CN117608474A - Method and device for unloading load to DPU based on local storage volume - Google Patents

Method and device for unloading load to DPU based on local storage volume Download PDF

Info

Publication number
CN117608474A
CN117608474A CN202311564539.5A CN202311564539A CN117608474A CN 117608474 A CN117608474 A CN 117608474A CN 202311564539 A CN202311564539 A CN 202311564539A CN 117608474 A CN117608474 A CN 117608474A
Authority
CN
China
Prior art keywords
dpu
file
file system
workload
target
Prior art date
Legal status (The legal status 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 status listed.)
Pending
Application number
CN202311564539.5A
Other languages
Chinese (zh)
Inventor
郎爽
黄明亮
鄢贵海
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Yusur Technology Co ltd
Original Assignee
Yusur Technology Co ltd
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 Yusur Technology Co ltd filed Critical Yusur Technology Co ltd
Priority to CN202311564539.5A priority Critical patent/CN117608474A/en
Publication of CN117608474A publication Critical patent/CN117608474A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0629—Configuration or reconfiguration of storage systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G06F9/5055—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering software capabilities, i.e. software resources associated or available to the machine

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Human Computer Interaction (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

The invention discloses a method and a device for unloading load to a DPU based on a local storage volume, wherein the method comprises the following steps: obtaining a file system path of a target file which is stored in a host by a local storage volume and is required to be unloaded to a DPU; establishing a shared file system between a file system path of the target file and the DPU; the workload unloaded to the DPU accesses the target file through the shared file system; the invention can make the DPU sense and access the appointed file or file catalog of host side in real time, thereby enriching the load types which can be unloaded to the DPU and reducing the resource consumption of host.

Description

Method and device for unloading load to DPU based on local storage volume
Technical Field
The invention relates to the technical field of computers, in particular to a method and a device for unloading load to a DPU (data processing unit) based on a local storage volume.
Background
The DPU network card has the functions of accelerating network traffic processing, improving the performance of containers and micro services, enhancing safety and providing higher performance and reliability for cloud native application in the cloud native environment. In the cloud native container platform, in order to further improve the performance of the service container, some loads need to be offloaded onto the DPU hardware, for example, a node agent of the container platform is offloaded onto the DPU, however, the node agent may need a specific path in the hostpath volume for communication with each plug-in, and the content under the path changes according to the deployment condition of k8s, while the DPU is outside the container cluster and cannot perform real-time synchronous sensing, so that normal work after load offloading cannot be realized.
That is, the DPU is a new favor in the cloud native environment, the industry has not formed a more mature and universal load-off scheme, and if only a poc test is performed, a developer can copy a file of a specific hostpath onto the DPU path directly through a scp command, or download the file through a wget command if the DPU network is reachable. However, the manual operation cannot be changed according to the file change in the pod images deployed by the clusters, and there is no way to enable host and DPU to simultaneously sense file modification or perform automatic synchronization, so that the method has no practical application value; meanwhile, the manual operation does not set a security access policy, and the data lacks security guarantee.
Disclosure of Invention
In view of the above, the present invention provides a method and apparatus for unloading load to a DPU based on a local storage volume, which can enable the DPU to sense and access a designated file or file directory on the host side in real time, thereby enriching load types that can be unloaded to the DPU and reducing resource consumption of the host.
In order to solve the technical problems, the invention is realized as follows:
a method of offloading load to a DPU based on a local storage volume, comprising:
obtaining a file system path of a target file which is stored in a host by a local storage volume and is required to be unloaded to a DPU;
establishing a shared file system between a file system path of the target file and the DPU;
the workload offloaded to the DPU accesses the target file through the shared file system.
Preferably, the establishing the shared file system is: the shared file system is established between the file synchronization client and the file synchronization server;
the file synchronization client side running on the DPU sends an access request to the file synchronization server side according to the running requirement of the workload unloaded to the DPU;
and the file synchronization server running on the host computer mounts the file system path of the target file to the local file system path of the DPU according to the access request so as to be accessed by the workload unloaded to the DPU.
Preferably, when the workload is workload kubelet, the accessing the target file through the shared file system is: target files required by the workload kubelet and stored in a host through a hostpath volume are hung on the DPU through the shared file system.
Preferably, the container group type supported by the target file includes: a device plug-in class container group, a container network plug-in class container group, and a container storage plug-in class container group.
Preferably, the file synchronization client and the file synchronization server communicate according to a network file system protocol NFS.
Preferably, the method further comprises: when the workload unloaded to the DPU accesses the target file through the shared file system, when the number of times that the file content of any appointed path under the host is accessed is larger than a set value, the DPU locally caches the file content of the appointed path.
Preferably, the method further comprises: and the file synchronization server judges whether to limit the access rights according to the type of the access request and the path rights to be accessed after receiving the access request.
An apparatus for offloading load to a DPU based on a local storage volume, the apparatus comprising: the system comprises a target path acquisition module, a shared file system module and a target file access module;
the target path acquisition module is used for acquiring a file system path of a target file of a work load to be unloaded to the DPU, which is stored in a host through a local storage volume;
the shared file system module establishes a shared file system between a file system path of the target file and the DPU;
and the target file access module enables the workload unloaded to the DPU to access the target file through the shared file system.
Preferably, the establishing the shared file system is: the shared file system is established between the file synchronization client and the file synchronization server;
the file synchronization client side running on the DPU sends an access request to the file synchronization server side according to the running requirement of the workload unloaded to the DPU;
and the file synchronization server running on the host computer mounts the file system path of the target file to the local file system path of the DPU according to the access request so as to be accessed by the workload unloaded to the DPU.
Preferably, when the workload is workload kubelet, the accessing the target file through the shared file system is: target files required by the workload kubelet and stored in a host through a hostpath volume are hung on the DPU through the shared file system.
The beneficial effects are that:
1. according to the invention, the shared file system is established to mount the file system path of the host side to the local file system path of the DPU, so that the DPU can access the file of the host side like accessing the local file through the shared file system; and when the file content is modified, the host and the DPU can realize perception at the same time, so that the problem of how to load and unload the load to the DPU hardware in the cloud prototheca platform is solved. So that the DPU can also sense and automatically synchronize files stored or modified by the local storage volumes through the containers deployed in the container cluster in real time.
2. According to the invention, the shared file system is established by adopting the interprocess communication technology, and the real-time perception of the DPU equipment originally positioned outside the k8s cluster on the files required by the cloud primary container platform load is realized by utilizing the file synchronization client side and the file synchronization server side of the shared file system, so that the difficulty in establishing the shared file system is reduced, and the flexibility in selecting the physical interface between the DPU and the host is improved.
3. The invention is particularly suitable for the application scene of unloading kubelet to the DPU by flexibly applying the local storage volume of the hostpath class, and reduces the resource consumption of the host by handing the kubelet load to the DPU for processing, thereby improving the service performance and enabling the host to have enough specific service realization required by the resource operation.
4. After the file synchronization service end receives the access request, whether to start access permission limitation is judged according to the type of the access request and the access path permission required by the current access request, namely, the security of host data is protected by setting a plurality of reasonable security options such as Access Control List (ACL), identity verification, encryption and the like.
5. According to the invention, the DPU file synchronization client locally caches the content of part of host files, and when the same files are accessed for multiple times, the DPU file synchronization client can acquire data from the cache, so that the request for the host server is reduced, and the performance of the host server is further improved.
Drawings
FIG. 1 is a general flow diagram of a method for offloading a DPU in accordance with an embodiment of the invention;
FIG. 2 is a schematic diagram of an implementation of the present invention for offloading kubelet load to a DPU, based on an embodiment;
fig. 3 is a flowchart illustrating a method for unloading load to a DPU according to an embodiment of the present invention;
fig. 4 is a schematic diagram of an apparatus for unloading load to a DPU according to an embodiment of the present invention.
Detailed Description
The invention will now be described in detail by way of example with reference to the accompanying drawings.
The invention provides a method for unloading load to a DPU based on a local storage volume, which has the following core ideas: according to the method, the problem that the load unloaded to the DPU cannot be perceived by the DPU outside the cloud primary container platform due to the fact that the load cannot work normally on the DPU and the load unloading of the DPU cannot be achieved is considered, the shared file system is established between the file system path of the target file required by the load and the DPU, the data communication between the DPU and the host is achieved through the shared file system, the file or the file catalog required by the load stored on the host through the local storage volume can be accessed by the DPU in real time, and the support is provided for the normal work of the load unloaded to the DPU.
Therefore, the invention mounts the file system path of the host side to the local file system path of the DPU by establishing the shared file system, so that the DPU can access the file of the host side like accessing the local file through the shared file system; when the file content is modified, the host and the DPU can realize perception at the same time, so that the problem of how to load and unload the load to DPU hardware in the cloud primary container platform is solved, and the DPU can also perceive and automatically synchronize files stored or modified by a container deployed in the container cluster through a local storage volume in real time. The invention reduces the resource consumption of host by giving the workload irrelevant to the service to the DPU, thereby improving the performance of the service and enabling the host to have enough service load required by the operation of the resource.
The present invention will be further described in detail with reference to the following examples.
The invention provides a method for unloading load to a DPU based on a local storage volume, which is shown in figure 1 and comprises the following steps:
s1: the file system path of the target file stored in the host through the local storage volume is obtained according to the workload offloaded to the DPU.
The local storage volume may be a storage volume type such as emptyDir, hostPath or local volume, and the workload unloaded to the DPU may be a node proxy kubelet load, or may be other workload that needs to be unloaded to the DPU and that uses the local storage volume for file storage. In some embodiments, when the workload is a workload kubelet, the file system path to obtain the target file that the workload to be offloaded to the DPU stores to the host through the local storage volume is: target files required by the workload kubelet and stored in a host through a hostpath volume are mounted on the DPU through a shared file system.
According to the work requirement of the load kubelet, the types of the container groups supported by the target file comprise: a device plug-in class container group, a container network plug-in (cni) class container group, and a container storage plug-in (csi) class container group. Depending on the type of container group, there will be a corresponding difference in the host side files required by kubelet. As shown in FIG. 2, the default path corresponding to the device plug-in class container group is/var/lib/kubrelet/device-pligins, the default path corresponding to the container network plug-in class container group is/opt/cni/bin or/etc/cni/net. D, and the default path corresponding to the container storage plug-in class container group is/usr/libexec/kubrelet-plug-ins/volume/exec.
S2: a shared file system is established between the file system path of the target file and the DPU. The method comprises the following steps:
a shared file system comprising: the file synchronization client and the file synchronization server. The file synchronization client operates on the DPU, and sends an access request to the file synchronization server according to the operating requirement of the workload unloaded to the DPU. The file synchronization server runs on a host, and according to the access request, the file system path of the target file is mounted on the local file system path of the DPU for access by the workload unloaded to the DPU. Wherein the access request comprises: read file request, write file request, list directory content request.
Through the shared file system, a target path on the host side can be hung on the DPU side according to requirements, so that a load unloaded to the DPU by kubelet and the like can sense a dp interface, a cni plug-in or other required file volume addresses under a required corresponding path.
S3: the workload offloaded to the DPU accesses the target file through the shared file system. The method comprises the following steps:
in some embodiments, when the workload is kubelet, if kubelet of the DPU needs to access the host file or directory on the specified path, the DPU sends a corresponding access request to the host server, such as reading the file, writing the file, listing the directory content, etc.; after receiving the request, the host server executes corresponding operations and transmits the data on the corresponding path back to the file synchronization client of the DPU, so as to realize the mounting of the target path and the real-time access to the target file, wherein the data transmission may involve the reading and writing of the data and the data format defined in the protocol. For a file synchronization client of a DPU, there is no distinction between accessing host shared files and accessing local files, and the workload kubelet running on the DPU can perceive access to files in real time through a specified path in the client virtual file system without knowing where the files are actually stored, and this transparency enables kubelet to be offloaded to the DPU without affecting the implementation of any function.
Therefore, the invention realizes the synchronous perception of the remote files at the host side and the DPU side through the network file protocol, so that the DPU can access the files stored at the host side in the local storage volume type such as the hostpath like accessing the local files through the file synchronous customer service side, and the problem of file perception when the load is unloaded to the DPU hardware in the cloud primary container platform is solved. When the node proxy load of the container platform is unloaded onto the DPU, the node proxy load on the DPU can acquire file catalogues with the same effect as other host nodes in the cluster; and under the condition that the deployment condition of k8s changes, the file content under the DPU can be synchronized with the deployment condition of the container cluster.
To promote the reliability of the shared file system, in some embodiments, the file synchronization client and the file synchronization server communicate in accordance with a network file system (Network File System, NFS) protocol. Different protocol versions exist in the NFS protocol, such as NFSv3, NFSv4, etc., and the NFS protocol defines specific implementations of a communication mode, a data transmission format, error handling, etc. between a file synchronization client and a file synchronization server. In some embodiments, the file synchronization client and the file synchronization server are connected through PCIE interfaces, and data transmission is performed by using an inter-process communication technology, and the file synchronization client and the file synchronization server perform communication of an access request and mount data through rpc Stub.
When the remote NFS shared directory is mounted, the file synchronization client mounts the remote file system to the local file system of the client to form a virtual directory structure, wherein the virtual directory structure is a part of the file system of the client and comprises files and directories shared on the remote file synchronization server, that is, the file synchronization client accesses the mounting path and accesses the virtual file system, then invokes the file synchronization client kernel, and the file synchronization client kernel uses the NFS protocol to communicate with the file synchronization server, which comprises encapsulating the NFS request into an NFS protocol message and transmitting the NFS message to the file synchronization server through a network, and the file synchronization server receives the request sent by the file synchronization client, parses the request and executes corresponding file system operations (virtual file system-file system), which may involve operations such as reading, writing, creating, deleting, etc. of the file; when the file synchronization server performs the requested file operation, it encapsulates the result of the operation into an NFS response message and sends the NFS response message back to the file synchronization client. The virtual file system obtained by the method provides a unified interface for the file systems of different hardware sides, so that an application program can access files and data without knowing the underlying file system.
To optimize communication efficiency and improve processing performance, in some embodiments, the file synchronization client of the DPU may locally cache the contents of a portion of the host file. Specifically, after implementing S3 that the workload accesses the target file through the shared file system, as shown in fig. 3, it further includes:
s301: judging whether the number of times of file content accessed on any appointed path in the client virtual file system is larger than a set value; if yes, continue to S302, otherwise wait for subsequent access requests to repeat S3.
S302: the file synchronization client of the DPU stores the file content of the specified path through a local cache.
It can be seen that when the same file is accessed multiple times, the file sync client of the DPU can retrieve data from the cache and forward it to the workload, thereby reducing the request to the host server. Meanwhile, data synchronization management is also needed to avoid data consistency problems possibly caused by caching.
In order to protect security of host data, in some embodiments, after the file synchronization server receives an access request, the present invention determines whether to limit access rights according to the type of the access request and the path rights to be accessed. Specifically, before S3 implementing that the workload accesses the target file through the shared file system, it further includes:
s300: and judging the access right after receiving the access request.
By setting reasonable security options such as Access Control List (ACL), authentication, encryption and the like for the file synchronization server, security of host data can be ensured. Limiting access rights includes: when the type of the access request is a read-only file and the path authority to be accessed is a write-in file, or the type of the access request is a write-in file and the path authority to be accessed is a read-in file, the file synchronization server starts the access authority to limit and suspend the access, so as to limit the access, verify the user identity and protect the privacy of the data transmission.
To avoid a malfunction after a device is restarted, in some embodiments, the file synchronization client on the DPU side will trigger an automatic mount after the device is restarted. The automatic mount attempt reestablishes communication with the file synchronization server to prevent the mount from failing after restarting to generate a functional failure. The automatic mounting is provided with an access security policy, so that only the DPU corresponding to the host has permission to access the files in the specific directory path, and the security of the data is ensured.
The invention also provides a device for unloading load to the DPU based on the local storage volume, as shown in fig. 4, the device comprises: the system comprises a target path acquisition module, a shared file system module and a target file access module.
And the target path acquisition module is used for acquiring a file system path of a target file of the workload to be unloaded to the DPU, which is stored in the host through the local storage volume.
In some embodiments, when the workload is a workload kubelet, accessing the target file through the shared file system is: target files required by the workload kubelet and stored in a host through a hostpath volume are mounted on the DPU through a shared file system. The type of the container group supported by the target file comprises: a device plug-in class container group, a container network plug-in class container group, and a container storage plug-in class container group.
And the shared file system module is used for establishing a shared file system between a file system path of the target file and the DPU.
In some embodiments, establishing the shared file system is: the shared file system is established between the file synchronization client and the file synchronization server. The file synchronization client side running on the DPU sends an access request to the file synchronization server side according to the running requirement of the workload unloaded to the DPU; and the file synchronization server side running on the host computer mounts the file system path of the target file onto the local file system path of the DPU according to the access request for the workload unloaded to the DPU to access. Wherein the access request comprises: read file request, write file request, list directory content request.
To promote the reliability of the shared file system, in some embodiments, the file synchronization client and the file synchronization server communicate in accordance with the NFS protocol. In some embodiments, the file synchronization client and the file synchronization server are connected through PCIE interfaces, and data transmission is performed by using an inter-process communication technology, and the file synchronization client and the file synchronization server perform communication of an access request and mount data through rpc Stub.
And the target file access module enables the workload unloaded to the DPU to access the target file through the shared file system.
To optimize communication efficiency and improve processing performance, in some embodiments, when a workload offloaded to the DPU accesses a target file through the shared file system, the DPU locally caches the file contents of a specified path when the number of times the file contents of any specified path under the host is accessed is greater than a set value.
To protect security of host data, in some embodiments, after receiving an access request, the file synchronization server determines whether to limit access rights according to the type of the access request and the path rights to be accessed.
To avoid a malfunction after a device is restarted, in some embodiments, the file synchronization client on the DPU side will trigger an automatic mount after the device is restarted.
In summary, the above embodiments are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (10)

1. A method of offloading load to a DPU based on a local storage volume, comprising:
obtaining a file system path of a target file which is stored in a host by a local storage volume and is required to be unloaded to a DPU;
establishing a shared file system between a file system path of the target file and the DPU;
the workload offloaded to the DPU accesses the target file through the shared file system.
2. The method of offloading loads to a DPU based on a local storage volume of claim 1, wherein establishing the shared file system is: the shared file system is established between the file synchronization client and the file synchronization server;
the file synchronization client side running on the DPU sends an access request to the file synchronization server side according to the running requirement of the workload unloaded to the DPU;
and the file synchronization server running on the host computer mounts the file system path of the target file to the local file system path of the DPU according to the access request so as to be accessed by the workload unloaded to the DPU.
3. The method of offloading loads to a DPU based on a local storage volume according to claim 1 or 2, wherein when the workload is a workload kubelet, the accessing the target file through the shared file system is: target files required by the workload kubelet and stored in a host through a hostpath volume are hung on the DPU through the shared file system.
4. The method for offloading loads to a DPU based on a local storage volume as recited in claim 3, wherein the container group type supported by the target file comprises: a device plug-in class container group, a container network plug-in class container group, and a container storage plug-in class container group.
5. The method of offloading loads to a DPU based on a local storage volume of claim 2, wherein the file synchronization client and the file synchronization server communicate in accordance with a network file system protocol NFS.
6. The method of offloading load to a DPU based on a local storage volume of claim 1 or 2, the method further comprising: when the workload unloaded to the DPU accesses the target file through the shared file system, when the number of times that the file content of any appointed path under the host is accessed is larger than a set value, the DPU locally caches the file content of the appointed path.
7. The method of offloading loads to a DPU based on a local storage volume of claim 2, further comprising: and the file synchronization server judges whether to limit the access rights according to the type of the access request and the path rights to be accessed after receiving the access request.
8. An apparatus for offloading load to a DPU based on a local storage volume, the apparatus comprising: the system comprises a target path acquisition module, a shared file system module and a target file access module;
the target path acquisition module is used for acquiring a file system path of a target file of a work load to be unloaded to the DPU, which is stored in a host through a local storage volume;
the shared file system module establishes a shared file system between a file system path of the target file and the DPU;
and the target file access module enables the workload unloaded to the DPU to access the target file through the shared file system.
9. The apparatus for offloading loads to a DPU based on a local storage volume as recited in claim 8, wherein establishing the shared file system is: the shared file system is established between the file synchronization client and the file synchronization server;
the file synchronization client side running on the DPU sends an access request to the file synchronization server side according to the running requirement of the workload unloaded to the DPU;
and the file synchronization server running on the host computer mounts the file system path of the target file to the local file system path of the DPU according to the access request so as to be accessed by the workload unloaded to the DPU.
10. The apparatus for offloading loads to a DPU based on a local storage volume according to claim 8 or 9, wherein when the workload is a workload kubelet, the accessing the target file through the shared file system is: target files required by the workload kubelet and stored in a host through a hostpath volume are hung on the DPU through the shared file system.
CN202311564539.5A 2023-11-22 2023-11-22 Method and device for unloading load to DPU based on local storage volume Pending CN117608474A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311564539.5A CN117608474A (en) 2023-11-22 2023-11-22 Method and device for unloading load to DPU based on local storage volume

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311564539.5A CN117608474A (en) 2023-11-22 2023-11-22 Method and device for unloading load to DPU based on local storage volume

Publications (1)

Publication Number Publication Date
CN117608474A true CN117608474A (en) 2024-02-27

Family

ID=89952676

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311564539.5A Pending CN117608474A (en) 2023-11-22 2023-11-22 Method and device for unloading load to DPU based on local storage volume

Country Status (1)

Country Link
CN (1) CN117608474A (en)

Similar Documents

Publication Publication Date Title
US5948062A (en) Network file server using a cached disk array storing a network file directory including file locking information and data mover computers each having file system software for shared read-write file access
US6298386B1 (en) Network file server having a message collector queue for connection and connectionless oriented protocols
US5893140A (en) File server having a file system cache and protocol for truly safe asynchronous writes
US9351153B2 (en) Apparatus for end-user transparent utilization of computational, storage, and network capacity of mobile devices, and associated methods
US5944789A (en) Network file server maintaining local caches of file directory information in data mover computers
CN110377395B (en) Pod migration method in Kubernetes cluster
EP2284725B1 (en) Client, brokerage server and method for providing cloud storage
US8601220B1 (en) Transparent data migration in a storage system environment
CN111338854B (en) Kubernetes cluster-based method and system for quickly recovering data
US8280944B2 (en) Methods, media and systems for managing a distributed application running in a plurality of digital processing devices
US7877411B1 (en) System and method for duplication of virtual private server files
CN109213571B (en) Memory sharing method, container management platform and computer readable storage medium
US20060123121A1 (en) System and method for service session management
US20020174265A1 (en) Method and apparatus for caching active computing environments
US10620871B1 (en) Storage scheme for a distributed storage system
CN106648462B (en) Date storage method and device
CN111782318A (en) Sharing access system and method for remotely mounting local disk to cloud desktop virtual machine
US20120311002A1 (en) Computer and data management method by the computer
CN112328366B (en) Efficient cloud platform host protection method and system
US20210055938A1 (en) Hydration in virtual machines
CN113961520A (en) Dynamic file mounting method, system, computer equipment and readable storage medium
US11614901B2 (en) Apparatus and method for processing sensitive data
CN111225003B (en) NFS node configuration method and device
CN117608474A (en) Method and device for unloading load to DPU based on local storage volume
US7827194B2 (en) Access to shared disk device on storage area network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication