CN115344348A - Intelligent network card virtual machine migration method based on libvirt and qemu - Google Patents
Intelligent network card virtual machine migration method based on libvirt and qemu Download PDFInfo
- Publication number
- CN115344348A CN115344348A CN202210831721.1A CN202210831721A CN115344348A CN 115344348 A CN115344348 A CN 115344348A CN 202210831721 A CN202210831721 A CN 202210831721A CN 115344348 A CN115344348 A CN 115344348A
- Authority
- CN
- China
- Prior art keywords
- qemu
- virtual machine
- network card
- intelligent network
- migration
- 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
Links
Images
Classifications
-
- 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/14—Session management
- H04L67/141—Setup of application sessions
-
- 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/4557—Distribution of virtual machine instances; Migration and load balancing
-
- 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/45583—Memory management, e.g. access or allocation
-
- 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/45595—Network integration; Enabling network access in virtual machine instances
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer And Data Communications (AREA)
Abstract
The application discloses an intelligent network card virtual machine migration method based on libvirt and qemu, which comprises the following steps: executing configuration on a source-end intelligent network card virtual machine to be migrated: configuration of an intelligent network card side: a dirty page recording module and a first Migrate-agent module; configuring a vfio migration agent module and a qmp interface on the qemu side; and executing configuration on the destination virtual machine receiving the migration: configuration of an intelligent network card side: a second Migrate-agent module; qemu side configuration: a vfio migration agent module, qmp interface. And after the configuration is completed, executing the virtual machine migration. By using the method of the embodiment of the application, the migration of the virtual machine with the intelligent network card equipment is realized based on libvirt and qemu.
Description
Technical Field
The invention relates to the technical field of computers, in particular to an intelligent network card virtual machine migration method based on libvirt and qemu.
Background
The intelligent network card is a novel network card which integrates network capacity and computing capacity, generally comprises a network interface controller and a multi-core CPU, and can be selectively configured with an FPGA or a GPU. The intelligent network card can be regarded as a programmable network card, and functions of network, storage, safety and the like are realized. Moreover, virtualization, load balancing, and data path optimization may also be implemented.
Smart cards integrate computing power, which can offload network, storage, and security functions from the host, thereby freeing up valuable CPU resources and the like. The functions that the intelligent network card can realize are many: offload network protocols include VXLAN, NVGRE, geneve protocol, and the like. And performing functions oF packet inspection, flow table processing, encryption, VXLAN coverage, NVMe-oF and the like. The intelligent network card can be used as a storage controller, is connected with source end storage, and unloads IO to the intelligent network card without the participation of a host CPU. The security aspect can realize the encryption and decryption functions of the data stream.
Libvirt is currently the most widely used tool and API to manage KVM virtual machines. Libvirtd is a daemon process and can be called by local virsh or remote virsh, and the Libvirtd calls an interface provided by qemu to operate a virtual machine.
Qemu is a set of software simulating a processor, and can realize the simulation of the whole set of computer system. The method is widely applied to the field of cloud computing at present, is combined with kvm, and can realize a high-performance virtual machine by using a hardware-assisted virtualization technology. The intelligent network card technology is an acceleration technology for unloading network and IO loads onto a network card, and is gradually widely used.
At present, the general use method of the intelligent network card is as follows: virtualizing a plurality of VF devices, and transmitting the VF devices to a virtual machine for use. For transparent transmission equipment, qemu cannot sense data change of VF, such as DMA data transmission between a virtual machine and an intelligent network card, the state of intelligent network card equipment, and the like. Thus, existing libvirt and qemu are not supported for migration of such virtual machines.
Disclosure of Invention
The embodiment of the invention provides an intelligent network card virtual machine migration method based on libvirt and qemu, which is used for realizing the migration of a virtual machine comprising intelligent network card equipment.
The embodiment of the invention provides an intelligent network card virtual machine migration method based on libvirt and qemu, which comprises the following steps:
executing the following configuration on the source-end intelligent network card virtual machine to be migrated:
configuration of an intelligent network card side:
the dirty page recording module is used for recording the dirty page information of the DMA address of the virtual machine to be migrated;
the first Migrate-agent module is used for establishing a socket monitoring port, receiving the network connection of a qemu side, transmitting the dirty page information and the equipment information of the virtual machine to be migrated to the qemu in the migration process, receiving the information of stopping the equipment after the migration is finished, and informing the intelligent network card;
qemu side configuration:
the vfio migration agent module is used for completing communication establishment and data transmission with the intelligent network card side migrate-agent;
the qmp interface receives ip and port information of the intelligent network card transmitted by libvirt;
and executing the following configuration on the destination virtual machine receiving the migration:
configuration of an intelligent network card side:
the second Migrate-agent module is used for establishing a socket monitoring port, receiving the network connection of the qemu side, receiving the equipment state and dirty page information transmitted by the qemu side in the migration process, receiving the message of starting the intelligent network card of the qemu side after the migration is finished, and starting the intelligent network card;
qemu side configuration:
the vfio migration agent module is used for completing communication establishment and data transmission with the intelligent network card side migrate-agent;
the qmp interface receives ip and port information of the intelligent network card transmitted by libvirt;
and after the configuration is completed, executing the virtual machine migration.
