843A1 WO 2021120843 A1 WO2021120843 A1 WO 2021120843A1 CN 2020123237 W CN2020123237 W CN 2020123237W WO 2021120843 A1 WO2021120843 A1 WO 2021120843A1
Authority
WO
WIPO (PCT)
Prior art keywords
memory
cloud host
capacity
user
memory capacity
Prior art date
Application number
PCT/CN2020/123237
Other languages
English (en)
French (fr)
Inventor
王鹏
Original Assignee
平安科技(深圳)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 平安科技(深圳)有限公司 filed Critical 平安科技(深圳)有限公司
Publication of WO2021120843A1 publication Critical patent/WO2021120843A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/50Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5011Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
    • G06F9/5016Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals the resource being the memory

Definitions

  • This application relates to the field of cloud computing technology, and in particular to a cloud host memory allocation method, cloud host, equipment, and storage medium.
  • Cloud services are the increase, use, and interaction modes of Internet-based related services, which usually involve the provision of dynamic, easily expandable and often virtualized resources through the Internet.
  • Cloud service refers to obtaining required services through the network in an on-demand and easily scalable manner.
  • memory is an indispensable part of cloud hosts, and memory communicates with the CPU Bridge, all programs are run in the memory, so the performance of the memory has a great impact on the cloud host.
  • DRAM memory Dynamic Random Access Memory (Dynamic Random Access Memory)
  • DRAM memory has the advantages of very low latency, usually nanoseconds, and ample bandwidth.
  • the inventor realized that DRAM memory is data volatile in abnormal scenarios such as power failure, so Data reliability is low, and the cost is high. Therefore, how to configure memory with fast reading speed, high reliability, low cost, and large capacity on the cloud host is an urgent need of users.
  • This application provides a cloud host memory allocation method, cloud host, equipment, and storage medium, which can achieve the purpose of improving the reliability of the cloud host memory while ensuring the reading speed, reducing the cost, and increasing the memory capacity.
  • a technical solution adopted in this application is to provide a cloud host memory allocation method, and the cloud host memory allocation method includes the following steps:
  • the storage space of the corresponding capacity in the non-volatile memory on the preset server is allocated to the cloud host as the memory of the cloud host, wherein the non-volatile memory on the server adopts 3D Xpoint storage technology memory.
  • the configuration instruction includes the number of CPUs and memory capacity configuration values of the cloud host requested by the user, and the memory capacity of the cloud host requested by the user is acquired according to the configuration instruction, include:
  • the memory capacity configuration value is empty, the memory capacity of the cloud host requested by the user is obtained according to the number of CPUs and the preset memory capacity corresponding to a single CPU.
  • the cloud host memory allocation method further includes the following steps:
  • the storage space of the corresponding capacity in the dynamic random access memory on the preset server is allocated to the cloud host as a cache of the cloud host.
  • the non-volatile memory includes multiple storage modes
  • the storage mode includes a memory mode
  • the cloud host memory allocation method further includes:
  • the storage mode of the non-volatile memory is set to the memory mode.
  • the preset memory capacity corresponding to the single CPU is 8G.
  • a technical solution adopted by this application is to provide a cloud host, the cloud host includes a memory, the memory uses a non-volatile memory, and the non-volatile memory uses a 3D Xpoint storage technology memory.
  • the cloud host further includes a processor and a cache
  • the cache adopts a dynamic random access memory, the cache is used to store first data, and the memory is used to store second data;
  • the processor is configured to find target data from the first data in the cache, and when the processor fails to find the target data from the first data, then retrieve the second data from the memory
  • the target data is searched in, and the target data is the data that needs to be called during the operation of the cloud host.
  • the cloud host further includes a memory manager, and when the processor fails to find the target data from the first data, the processor controls the memory manager from Searching for the target data in the second data in the memory.
  • a technical solution adopted in this application is to provide a cloud host memory allocation device, the cloud host memory allocation device including a processor and a memory coupled to the processor, wherein:
  • the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the following steps:
  • the storage space of the corresponding capacity in the non-volatile memory on the preset server is allocated to the cloud host as the memory of the cloud host, wherein the non-volatile memory on the server adopts 3D Xpoint storage technology memory.
  • a technical solution adopted by this application is to provide a storage medium storing computer-readable instructions.
  • the computer-readable instructions are executed by one or more processors, one or Multiple processors perform the following steps:
  • the storage space of the corresponding capacity in the non-volatile memory on the preset server is allocated to the cloud host as the memory of the cloud host, wherein the non-volatile memory on the server adopts 3D Xpoint storage technology memory.
  • the cloud host memory allocation method and cloud host, equipment and storage medium proposed in this application use the memory of 3D Xpoint storage technology as the non-volatile memory on the server and allocate it to the user as the cloud host memory according to user configuration instructions.
  • the user provides a cost-effective cloud host with high memory reliability, fast reading speed, and low cost.
  • FIG. 1 is a schematic flowchart of a cloud host memory allocation method according to an embodiment of the present application
  • FIG. 2 is a schematic structural diagram of a cloud host according to an embodiment of the present application.
  • FIG. 3 is a schematic diagram of the principle of searching for target data by the cloud host described in FIG. 1;
  • FIG. 4 is a schematic structural diagram of a cloud host memory allocation device according to an embodiment of the present application.
  • FIG. 5 is a schematic structural diagram of a storage medium according to an embodiment of the present application.
  • FIG. 1 is a schematic flowchart of a method for allocating a memory of a cloud host according to an embodiment of the present application.
  • the cloud host memory allocation method can be run on a local server, a remote server, or a remote cloud platform, and the local server, the remote server, or the remote cloud platform can communicate with users through a wireless network such as WIFI or a wired network such as Ethernet.
  • the user terminal may be any terminal that can be operated by the user, such as a mobile phone, a tablet computer, a notebook, and the like. It should be noted that if there is substantially the same result, the method of the present application is not limited to the sequence of the process shown in FIG. 1. As shown in Figure 1, the method includes steps:
  • Step S101 Obtain a configuration instruction for a user to apply for creating a cloud host.
  • the user can send a configuration instruction for creating a cloud host to the server through input devices such as mouse, keyboard, etc., using script commands, graphical interfaces, etc., on the user side.
  • the configuration instruction may include the configuration instruction requested by the user. The number of CPUs and memory capacity configuration values of the cloud host.
  • Step S102 Acquire the memory capacity of the cloud host requested by the user according to the configuration instruction.
  • a configuration interface can be provided to the user, so that the user can input the configuration instruction specification through the configuration interface, including the configuration value of the memory capacity of the cloud host, and the server can The memory capacity of the cloud host requested by the user can be obtained according to the configuration instruction.
  • the user can input the configuration according to the user's needs through the provided configuration interface.
  • the memory capacity configuration value may also not input the designated memory capacity configuration value, so that the memory capacity configuration value is empty.
  • the memory capacity configuration value is used as the memory capacity of the cloud host requested by the user; if the memory capacity configuration value is empty, it can be The number of CPUs and the preset memory capacity corresponding to a single CPU obtain the memory capacity of the cloud host requested by the user. Among them, the preset memory capacity corresponding to a single cpu can be manually set, and a reasonable preset memory capacity can be set according to the user's application requirements.
  • the preset memory capacity corresponding to the single cpu is 8G, that is, when the user passes When the configuration instruction specifies that the number of CPUs of the cloud host is 1, if the memory capacity configuration value in the configuration instruction entered by the user is empty, the memory capacity is considered to be 8G, and when the user passes the When the configuration instruction specifies that the number of CPUs of the cloud host is 2, if the memory capacity configuration value in the configuration instruction input by the user is empty, the memory capacity is considered to be 16G, and so on.
  • Step S103 According to the memory capacity, the storage space of the corresponding capacity in the non-volatile memory on the preset server is allocated to the cloud host as the memory of the cloud host, wherein the non-volatile memory on the server To adopt 3D Xpoint storage technology memory.
  • a non-volatile memory is installed on the server.
  • the non-volatile memory can also be installed on other storage devices and communicated with the server.
  • the server is based on the configuration instruction
  • the acquired memory capacity or the memory capacity calculated from the configuration instruction allocates the storage space of the corresponding capacity in the non-volatile memory on the preset server to the cloud host as the cloud host Memory.
  • the non-volatile memory on the server is a memory using 3D Xpoint storage technology.
  • the 3D The memory of Xpoint storage technology can be Optane non-volatile memory DCPM (optane DC (data center) persistent memory).
  • the Optane non-volatile memory DCPM can provide enough in scenarios that require a large amount of memory such as virtualization. Due to its non-volatility, the loading time of data and processes can be avoided after the server restarts, and the Optane non-volatile memory DCPM is more expensive than dynamic random access memory DRAM memory.
  • the Optane non-volatile memory DCPM can provide relatively large storage capacity while providing read and write speeds and latency close to that of dynamic random access memory DRAM, which is traditional SSD is incomparable.
  • the cloud host memory allocation method of this embodiment uses the memory of 3D Xpoint storage technology as the non-volatile memory on the server, and allocates it to users as cloud host memory according to user configuration instructions, which can provide users with a high memory reliability. , A cost-effective cloud host with fast reading speed and low cost.
  • the cloud host memory can be allocated to the user according to the number of CPUs and the preset memory capacity corresponding to a single cpu, which ensures the performance of the cloud host and improves the user experience.
  • the cloud host memory allocation method further includes the following steps:
  • Step S104 Calculate the cache capacity required by the cloud host according to the memory capacity and according to a preset ratio.
  • a dynamic random access memory DRAM is provided on the server, and the non-volatile memory includes multiple storage modes.
  • the server is not The volatile memory is set to a memory mode (Memory Mode), and the memory mode (Memory Mode) is mainly applied to a scene that requires a large-capacity memory.
  • the memory mode (Memory Mode)
  • the Optane non-volatile memory DCPM is recognized by the cloud host system as the dynamic random access memory DRAM memory, and the actual dynamic random access memory DRAM memory can be used as the Optane non-volatile memory
  • the volatile memory DCPM cache The CPU of the cloud host uses a dynamic random access memory DRAM memory as a cache, and the Optane non-volatile memory DCPM is used as a memory.
  • the cache capacity required by the cloud host may be calculated according to the memory capacity and according to a preset ratio, and the preset ratio may be artificially set according to the requirements of the application on the cloud host.
  • Step S105 According to the cache capacity, the storage space of the corresponding capacity in the dynamic random access memory on the preset server is allocated to the cloud host as a cache of the cloud host.
  • step S105 after calculating the cache capacity, the storage space of the corresponding capacity in the dynamic random access memory DRAM on the server may be allocated to the cloud host as the cache of the cloud host, so that the When the CPU of the cloud host reads data, the CPU of the cloud host may first search for the data from the dynamic random access memory DRAM. When the CPU of the cloud host reads the data from the dynamic random access memory DRAM, After reading the data fails, the data is searched from the non-volatile memory.
  • the memory allocation method of a cloud host in this embodiment calculates the cache capacity required by the cloud host according to a preset ratio according to the memory capacity and allocates the cache to the cloud host, thereby improving the processing performance of the cloud host and further improving users Experience.
  • the CPU of the cloud host may first search for the data from the dynamic random access memory DRAM, and when the CPU of the cloud host fails to read the data from the dynamic random access memory DRAM , Searching for the data from the non-volatile memory can increase the computing speed of the cloud host and improve user experience.
  • FIG. 2 is a schematic structural diagram of a cloud host 1 according to an embodiment of the present application
  • FIG. 3 is a schematic diagram of a principle of searching for target data by the cloud host 1 in FIG. 1.
  • the application also provides a cloud host 1, which includes a memory 10 and a processor 11.
  • the memory 10 is a non-volatile memory, and the non-volatile memory is a memory using 3D Xpoint storage technology.
  • the memory using 3D Xpoint storage technology may be an Optane DC (data center) persistent memory, and the Optane non-volatile memory DCPM may be used in virtualization, etc. Sufficient capacity can be provided in scenarios that require a large amount of memory.
  • the Optane non-volatile memory DCPM Due to its non-volatile nature, the loading time of data and processes can be avoided after the server restarts, and the Optane non-volatile memory DCPM is relative to Dynamic random access memory DRAM has a low cost, which can reduce the total cost of the server and the cloud host 1, and provide sufficient memory 10, thereby reducing the number of devices, energy consumption, etc., especially for read and write performance requirements For very high scenarios, the speed and latency gap between dynamic random access memory DRAM and SSD is still too large.
  • the Optane non-volatile memory DCPM can provide relatively large storage capacity while providing close to dynamic random access. The read and write speed and latency of memory DRAM are unmatched by traditional SSDs.
  • the processor 11 reads target data from the memory 10, and the target data is data that needs to be called during the operation of the cloud host 1.
  • the cloud host further includes a cache 12, the cache 12 adopts a dynamic random access memory, the cache 12 is used to store first data, and the memory 10 is used to store second data.
  • Data wherein, the memory 10 using a non-volatile memory includes multiple storage modes, in this embodiment, the non-volatile memory is set to a memory mode (Memory Mode), the memory mode (Memory Mode) is mainly applied to scenes that require large-capacity memory.
  • the Optane non-volatile memory DCPM is recognized by the cloud host system as the dynamic random access memory DRAM memory, and the actual dynamic random access memory DRAM memory can be used As the cache of the Optane non-volatile memory DCPM.
  • the CPU of the cloud host 1 uses a dynamic random access memory DRAM memory as a cache, and the Optane non-volatile memory DCPM is used as a memory.
  • the processor 11 searches for target data from the first data in the cache 12. As shown in FIG. 3(a), when the processor 11 fails to find the target data from the first data, Then, the target data is searched from the second data in the memory 10.
  • the cloud host further includes a memory manager 13.
  • the processor 11 fails to find the target data from the first data
  • the processor 11 controls the The memory manager 13 searches for the target data from the second data of the memory 10, and returns the searched target data to the processor 11, as shown in FIG. 3(b).
  • the cloud host 1 proposed in this application uses the memory of 3D Xpoint storage technology as the non-volatile memory on the server and allocates it to the user as the memory 10 of the cloud host 1, which can provide users with a memory 10 with high reliability and readability. Fetch high-speed, low-cost and cost-effective cloud hosts1.
  • a hybrid configuration of the DRAM cache 12 is adopted to improve the processing performance of the cloud host 1 and further improve the user experience.
  • the processor 11 of the cloud host searches for target data from the first data in the cache 12, and when the processor 11 fails to find the target data from the first data, then Searching for the target data in the second data in the memory 10 can increase the computing speed of the cloud host 1 and improve user experience.
  • FIG. 4 is a schematic structural diagram of a cloud host memory allocation device 30 according to an embodiment of the present application.
  • the cloud host memory allocation device 30 includes a processor 32 and a memory 31 coupled to the processor 32, wherein:
  • the memory 31 stores computer-readable instructions, and when the computer-readable instructions are executed by the processor 32, the following steps are implemented: obtaining a configuration instruction for a user to apply to create a cloud host; and obtaining all the instructions requested by the user according to the configuration instruction.
  • the configuration instruction includes the number of CPUs and memory capacity configuration values of the cloud host requested by the user, and the obtaining of the memory capacity of the cloud host requested by the user according to the configuration instruction includes: If the memory capacity configuration value is not empty, the memory capacity configuration value is used as the memory capacity of the cloud host requested by the user; if the memory capacity configuration value is empty, it is based on the number of CPUs and The preset memory capacity corresponding to a single cpu acquires the memory capacity of the cloud host requested by the user.
  • the cloud host memory allocation method further includes the following steps: calculating the cache capacity required by the cloud host according to a preset ratio according to the memory capacity; and calculating the preset cache capacity according to the cache capacity
  • the storage space of the corresponding capacity in the dynamic random access memory on the server is allocated to the cloud host as a cache of the cloud host.
  • the non-volatile memory includes multiple storage modes
  • the storage mode includes a memory mode
  • the cloud host memory allocation method further includes: setting the storage mode of the non-volatile memory Is the memory mode.
  • FIG. 5 is a schematic structural diagram of a storage medium according to an embodiment of the present application.
  • a storage medium storing computer-readable instructions 41.
  • the computer-readable instructions 41 are executed by one or more processors, one or more processors are caused to perform the following steps: Configuration instruction; obtaining the memory capacity of the cloud host requested by the user according to the configuration instruction; and assigning a corresponding capacity of storage space in the non-volatile memory on the preset server to the cloud host according to the memory capacity
  • the configuration instruction includes the number of CPUs and memory capacity configuration values of the cloud host requested by the user, and the obtaining of the memory capacity of the cloud host requested by the user according to the configuration instruction includes: If the memory capacity configuration value is not empty, the memory capacity configuration value is used as the memory capacity of the cloud host requested by the user; if the memory capacity configuration value is empty, it is based on the number of CPUs and The preset memory capacity corresponding to a single cpu acquires the memory capacity of the cloud host requested by the user.
  • the cloud host memory allocation method further includes the following steps: calculating the cache capacity required by the cloud host according to a preset ratio according to the memory capacity; and calculating the preset cache capacity according to the cache capacity
  • the storage space of the corresponding capacity in the dynamic random access memory on the server is allocated to the cloud host as a cache of the cloud host.
  • the non-volatile memory includes multiple storage modes
  • the storage mode includes a memory mode
  • the cloud host memory allocation method further includes: setting the storage mode of the non-volatile memory Is the memory mode.
  • the computer program can be stored in a computer readable storage medium. When executed, it may include the procedures of the above-mentioned method embodiments.
  • the aforementioned storage medium may be a magnetic disk, an optical disk, or a read-only storage memory (Read-Only Memory, ROM) and other non-volatile storage media, can also be random storage memory (Random Access Memory, RAM) and other volatile storage media.

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Memory System Of A Hierarchy Structure (AREA)

Abstract

一种云主机内存分配方法及云主机、设备及存储介质,涉及云计算技术领域,所述云主机内存分配方法包括:获取用户申请创建云主机的配置指令(S101);依据所述配置指令获取用户申请的所述云主机的内存容量(S102);及依据所述内存容量将预设服务器上的非易失性存储器中相应容量的储存空间分配给所述云主机作为所述云主机的内存(S103),其中,所述服务器上非易失性存储器为采用3D Xpoint存储技术的存储器。该方法能够达到提高云主机内存可靠性的同时,保证读取速度,降低成本,提高内存容量的目的。

Description

云主机内存分配方法及云主机、设备及存储介质
本申请以2020年7月21日提交的申请号为202010707528.8,发明名称为“云主机内存分配方法及云主机、设备及存储介质”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及云计算技术领域,尤其涉及一种云主机内存分配方法及云主机、设备及存储介质。
背景技术
云服务是基于互联网的相关服务的增加、使用和交互模式,通常涉及通过互联网来提供动态易扩展且经常是虚拟化的资源。云服务指通过网络以按需、易扩展的方式获得所需服务。随着科技的发展,现在越来越多用户使用更易扩展,成本更低的云主机替代本地计算机,对于云主机而言,内存是云主机不可或缺的组成部分,内存是与CPU进行沟通的桥梁,所有程序的运行都是在内存中进行的,因此内存的性能对云主机的影响非常大。
现有技术中,云主机通常采用DRAM内存(Dynamic Random Access Memory, 动态随机存取存储器),DRAM内存具有延迟很低,通常为纳秒级别、带宽较为充裕等优点,发明人意识到,DRAM内存在断电等异常场景下具有数据易失性,故数据可靠性低,且成本较高。 因此,如何在云主机上配置读取速度快、可靠性高、成本低、容量大的内存是用户的迫切需求。
技术问题
本申请提供一种云主机内存分配方法及云主机、设备及存储介质,能够达到提高云主机内存可靠性的同时,保证读取速度,降低成本,提高内存容量的目的。
技术解决方案
为解决上述技术问题,本申请采用的一个技术方案是:提供一种云主机内存分配方法,所述云主机内存分配方法包括以下步骤:
获取用户申请创建云主机的配置指令;
依据所述配置指令获取用户申请的所述云主机的内存容量;及
依据所述内存容量将预设服务器上的非易失性存储器中相应容量的储存空间分配给所述云主机作为所述云主机的内存,其中,所述服务器上非易失性存储器为采用3D Xpoint存储技术的存储器。
根据本申请的一种实施例,所述配置指令包括用户申请的所述云主机的cpu个数、内存容量配置值,所述依据所述配置指令获取用户申请的所述云主机的内存容量,包括:
若所述内存容量配置值为非空,则将所述内存容量配置值作为用户申请的所述云主机的所述内存容量;
若所述内存容量配置值为空,则依据所述cpu个数及单个cpu对应的预设内存容量获取用户申请的所述云主机的所述内存容量。
根据本申请的一种实施例,所述云主机内存分配方法还包括以下步骤:
依据所述内存容量并按预设比例计算所述云主机所需的缓存容量;及
依据所述缓存容量将所述预设服务器上的动态随机存取存储器中相应容量的储存空间分配给所述云主机作为所述云主机的缓存。
根据本申请的一种实施例,所述非易失性存储器包括多种存储模式,所述存储模式包括内存模式,所述云主机内存分配方法还包括:
设置所述非易失性存储器的所述存储模式为所述内存模式。
根据本申请的一种实施例,所述单个cpu对应的预设内存容量为8G。
此外,为解决上述技术问题,本申请还采用的一个技术方案是:提供一种云主机,所述云主机包括内存,所述内存采用非易失存储器,所述非易失性存储器为采用3D Xpoint存储技术的存储器。
根据本申请的一种实施例,所述云主机还包括处理器、缓存,
所述缓存采用动态随机存取存储器,所述缓存用于存储第一数据,所述内存用于存储第二数据;
所述处理器用于从所述缓存的所述第一数据中查找目标数据,当所述处理器从所述第一数据中查找所述目标数据失败,则从所述内存的所述第二数据中查找所述目标数据,所述目标数据为所述云主机运行过程中需要调用的数据。
根据本申请的一种实施例,所述云主机还包括内存管理器,当所述处理器从所述第一数据中查找所述目标数据失败,则所述处理器控制所述内存管理器从所述内存的所述第二数据中查找所述目标数据。
此外,为解决上述技术问题,本申请还采用的一个技术方案是:提供一种云主机内存分配设备,所述云主机内存分配设备包括处理器、与所述处理器耦接的存储器,其中,
所述存储器中存储有计算机可读指令,所述计算机可读指令被所述处理器执行时,使得所述处理器执行如下步骤:
获取用户申请创建云主机的配置指令;
依据所述配置指令获取用户申请的所述云主机的内存容量;及
依据所述内存容量将预设服务器上的非易失性存储器中相应容量的储存空间分配给所述云主机作为所述云主机的内存,其中,所述服务器上非易失性存储器为采用3D Xpoint存储技术的存储器。
此外,为解决上述技术问题,本申请还采用的一个技术方案是:提供一种存储有计算机可读指令的存储介质,所述计算机可读指令被一个或多个处理器执行时,使得一个或多个处理器执行如下步骤:
获取用户申请创建云主机的配置指令;
依据所述配置指令获取用户申请的所述云主机的内存容量;及
依据所述内存容量将预设服务器上的非易失性存储器中相应容量的储存空间分配给所述云主机作为所述云主机的内存,其中,所述服务器上非易失性存储器为采用3D Xpoint存储技术的存储器。
有益效果
本申请提出的云主机内存分配方法及云主机、设备及存储介质,通过采用3D Xpoint存储技术的存储器作为服务器上非易失性存储器,并按用户配置指令分配给用户作为云主机内存,可以为用户提供一种内存可靠性高、读取速度快、成本低的高性价比云主机。
附图说明
图1是本申请一种实施例的云主机内存分配方法的流程示意图;
图2是本申请一种实施例的云主机的结构示意图;
图3是图1所述云主机查找目标数据的原理示意图;
图4是本申请一种实施例的云主机内存分配设备的结构示意图;
图5是本申请一种实施例的存储介质的结构示意图。
本发明的实施方式
下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅是本申请的一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本申请保护的范围。
图1是本申请一种实施例的云主机内存分配方法的流程示意图。所述云主机内存分配方法可以运行于本地服务器、远程服务器或远程云平台上,所述本地服务器、所述远程服务器或所述远程云平台通过WIFI等无线网络或者以太网等有线网络可以与用户端通信,所述用户端可以是能够提供被用户操作的任意终端,例如为手机、平板电脑、笔记本等。需注意的是,若有实质上相同的结果,本申请的方法并不以图1所示的流程顺序为限。如图1所示,该方法包括步骤:
步骤S101:获取用户申请创建云主机的配置指令。
用户通过可以在用户端通过鼠标、键盘等输入设备,使用脚本命令、图形界面等方式向所述服务器发送创建云主机的配置指令,本实施例中,所述配置指令可以包括用户申请的所述云主机的cpu个数、内存容量配置值。
步骤S102:依据所述配置指令获取用户申请的所述云主机的内存容量。
根据本申请的一种实施例,可以给用户提供配置接口,使用户可以通过所述配置接口输入所述配置指令指定,包括对所述云主机的所述内存容量的配置值,所述服务器可以依据所述配置指令可以获取用户申请的所述云主机的内存容量。
需要说明的是,可以理解,本实施例中,为了不对用户操作做过多的限制,在满足用户的操作习惯的同时,提供简便的操作,用户可以通过提供的配置接口输入按用户需求配置的所述内存容量配置值,也可以不输入指定的所述内存容量配置值,使所述内存容量配置值为空。
具体的,若所述内存容量配置值为非空,则将所述内存容量配置值作为用户申请的所述云主机的所述内存容量;若所述内存容量配置值为空,则可以依据所述cpu个数及单个cpu对应的预设内存容量获取用户申请的所述云主机的所述内存容量。其中,单个cpu对应的预设内存容量可以人为设置,可以根据用户的应用需求设置合理的预设内存容量,本实施例中,所述单个cpu对应的预设内存容量为8G,即当用户通过所述配置指令指定所述云主机的cpu个数为1个时,若用户输入的所述配置指令中所述内存容量配置值为空,则认为所述内存容量为8G,当用户通过所述配置指令指定所述云主机的cpu个数为2个时,若用户输入的所述配置指令中所述内存容量配置值为空,则认为所述内存容量为16G,依此类推。
步骤S103: 依据所述内存容量将预设服务器上的非易失性存储器中相应容量的储存空间分配给所述云主机作为所述云主机的内存,其中,所述服务器上非易失性存储器为采用3D Xpoint存储技术的存储器。
步骤S103中,所述服务器上安装有非易失性存储器,所述非易失性存储器也可以安装在其他存储设备上,并与所述服务器通信连接,所述服务器依据从所述配置指令中获取的所述内存容量或者从所述配置指令中计算出的所述内存容量将预设服务器上的所述非易失性存储器中相应容量的储存空间分配给所述云主机作为所述云主机的内存。
根据本申请的一种实施例,所述服务器上非易失性存储器为采用3D Xpoint存储技术的存储器。本实施例中,所述采用3D Xpoint存储技术的存储器可以是傲腾非易失性内存DCPM(optane DC(data center)persistent memory),所述傲腾非易失性内存DCPM可在虚拟化等需要大量内存的场景下可提供足够的容量,由于其具有非易失性,可在所述服务器重启后免去数据和进程的载入时间,并且所述傲腾非易失性内存DCPM相对于动态随机存取存储器DRAM内存成本较低,可以降低所述服务器及所述云主机的总成本,并提供足够的内存,从而减少设备数量、能源消耗等,特别的,对读写性能要求很高的场景来说动态随机存取存储器DRAM和SSD的速度和延迟差距还是过大,所述傲腾非易失性内存DCPM可提供相对较大的存储容量的同时可提供接近动态随机存取存储器DRAM的读写速度和延迟,是传统SSD无法比拟的。
本实施例的云主机内存分配方法,通过采用3D Xpoint存储技术的存储器作为服务器上非易失性存储器,并按用户配置指令分配给用户作为云主机内存,可以为用户提供一种内存可靠性高、读取速度快、成本低的高性价比云主机。
进一步地,在用户没有指定具体内存容量时,可以按cpu个数及单个cpu对应的预设内存容量为用户分配云主机内存,保证了云主机的性能,提高用户体验。
在另一实施例中,所述云主机内存分配方法还包括以下步骤:
步骤S104:依据所述内存容量并按预设比例计算所述云主机所需的缓存容量。
需要说明的是,根据本申请的一种实施例,所述服务器上设有动态随机存取存储器DRAM,所述非易失性存储器包括多种存储模式,本实施例中,所述服务器上非易失性存储器设置为内存模式(Memory Mode),所述内存模式(Memory Mode)主要应用于需要大容量内存的场景。所述内存模式(Memory Mode)下所述傲腾非易失性内存DCPM被所述云主机系统认成所述动态随机存取存储器DRAM内存,而实际的动态随机存取存储器DRAM内存可以被用作为所述傲腾非易失性内存DCPM的缓存。所述云主机的CPU将动态随机存取存储器DRAM内存用作高速缓存,将所述傲腾非易失性内存DCPM用作内存。
具体的,步骤S104中,可以依据所述内存容量并按预设比例计算所述云主机所需的缓存容量,所述预设比例可以根据所述云主机上的应用程序的需求进行人为设置。
步骤S105:依据所述缓存容量将所述预设服务器上的动态随机存取存储器中相应容量的储存空间分配给所述云主机作为所述云主机的缓存。
步骤S105中,计算出所述缓存容量后,可以将所述服务器上的所述动态随机存取存储器DRAM中相应容量的储存空间分配给所述云主机作为所述云主机的缓存,使所述云主机的CPU在读取数据时,所述云主机的CPU可以先从所述动态随机存取存储器DRAM中查找所述数据,当所述云主机的CPU从所述动态随机存取存储器DRAM中读取所述数据失败后,再从所述非易失性存储器中查找所述数据。
本实施例的一种云主机内存分配方法,依据所述内存容量并按预设比例计算所述云主机所需的缓存容量并为云主机分配缓存,提升所述云主机处理性能,进一步提高用户体验。
进一步地,所述云主机的CPU可以先从所述动态随机存取存储器DRAM中查找所述数据,当所述云主机的CPU从所述动态随机存取存储器DRAM中读取所述数据失败后,再从所述非易失性存储器中查找所述数据,可以提升所述云主机的运算速度,提升用户体验。
请一并参阅图2、图3,图2是本申请一种实施例的云主机1的结构示意图;图3是图1所述云主机1查找目标数据的原理示意图。本申请还提供一种云主机1,所述云主机包括内存10、处理器11。
所述内存10采用非易失存储器,所述非易失性存储器为采用3D Xpoint存储技术的存储器。本实施例中,所述采用3D Xpoint存储技术的存储器可以是傲腾非易失性内存DCPM(optane DC(data center)persistent memory),所述傲腾非易失性内存DCPM可在虚拟化等需要大量内存的场景下可提供足够的容量,由于其具有非易失性,可在所述服务器重启后免去数据和进程的载入时间,并且所述傲腾非易失性内存DCPM相对于动态随机存取存储器DRAM内存成本较低,可以降低所述服务器及所述云主机1的总成本,并提供足够的内存10,从而减少设备数量、能源消耗等,特别的,对读写性能要求很高的场景来说动态随机存取存储器DRAM和SSD的速度和延迟差距还是过大,所述傲腾非易失性内存DCPM可提供相对较大的存储容量的同时可提供接近动态随机存取存储器DRAM的读写速度和延迟,是传统SSD无法比拟的。
所述处理器11从所述内存10中读取目标数据,所述目标数据为所述云主机1运行过程中需要调用的数据。
根据本申请的另一种实施例,所述云主机还包括缓存12,所述缓存12采用动态随机存取存储器,所述缓存12用于存储第一数据,所述内存10用于存储第二数据;其中,采用非易失存储器的所述内存10包括多种存储模式,本实施例中,所述非易失性存储器设置为内存模式(Memory Mode),所述内存模式(Memory Mode)主要应用于需要大容量内存的场景。所述内存模式(Memory Mode)下所述傲腾非易失性内存DCPM被所述云主机系统认成所述动态随机存取存储器DRAM内存,而实际的动态随机存取存储器DRAM内存可以被用作为所述傲腾非易失性内存DCPM的缓存。所述云主机1的CPU将动态随机存取存储器DRAM内存用作高速缓存,将所述傲腾非易失性内存DCPM用作内存。
所述处理器11从所述缓存12的所述第一数据中查找目标数据,如图3(a)所示,当所述处理器11从所述第一数据中查找所述目标数据失败,则从所述内存10的所述第二数据中查找所述目标数据。
根据本申请的另一种实施例,所述云主机还包括内存管理器13,当所述处理器11从所述第一数据中查找所述目标数据失败,则所述处理器11控制所述内存管理器13从所述内存10的所述第二数据中查找所述目标数据,并将查找到的所述目标数据返回给所述处理器11,如图3(b)所示。
本申请提出的云主机1,通过采用3D Xpoint存储技术的存储器作为服务器上非易失性存储器,并分配给用户作为云主机1的内存10,可以为用户提供一种内存10可靠性高、读取速度快、成本低的高性价比云主机1。
进一步地,采用DRAM缓存12混合配置,提升云主机1处理性能,进一步提高用户体验。
进一步地,对于云主机的所述处理器11从所述缓存12的所述第一数据中查找目标数据,当所述处理器11从所述第一数据中查找所述目标数据失败,则从所述内存10的所述第二数据中查找所述目标数据,可以提升云主机1的运算速度,提升用户体验。
请参阅图4,图4是本申请一种实施例的云主机内存分配设备30的结构示意图。所述云主机内存分配设备30包括处理器32、与所述处理器32耦接的存储器31,其中,
所述存储器31中存储有计算机可读指令,所述计算机可读指令被所述处理器32执行时实现以下步骤:获取用户申请创建云主机的配置指令;依据所述配置指令获取用户申请的所述云主机的内存容量;及依据所述内存容量将预设服务器上的非易失性存储器中相应容量的储存空间分配给所述云主机作为所述云主机的内存,其中,所述服务器上非易失性存储器为采用3D Xpoint存储技术的存储器。
在一个实施例中,所述配置指令包括用户申请的所述云主机的cpu个数、内存容量配置值,所述依据所述配置指令获取用户申请的所述云主机的内存容量,包括:若所述内存容量配置值为非空,则将所述内存容量配置值作为用户申请的所述云主机的所述内存容量;若所述内存容量配置值为空,则依据所述cpu个数及单个cpu对应的预设内存容量获取用户申请的所述云主机的所述内存容量。
在一个实施例中,所述云主机内存分配方法还包括以下步骤:依据所述内存容量并按预设比例计算所述云主机所需的缓存容量;及依据所述缓存容量将所述预设服务器上的动态随机存取存储器中相应容量的储存空间分配给所述云主机作为所述云主机的缓存。
在一个实施例中,所述非易失性存储器包括多种存储模式,所述存储模式包括内存模式,所述云主机内存分配方法还包括:设置所述非易失性存储器的所述存储模式为所述内存模式。
请参阅图5,图5是本申请一种实施例的存储介质的结构示意图。如图5所示存储有计算机可读指令41的存储介质,该计算机可读指令41被一个或多个处理器执行时,使得一个或多个处理器执行以下步骤:获取用户申请创建云主机的配置指令;依据所述配置指令获取用户申请的所述云主机的内存容量;及依据所述内存容量将预设服务器上的非易失性存储器中相应容量的储存空间分配给所述云主机作为所述云主机的内存,其中,所述服务器上非易失性存储器为采用3D Xpoint存储技术的存储器。
在一个实施例中,所述配置指令包括用户申请的所述云主机的cpu个数、内存容量配置值,所述依据所述配置指令获取用户申请的所述云主机的内存容量,包括:若所述内存容量配置值为非空,则将所述内存容量配置值作为用户申请的所述云主机的所述内存容量;若所述内存容量配置值为空,则依据所述cpu个数及单个cpu对应的预设内存容量获取用户申请的所述云主机的所述内存容量。
在一个实施例中,所述云主机内存分配方法还包括以下步骤:依据所述内存容量并按预设比例计算所述云主机所需的缓存容量;及依据所述缓存容量将所述预设服务器上的动态随机存取存储器中相应容量的储存空间分配给所述云主机作为所述云主机的缓存。
在一个实施例中,所述非易失性存储器包括多种存储模式,所述存储模式包括内存模式,所述云主机内存分配方法还包括:设置所述非易失性存储器的所述存储模式为所述内存模式。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程,是可以通过计算机程序来指令相关的硬件来完成,该计算机程序可存储于一计算机可读取存储介质中,该程序在执行时,可包括如上述各方法的实施例的流程。其中,前述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory,ROM)等非易失性存储介质,也可以是随机存储记忆体(Random Access Memory,RAM)等易失性存储介质。
以上所述实施例的各技术特征可以进行任意的组合,为使描述简洁,未对上述实施例中的各个技术特征所有可能的组合都进行描述,然而,只要这些技术特征的组合不存在矛盾,都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本申请的几种实施方式,其描述较为具体和详细,但并不能因此而理解为对本申请专利范围的限制。应当指出的是,对于本领域的普通技术人员来说,在不脱离本申请构思的前提下,还可以做出若干变形和改进,这些都属于本申请的保护范围。因此,本申请专利的保护范围应以所附权利要求为准。

Claims (20)

1、一种云主机内存分配方法,其中,所述云主机内存分配方法包括以下步骤:
获取用户申请创建云主机的配置指令;
依据所述配置指令获取用户申请的所述云主机的内存容量;及
依据所述内存容量将预设服务器上的非易失性存储器中相应容量的储存空间分配给所述云主机作为所述云主机的内存,其中,所述服务器上非易失性存储器为采用3D Xpoint存储技术的存储器。
2、根据权利要求1所述云主机内存分配方法,其中,所述配置指令包括用户申请的所述云主机的cpu个数、内存容量配置值,所述依据所述配置指令获取用户申请的所述云主机的内存容量,包括:
若所述内存容量配置值为非空,则将所述内存容量配置值作为用户申请的所述云主机的所述内存容量;
若所述内存容量配置值为空,则依据所述cpu个数及单个cpu对应的预设内存容量获取用户申请的所述云主机的所述内存容量。
3、根据权利要求1所述云主机内存分配方法,其中,所述云主机内存分配方法还包括以下步骤:
依据所述内存容量并按预设比例计算所述云主机所需的缓存容量;及
依据所述缓存容量将所述预设服务器上的动态随机存取存储器中相应容量的储存空间分配给所述云主机作为所述云主机的缓存。
4、根据权利要求1所述云主机内存分配方法,其中,所述非易失性存储器包括多种存储模式,所述存储模式包括内存模式,所述云主机内存分配方法还包括:
设置所述非易失性存储器的所述存储模式为所述内存模式。
5、根据权利要求2所述云主机内存分配方法,其中,所述单个cpu对应的预设内存容量为8G。
6、一种云主机,其中,所述云主机包括内存,所述内存采用非易失存储器,所述非易失性存储器为采用3D Xpoint存储技术的存储器。
7、根据权利要求6所述云主机,其中,所述云主机还包括处理器、缓存,
所述缓存采用动态随机存取存储器,所述缓存用于存储第一数据,所述内存用于存储第二数据;
所述处理器用于从所述缓存的所述第一数据中查找目标数据,当所述处理器从所述第一数据中查找所述目标数据失败,则从所述内存的所述第二数据中查找所述目标数据,所述目标数据为所述云主机运行过程中需要调用的数据。
8、根据权利要求7所述云主机,其中,所述云主机还包括内存管理器,当所述处理器从所述第一数据中查找所述目标数据失败,则所述处理器控制所述内存管理器从所述内存的所述第二数据中查找所述目标数据。
9、一种云主机内存分配设备,其中,所述云主机内存分配设备包括处理器、与所述处理器耦接的存储器,其中,
所述存储器中存储有计算机可读指令,所述计算机可读指令被所述处理器执行时,使得所述处理器执行如下步骤:
获取用户申请创建云主机的配置指令;
依据所述配置指令获取用户申请的所述云主机的内存容量;及
依据所述内存容量将预设服务器上的非易失性存储器中相应容量的储存空间分配给所述云主机作为所述云主机的内存,其中,所述服务器上非易失性存储器为采用3D Xpoint存储技术的存储器。
10、根据权利要求9所述云主机内存分配设备,其中,所述配置指令包括用户申请的所述云主机的cpu个数、内存容量配置值,所述依据所述配置指令获取用户申请的所述云主机的内存容量,包括:
若所述内存容量配置值为非空,则将所述内存容量配置值作为用户申请的所述云主机的所述内存容量;
若所述内存容量配置值为空,则依据所述cpu个数及单个cpu对应的预设内存容量获取用户申请的所述云主机的所述内存容量。
11、根据权利要求9所述云主机内存分配设备,其中,所述云主机内存分配方法还包括以下步骤:
依据所述内存容量并按预设比例计算所述云主机所需的缓存容量;及
依据所述缓存容量将所述预设服务器上的动态随机存取存储器中相应容量的储存空间分配给所述云主机作为所述云主机的缓存。
12、根据权利要求9所述云主机内存分配设备,其中,所述非易失性存储器包括多种存储模式,所述存储模式包括内存模式,所述云主机内存分配方法还包括:
设置所述非易失性存储器的所述存储模式为所述内存模式。
13、根据权利要求10所述云主机内存分配设备,其中,所述单个cpu对应的预设内存容量为8G。
14、根据权利要求9所述云主机内存分配设备,其中,所述云主机包括内存,所述内存采用非易失存储器,所述非易失性存储器为采用3D Xpoint存储技术的存储器。
15、一种存储有计算机可读指令的存储介质,所述计算机可读指令被一个或多个处理器执行时,使得一个或多个处理器执行如如下步骤:
获取用户申请创建云主机的配置指令;
依据所述配置指令获取用户申请的所述云主机的内存容量;及
依据所述内存容量将预设服务器上的非易失性存储器中相应容量的储存空间分配给所述云主机作为所述云主机的内存,其中,所述服务器上非易失性存储器为采用3D Xpoint存储技术的存储器。
16、根据权利要求15所述云主机内存分配设备,其中,所述配置指令包括用户申请的所述云主机的cpu个数、内存容量配置值,所述依据所述配置指令获取用户申请的所述云主机的内存容量,包括:
若所述内存容量配置值为非空,则将所述内存容量配置值作为用户申请的所述云主机的所述内存容量;
若所述内存容量配置值为空,则依据所述cpu个数及单个cpu对应的预设内存容量获取用户申请的所述云主机的所述内存容量。
17、根据权利要求15所述云主机内存分配设备,其中,所述云主机内存分配方法还包括以下步骤:
依据所述内存容量并按预设比例计算所述云主机所需的缓存容量;及
依据所述缓存容量将所述预设服务器上的动态随机存取存储器中相应容量的储存空间分配给所述云主机作为所述云主机的缓存。
18、根据权利要求15所述云主机内存分配设备,其中,所述非易失性存储器包括多种存储模式,所述存储模式包括内存模式,所述云主机内存分配方法还包括:
设置所述非易失性存储器的所述存储模式为所述内存模式。
19、根据权利要求16所述云主机内存分配设备,其中,所述单个cpu对应的预设内存容量为8G。
20、根据权利要求15所述云主机内存分配设备,其中,所述云主机包括内存,所述内存采用非易失存储器,所述非易失性存储器为采用3D Xpoint存储技术的存储器。
PCT/CN2020/123237 2020-07-21 2020-10-23 云主机内存分配方法及云主机、设备及存储介质 WO2021120843A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010707528.8A CN111831451A (zh) 2020-07-21 2020-07-21 云主机内存分配方法及云主机、设备及存储介质
CN202010707528.8 2020-07-21

Publications (1)

Publication Number Publication Date
WO2021120843A1 true WO2021120843A1 (zh) 2021-06-24

Family

ID=72923859

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/123237 WO2021120843A1 (zh) 2020-07-21 2020-10-23 云主机内存分配方法及云主机、设备及存储介质

Country Status (2)

Country Link
CN (1) CN111831451A (zh)
WO (1) WO2021120843A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113419671B (zh) * 2021-05-20 2022-02-18 佛山市禅城区政务服务数据管理局 一种个人数据空间访问控制方法
CN113569137B (zh) * 2021-07-06 2024-06-18 北京汇钧科技有限公司 云主机规格的推荐方法及装置、存储介质、电子装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180107595A1 (en) * 2016-10-18 2018-04-19 Micron Technology, Inc. Apparatuses and methods for an operating system cache in a solid state device
CN109558421A (zh) * 2018-10-29 2019-04-02 中国建设银行股份有限公司 基于缓存的数据处理方法、系统、装置及存储介质
CN110209490A (zh) * 2018-04-27 2019-09-06 腾讯科技(深圳)有限公司 一种内存管理方法及相关设备
CN110896406A (zh) * 2018-09-13 2020-03-20 华为技术有限公司 数据存储方法、装置及服务器
CN111159062A (zh) * 2019-12-20 2020-05-15 海光信息技术有限公司 一种缓存数据的调度方法、装置、cpu芯片及服务器

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110287127A (zh) * 2019-05-14 2019-09-27 江苏大学 一种多粒度多核可扩展的非易失性内存管理方法及系统
CN110297713A (zh) * 2019-05-22 2019-10-01 菜鸟智能物流控股有限公司 一种云主机的配置管理系统和方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180107595A1 (en) * 2016-10-18 2018-04-19 Micron Technology, Inc. Apparatuses and methods for an operating system cache in a solid state device
CN110209490A (zh) * 2018-04-27 2019-09-06 腾讯科技(深圳)有限公司 一种内存管理方法及相关设备
CN110896406A (zh) * 2018-09-13 2020-03-20 华为技术有限公司 数据存储方法、装置及服务器
CN109558421A (zh) * 2018-10-29 2019-04-02 中国建设银行股份有限公司 基于缓存的数据处理方法、系统、装置及存储介质
CN111159062A (zh) * 2019-12-20 2020-05-15 海光信息技术有限公司 一种缓存数据的调度方法、装置、cpu芯片及服务器

Also Published As

Publication number Publication date
CN111831451A (zh) 2020-10-27

Similar Documents

Publication Publication Date Title
CN108027803B (zh) 分布式无锁基于rdma的存储器分配和解分配
US10496613B2 (en) Method for processing input/output request, host, server, and virtual machine
EP3608792A1 (en) Managed switching between one or more hosts and solid state drives (ssds) based on the nvme protocol to provide host storage services
JP7280508B2 (ja) 情報処理装置、情報処理方法および仮想マシン接続管理プログラム
US11966612B2 (en) Solid-state disk (SSD) data migration
US20200310859A1 (en) System and method for an object layer
US10733027B2 (en) Memory allocator
WO2021098267A1 (zh) 一种磁盘处理的方法、系统、设备及可读存储介质
US9807014B2 (en) Reactive throttling of heterogeneous migration sessions in a virtualized cloud environment
CN110196681B (zh) 业务写操作的磁盘数据写入控制方法及装置、电子设备
TW201220197A (en) for improving the safety and reliability of data storage in a virtual machine based on cloud calculation and distributed storage environment
US10248460B2 (en) Storage management computer
US9735984B2 (en) Bursting cloud resources to affect state change performance
JP6262360B2 (ja) 計算機システム
US20220066928A1 (en) Pooled memory controller for thin-provisioning disaggregated memory
US20080229040A1 (en) Network storage system, management method therefor, and control program product therefor
WO2021120843A1 (zh) 云主机内存分配方法及云主机、设备及存储介质
US20240143377A1 (en) Overlay container storage driver for microservice workloads
WO2023029610A1 (zh) 数据访问方法、装置及存储介质
CN107577733B (zh) 一种数据复制的加速方法及系统
CN107329798B (zh) 数据复制的方法、装置和虚拟化系统
WO2016119618A1 (zh) 一种远端内存分配方法、装置和系统
CN114726657B (zh) 中断管理和数据收发管理的方法、装置及智能网卡
WO2022262623A1 (zh) 一种数据交换方法及装置
EP4109255A1 (en) Scheduling in a container orchestration system utilizing hardware topology hints

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20900763

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20900763

Country of ref document: EP

Kind code of ref document: A1