Optionally, the vfio migration agent module is embedded in the qemu vfio migration framework, and before executing the virtual machine migration, the dirty page recording function is started.
Optionally, executing the virtual machine migration includes the following processes:
the source-end intelligent network card virtual machine receives a virsh migrate command, performs parameter analysis, and generates an xml file of a destination-end virtual machine according to the equipment information of the source-end intelligent network card virtual machine;
the source end libvirtd transmits the xml file to the destination end libvirtd;
the target end libvirtd performs environment preparation according to the xml file, generates a qemu starting parameter, draws a qemu process, and establishes a monitoring port;
and calling a qemu qmp interface by the source end libvirtd to perform virtual machine migration.
Optionally, in the process of executing virtual machine migration, the method further includes:
the source end libvirt sends the header information and the migration configuration to a target qemu;
the source terminal qemu marks all memory blocks as dirty and submits all memory blocks to the destination terminal qemu;
and traversing all the memory blocks to transmit the memory contents.
Optionally, the performing the memory content transmission further includes:
in the process of transmitting the memory content, calculating the residual amount;
and under the condition that the residual quantity can be transmitted at one time, suspending the virtual machine, and transmitting the content of the residual memory block and the state information of the equipment to the destination terminal.
The application also provides a terminal device, which comprises a processor and a memory, wherein the memory stores a computer program, and the computer program is executed by the processor to realize the steps of the intelligent network card virtual machine migration method based on libvirt and qemu.
The application also provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the intelligent network card virtual machine migration method based on libvirt and qemu are realized.
By using the method of the embodiment of the invention, the migration of the virtual machine with the intelligent network card equipment is realized based on libvirt and qemu.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1 is a basic flowchart of an intelligent network card virtual machine migration method according to an embodiment of the present application.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
The embodiment of the invention provides an intelligent network card virtual machine migration method based on libvirt and qemu, which comprises the following steps as shown in fig. 1:
in step S101, the following configuration is executed on the source-end smart network card virtual machine to be migrated:
configuration of the source end intelligent network card side:
the dirty page recording module is configured to record dirty page information of a DMA address of the virtual machine to be migrated, specifically, after the VF is passed through to the virtual machine, the intelligent network card side has DMA information of network card operation, where the DMA information in this example includes a GPA address and a length of a dirty page.
The first Migrate-agent module is used for establishing a socket monitoring port, receiving the network connection of a qemu side, transmitting the dirty page information and the equipment information of the virtual machine to be migrated to the qemu in the migration process, receiving the information of stopping the equipment after the migration is finished, and informing the intelligent network card. After the intelligent network card is informed, the intelligent network card completes the final data packet sending.
Source qemu side configuration:
and adding a vfio migration agent module in qemu for completing communication establishment and data transmission with the smart network card side migrate-agent. Specifically, the method comprises the steps of transmitting DMA dirty page information and equipment states; and the method also comprises the steps of opening and closing the dirty page recording action and the like. Meanwhile, various interfaces for qemu equipment migration are provided, and migration actions are completed.
The qmp interface receives ip and port information of the intelligent network card transmitted by libvirt.
In step S102, the following configuration is performed at the destination virtual machine receiving the migration:
and (3) configuring the intelligent network card side of the destination:
and the second Migrate-agent module is used for establishing a socket monitoring port, receiving the network connection of the qemu side, receiving the equipment state and dirty page information transmitted by the qemu side in the migration process, receiving the message of starting the intelligent network card of the qemu side after the migration is finished, and starting the intelligent network card.
And (3) configuring a destination qemu side:
the vfio migration agent module is used for completing communication establishment and data transmission with the smart network card side migrate-agent;
a qmp interface, which receives ip and port information of the intelligent network card transmitted by libvirt;
in step S103, after the configuration is completed, the virtual machine migration is executed.
Specifically, parameters such as srcsmirtnictnip, dsrsmarstnictnip, and dstsmarstnictnip added to the virsh migrate can be used to respectively represent the ip and port information of the source-end intelligent network card and the destination-end intelligent network card.
The libvirtd sends the monitoring port information ip and port of the intelligent network card to qemu, and the qemu at the two ends can establish network connection with the intelligent network card in the migration stage.
The virtual machine migration process is then performed.
By using the method of the embodiment of the invention, the migration of the virtual machine with the intelligent network card equipment is realized based on libvirt and qemu.
In some embodiments, the vfio migration agent module is embedded in the qemu vfio migration framework, and before virtual machine migration is executed, a dirty page recording function is started. Specifically, the vfio migration agent module is embedded in the qemu vfio migration framework.
And starting a dirty page recording function before migration, acquiring information of dirty pages including GPA (general purpose address) addresses and sizes from the intelligent network card side during memory migration, and migrating the memory to a destination end during migration.
And reading and migrating the equipment state of the intelligent network card to a destination end at the final stage of migration, triggering the intelligent network card to send data outwards, sending the network data of the source end to the network, and finally stopping the intelligent network card.
And the destination terminal receives the dirty page memory, recovers the dirty page information and the equipment state of the intelligent network card side, and finally starts the intelligent network card.
In some embodiments, performing virtual machine migration includes the process of:
the source-end intelligent network card virtual machine receives a virsh migrate command, performs parameter analysis, checks the validity of migration parameters, checks whether the current qemu version supports the parameters, and generates an xml file of a destination-end virtual machine according to the device information of the source-end intelligent network card virtual machine under the condition that the parameters support.
And the source end libvirtd transmits the xml file to the destination end libvirtd.
And the target end libvirtd performs environment preparation, such as the creation of a magnetic disk device and the like, according to the xml file. And then generating a qemu starting parameter, pulling a qemu process, and establishing a monitoring port by the qemu process.
And calling a qemu qmp interface by the source libvirtual to perform virtual machine migration.
In some embodiments, in performing the virtual machine migration, the method further includes:
the source end libvirt sends the header information and the migration configuration to a target qemu;
the source terminal qemu identifies all memory blocks as dirty, delivers all memory blocks to the destination terminal qemu, specifically identifies all memory blocks as dirty, represents that all memory blocks need to be delivered, and delivers all information such as names and sizes of the memory blocks to the destination terminal qemu.
And traversing all the memory blocks to transmit the memory contents.
In some embodiments, the performing the memory content transfer further comprises:
in the process of transmitting the memory content, calculating the residual amount;
and under the condition that the remaining amount can be transmitted at one time, suspending the virtual machine, and transmitting the content of the memory block of the remaining part and the state information of the equipment to the destination terminal.
And the destination qemu receives the migration data of the memory and recovers while receiving, and when the migration is completed, the vcpu is started, and the virtual machine of the destination starts to run.
The source libvirtd drops the qemu process stop, removes the virtual machine information in the libvirtd and the like, and completes the migration work.
According to the scheme, the migration of the intelligent network card virtual machine can be completed by using a libvirt + qemu native mode, excessive adaptation of a control surface is not needed, and the compatibility is good. The method for transmitting the information of the intelligent network card by libvirt and qemu can transmit the required value according to the actual situation. The dirty page recording module and the migrate-agent module provided by the scheme of the application are not limited to individual intelligent network cards, and have universality.
The application also provides a terminal device, which comprises a processor and a memory, wherein the memory stores a computer program, and the computer program is executed by the processor to realize the steps of the intelligent network card virtual machine migration method based on libvirt and qemu.
The application also provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the intelligent network card virtual machine migration method based on libvirt and qemu are realized.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in a process, method, article, or apparatus that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present invention or portions thereof contributing to the prior art may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal (which may be a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
While the present invention has been described with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, which are illustrative and not restrictive, and it will be apparent to those skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (7)
1. An intelligent network card virtual machine migration method based on libvirt and qemu is characterized by comprising the following steps:
executing the following configuration on the source-end intelligent network card virtual machine to be migrated:
configuration of an intelligent network card side:
the dirty page recording module is used for recording the dirty page information of the DMA address of the virtual machine to be migrated;
the first Migrate-agent module is used for establishing a socket monitoring port, receiving the network connection of a qemu side, transmitting the dirty page information and the equipment information of the virtual machine to be migrated to the qemu in the migration process, receiving the information of stopping the equipment after the migration is finished, and informing the intelligent network card;
qemu side configuration:
the vfio migration agent module is used for completing communication establishment and data transmission with the intelligent network card side migrate-agent;
the qmp interface receives ip and port information of the intelligent network card transmitted by libvirt;
and executing the following configuration on the destination virtual machine receiving the migration:
configuration of an intelligent network card side:
the second Migrate-agent module is used for establishing a socket monitoring port, receiving the network connection of the qemu side, receiving the equipment state and dirty page information transmitted by the qemu side in the migration process, receiving the message of starting the intelligent network card from the qemu side after the migration is finished, and starting the intelligent network card;
qemu side configuration:
the vfio migration agent module is used for completing communication establishment and data transmission with the intelligent network card side migrate-agent;
the qmp interface receives ip and port information of the intelligent network card transmitted by libvirt;
and after the configuration is completed, executing the virtual machine migration.
2. The libvirt and qemu-based intelligent network card virtual machine migration method according to claim 1, wherein the vfio migration agent module is embedded in a qemu vfio migration frame, and before virtual machine migration is executed, a dirty page recording function is started.
3. The libvirt and qemu-based intelligent network card virtual machine migration method according to claim 2, wherein executing virtual machine migration comprises the following processes:
the source-end intelligent network card virtual machine receives a virsh migrate command, performs parameter analysis, and generates an xml file of a destination-end virtual machine according to the equipment information of the source-end intelligent network card virtual machine;
the source end libvirtd transmits the xml file to a destination end libvirtd;
the target end libvirtd performs environment preparation according to the xml file, generates qemu starting parameters, draws a qemu process, and establishes a monitoring port;
and calling a qemu qmp interface by the source libvirtual to perform virtual machine migration.
4. The libvirt and qemu-based intelligent network card virtual machine migration method according to claim 3, wherein in the process of executing virtual machine migration, the method further comprises:
the source end libvirt sends the head information and the migration configuration to a target qemu;
the source end qemu identifies all memory blocks as dirty and delivers all memory blocks to the destination end qemu;
and traversing all the memory blocks to transmit the memory contents.
5. The libvirt and qemu-based intelligent network card virtual machine migration method of claim 4, wherein the performing memory content transmission further comprises:
in the process of transmitting the memory content, calculating the residual amount;
and under the condition that the remaining amount can be transmitted at one time, suspending the virtual machine, and transmitting the content of the memory block of the remaining part and the state information of the equipment to the destination terminal.
6. A terminal device, characterized by comprising a processor and a memory, wherein the memory stores a computer program, and the computer program when executed by the processor implements the steps of the libvirt and qemu-based smart network card virtual machine migration method according to any one of claims 1 to 5.
7. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program, which when executed by a processor implements the steps of the libvirt and qemu-based intelligent network card virtual machine migration method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210831721.1A CN115344348A (en) | 2022-07-15 | 2022-07-15 | Intelligent network card virtual machine migration method based on libvirt and qemu |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210831721.1A CN115344348A (en) | 2022-07-15 | 2022-07-15 | Intelligent network card virtual machine migration method based on libvirt and qemu |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115344348A true CN115344348A (en) | 2022-11-15 |
Family
ID=83948414
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210831721.1A Pending CN115344348A (en) | 2022-07-15 | 2022-07-15 | Intelligent network card virtual machine migration method based on libvirt and qemu |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115344348A (en) |
-
2022
- 2022-07-15 CN CN202210831721.1A patent/CN115344348A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110071900B (en) | Data sending method and device | |
| US10698717B2 (en) | Accelerator virtualization method and apparatus, and centralized resource manager | |
| EP3176980B1 (en) | Unified server communication method, baseboard management controller and server | |
| EP4160424A2 (en) | Zero-copy processing | |
| CN112039722B (en) | MQTT protocol simulation method and simulation equipment | |
| CN108491278B (en) | Method and network device for processing service data | |
| CN104199716B (en) | A kind of method of live migration of virtual machine, physical host and system | |
| US9621633B2 (en) | Flow director-based low latency networking | |
| JP6195465B2 (en) | Remote card content management using synchronous server-side scripting | |
| WO2017166997A1 (en) | Inic-side exception handling method and device | |
| CN115729880A (en) | Data processing method, device, equipment and storage medium | |
| CN115103036A (en) | Efficient TCP/IP datagram processing method and system | |
| CN107277163B (en) | Equipment remote mapping method and device | |
| CN110795156B (en) | Mobile memory loading method, thin client, storage medium and device | |
| CN115344348A (en) | Intelligent network card virtual machine migration method based on libvirt and qemu | |
| CN104717175A (en) | Virtual desktop processing method and system | |
| CN115022424B (en) | Hydropower LCU controller network card virtual control method, system, equipment and medium thereof | |
| CN109039765B (en) | Network configuration method and device | |
| CN101917780B (en) | Function implementation method and device of wireless communication terminal network equipment | |
| CN113885816A (en) | Method, system, device and storage medium for printing by virtual printing device | |
| CN111669358B (en) | Method and device for processing vrouter network isolation spaces in batch | |
| CN113721995B (en) | Method and device for loading smart daughter card and storage medium | |
| CN116743587B (en) | Virtual network interface implementation method and device based on heterogeneous computing accelerator card | |
| CN111107663B (en) | Data transmission method and device, storage medium and electronic device | |
| CN114328317B (en) | Method, device and medium for improving communication performance of storage system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |