CN115296994A - Starting configuration method, device and medium for pooling heterogeneous computing resources - Google Patents

Starting configuration method, device and medium for pooling heterogeneous computing resources Download PDF

Info

Publication number
CN115296994A
CN115296994A CN202210967888.0A CN202210967888A CN115296994A CN 115296994 A CN115296994 A CN 115296994A CN 202210967888 A CN202210967888 A CN 202210967888A CN 115296994 A CN115296994 A CN 115296994A
Authority
CN
China
Prior art keywords
address
heterogeneous computing
application server
pooled
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.)
Granted
Application number
CN202210967888.0A
Other languages
Chinese (zh)
Other versions
CN115296994B (en
Inventor
黄宬
王彦伟
阚宏伟
樊嘉恒
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.)
Inspur Beijing Electronic Information Industry Co Ltd
Original Assignee
Inspur Beijing Electronic Information Industry 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 Inspur Beijing Electronic Information Industry Co Ltd filed Critical Inspur Beijing Electronic Information Industry Co Ltd
Priority to CN202210967888.0A priority Critical patent/CN115296994B/en
Publication of CN115296994A publication Critical patent/CN115296994A/en
Application granted granted Critical
Publication of CN115296994B publication Critical patent/CN115296994B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • 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
    • H04L41/0806—Configuration setting for initial configuration or provisioning, e.g. plug-and-play
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The application discloses a start configuration method, a start configuration device and a start configuration medium of pooled heterogeneous computing resources, which are applied to the technical field of communication. If the IP address needs to be allocated after the pooled heterogeneous computing resources are started, the self heterogeneous computing resource information and the IP address request information are simultaneously broadcasted to the application server. And the application server allocates the IP address after receiving the IP address request information, and one IP address is selected as a target IP address by the pooling heterogeneous computing resource. And then adding the target IP address and the IP address of the corresponding application server into an option field in the DHCP Request message by the pooling heterogeneous computing resource, and broadcasting the target IP address and the IP address of the corresponding application server to the application server so that the corresponding application server can return a DHCP ACK message. Compared with the original scheme, the method and the device have the advantages that the heterogeneous computing resource information and the IP address request information of the device are broadcast at the same time, and the starting efficiency of the pooled heterogeneous computing resources is improved.

Description

Starting configuration method, device and medium for pooling heterogeneous computing resources
Technical Field
The present application relates to the field of communications technologies, and in particular, to a method, an apparatus, and a medium for starting configuration of pooled heterogeneous computing resources.
Background
Compared with the traditional heterogeneous computing, the pooling heterogeneous computing resources of the pooling heterogeneous computing belong to a network connection state, are not attached to a certain server, are managed in a centralized manner, and are connected with the server through the communication capability of the pooling heterogeneous computing resources, instead of interfaces on a mainboard of a high-speed serial computer extended bus standard (PCIE). When the pooled heterogeneous computing resources are started, corresponding Internet Protocol (IP) addresses need to be configured, and then the pooled heterogeneous computing resources communicate with an application server, and the specific information of the pooled heterogeneous computing resources is interacted, including: the type, model, current state (e.g., application logic), etc. of the pooled heterogeneous computing resources. FIG. 1 is a flow diagram illustrating an exemplary method for implementing boot configuration by interacting a pooled heterogeneous computing resource with a server; as shown in fig. 1, the method comprises the following steps: s10: broadcasting application to obtain an IP address; s11: allocating an IP address; s12: broadcasting heterogeneous computing resource information; s13: the application server replies; s14: a heterogeneous computing request; s15: and returning the heterogeneous computation. It can be seen that, after the current pooled heterogeneous computing resources are started, an IP address request is initiated, a Dynamic Host Configuration Protocol (DHCP) server responds to the request to allocate an IP address of the pooled heterogeneous computing resources, after the pooled heterogeneous computing resources acquire the IP address, the pooled heterogeneous computing resources send their own heterogeneous computing resource information to an application server, and the application server performs related applications according to the heterogeneous computing resource information.
However, in the local area network, after the IP address is obtained by the DHCP server, the application server is notified of the heterogeneous computing resource information by broadcasting, and two times of broadcasting occupy unnecessary system bandwidth, thereby prolonging the time for starting the heterogeneous pooling resource and reducing the starting efficiency of the pooling heterogeneous computing resource.
Therefore, how to improve the starting efficiency of the pooled heterogeneous computing resources is an urgent problem to be solved by those skilled in the art.
Disclosure of Invention
The application aims to provide a start configuration method, a start configuration device and a start configuration medium of pooled heterogeneous computing resources so as to improve the start efficiency of the pooled heterogeneous computing resources.
In order to solve the above technical problem, the present application provides a method for starting configuration of pooled heterogeneous computing resources, including:
after the pooling heterogeneous computing resource is started, if the IP address needs to be allocated, broadcasting own heterogeneous computing resource information and IP address request information to the application server at the same time; the application server receives the IP address request information and then allocates an IP address to the pooling heterogeneous computing resource;
selecting an IP address returned by the application server as a target IP address;
and adding the target IP address and the corresponding IP address of the application server into an option field in a DHCP Request message, and broadcasting the DHCP Request message to the application server so as to facilitate the application server corresponding to the target IP address to return a DHCP ACK message.
Preferably, after receiving the broadcasted DHCP Request message, the other application servers except the application server corresponding to the target IP address clear their own IP address allocation records.
Preferably, after the pooled heterogeneous computing resource successfully acquires the target IP address, the method further includes:
if the target IP address needs to be released by the pooling heterogeneous computing resource, sending a DHCP Release message to the corresponding application server; and after receiving the DHCP Release message, the application server recovers the corresponding target IP address.
Preferably, the selecting an IP address returned by the application server as a target IP address includes:
and selecting the IP address returned by the application server which is received firstly as the target IP address.
Preferably, after the pooled heterogeneous computing resource receives the DHCP ACK packet, the method further includes:
and if the target IP address distributed by the application server is used, sending a DHCP Decline message to the corresponding application server to inform the application server to forbid the corresponding target IP address.
Preferably, the option field of the DHCP ACK packet returned by the application server includes the use lease information of the target IP address.
Preferably, if the IP address of the pooled heterogeneous computing resource is already applied for or is a fixed IP address, it is determined that the IP address does not need to be allocated.
In order to solve the foregoing technical problem, the present application further provides a device for starting and configuring pooled heterogeneous computing resources, including:
the first broadcasting module is used for broadcasting self heterogeneous computing resource information and IP address request information to the application server simultaneously if the first broadcasting module needs to allocate an IP address after the start of the pooled heterogeneous computing resources; the application server receives the IP address request information and then allocates an IP address to the pooling heterogeneous computing resource;
the selection module is used for selecting one IP address returned by the application server as a target IP address;
and the second broadcasting module is used for adding the target IP address and the corresponding IP address of the application server into an option field in a DHCP Request message, and broadcasting the DHCP Request message to the application server so as to facilitate the application server corresponding to the target IP address to return a DHCP ACK message.
Preferably, the apparatus for starting configuration of the pooled heterogeneous computing resources further comprises: the first sending module is used for sending a DHCP Release message to the corresponding application server if the target IP address needs to be released by the pooling heterogeneous computing resource after the target IP address is successfully obtained by the pooling heterogeneous computing resource; and after receiving the DHCP Release message, the application server recovers the corresponding target IP address.
Preferably, the apparatus for starting and configuring the pooled heterogeneous computing resources further comprises: a second sending module, configured to, after the pooled heterogeneous computing resource receives the DHCP ACK packet, send a DHCP Decline packet to the corresponding application server to notify the application server to disable the corresponding target IP address if the target IP address allocated by the application server is used.
In order to solve the above technical problem, the present application further provides a device for starting and configuring pooled heterogeneous computing resources, including: a memory for storing a computer program;
and the processor is used for realizing the steps of the starting configuration method of the pooling heterogeneous computing resources when executing the computer program.
In order to solve the above technical problem, the present application further provides a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the computer program implements the steps of the foregoing method for starting configuration of pooled heterogeneous computing resources.
According to the starting configuration method of the pooling heterogeneous computing resource, after the pooling heterogeneous computing resource is started, the pooling heterogeneous computing resource judges whether the pooling heterogeneous computing resource needs to be allocated with an IP address or not, and if the pooling heterogeneous computing resource needs to be allocated with the IP address, the pooling heterogeneous computing resource simultaneously broadcasts the self heterogeneous computing resource information and the IP address request information to an application server; according to the scheme, the original DHCP server and the original application server are combined, one application server is adopted to complete corresponding functions, and the application server receives IP address request information and then allocates IP addresses to the pooling heterogeneous computing resources. The method comprises the steps that a plurality of common application servers are provided, a plurality of returned IP addresses are provided, one IP address returned by the application server is selected by the pooling heterogeneous computing resource to serve as a target IP address, then the target IP address and the IP address of the corresponding application server are added into an option field in a DHCP Request message by the pooling heterogeneous computing resource, and the DHCP Request message is broadcasted to the application server, so that the application server corresponding to the target IP address can return a DHCP ACK message. If the corresponding target IP address is not occupied, after the pooling heterogeneous computing resource receives the DHCP ACK message, representing that the target IP address is successfully obtained, and completing the starting configuration. Compared with the method that after the IP address is obtained, the heterogeneous computing resource information of the application server is broadcasted and notified, the pooling heterogeneous computing resource simultaneously broadcasts the heterogeneous computing resource information of the pooling heterogeneous computing resource and the IP address request information, redundant system bandwidth cannot be occupied by single broadcasting, starting time of the pooling heterogeneous computing resource is shortened, and starting efficiency of the pooling heterogeneous computing resource is greatly improved.
The application also provides a starting configuration device and a computer readable storage medium of the pooling heterogeneous computing resource, which correspond to the method, so that the method has the same beneficial effects as the method.
Drawings
In order to more clearly illustrate the embodiments of the present application, the drawings needed for the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings can be obtained by those skilled in the art without inventive effort.
FIG. 1 is a flow diagram illustrating an exemplary method for implementing boot configuration by interacting a pooled heterogeneous computing resource with a server;
fig. 2 is a schematic structural diagram of a connection between a server and an FPGA pool according to an embodiment of the present disclosure;
fig. 3 is a flowchart of a method for starting configuration of pooled heterogeneous computing resources according to an embodiment of the present application;
fig. 4 is a flowchart illustrating a method for interacting a pooled heterogeneous computing resource with a server to complete a boot configuration according to an embodiment of the present application;
fig. 5 is a block diagram of an apparatus for starting configuration of pooled heterogeneous computing resources according to an embodiment of the present application;
fig. 6 is a block diagram of an apparatus for configuring startup of a pooled heterogeneous computing resource according to another embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present application without any creative effort belong to the protection scope of the present application.
The core of the application is to provide a method, a device and a medium for starting and configuring pooled heterogeneous computing resources, so as to improve the starting efficiency of the pooled heterogeneous computing resources.
In order that those skilled in the art will better understand the disclosure, the following detailed description will be given with reference to the accompanying drawings.
Heterogeneous computing translates heterogeneous operations, which mainly refers to a computing method of a system composed of computing units using different types of instruction sets and architectures, and common computing unit categories include: a Central Processing Unit (CPU) of x86 architecture, a coprocessor such as a Graphics Processing Unit (GPU), a Field Programmable Gate Array (FPGA), a Digital Signal Processor (DSP), and an Application Specific Integrated Circuit (ASIC). Heterogeneous computing has gained more attention in recent years, primarily because traditional ways of increasing computing power by increasing CPU clock frequency and number of cores have met thermal dissipation and power consumption bottlenecks. Meanwhile, although the working frequency of special computing units such as the GPU is low, the special computing units have more kernel numbers and parallel computing capability, and the overall performance-chip area ratio and the performance-power consumption ratio are high, the special computing units are far from being fully utilized. The heterogeneous accelerator virtualization technology is a very popular research field in recent years, and for cloud computing manufacturers, if virtualized computing resources can be further pooled, similar to storage pooling, a lot of benefits will be brought, namely, a more flexible heterogeneous accelerator resource configuration is provided, and in a physical machine, the requirements of 1 CPU on which 1 GPU or 16 GPUs are hung, and 64GB memory or 512GB memory on different workloads are usually inconsistent. For cloud vendors, an unflustered solution may require frequent installation to implement in the face of different preferences of different customers. And secondly, the GPU resource utilization rate is higher, a plurality of tasks such as machine learning reasoning tasks do not need the whole GPU card, the GPU card is idle in time and space due to different time periods and different using modes, and the possibility of oversale of the GPU can be provided through GPU virtualization or pooling. And thirdly, the GPU resource management is more convenient, the GPU resource monitoring is difficult due to the server model, the GPU equipment model, the resource fragment of computing power and the like, and the difficulty is expected to be reduced through pooling centralized management. The pooled heterogeneous computing resources are heterogeneous accelerators that directly interact with the host application via the network to perform complex operations, such as: pooled GPUs, pooled FPGAs, etc., and the host application is an application run by the CPU on the server. In addition, a Dynamic Host Configuration Protocol (DHCP) is a network Protocol of a local area network, and means that a server controls a segment of IP address range, a client (pooled heterogeneous computing resources) can automatically obtain an IP address and a subnet mask allocated by the server when logging in the server, the client performs broadcasting after logging in, and the DHCP server allocates a relative IP address to the client after receiving a message.
Fig. 2 is a schematic structural diagram of a connection between a server and an FPGA pool according to an embodiment of the present disclosure; as shown in fig. 2, the system comprises a server 1 and an FPGA pool 2, where the FPGA pool 2 establishes communication connections with a plurality of servers 1, that is, all the pooled heterogeneous computing resources belong to a network connection state, are not attached to a certain server, but are managed in a centralized manner, and are connected to the servers through their own communication capabilities. At present, because the DHCP protocol is independent of the process of heterogeneous computing resource information interaction, that is, after the DHCP server obtains the IP address, the application server broadcasts the heterogeneous computing resource information of the application server itself, which may result in low starting efficiency of the pooled heterogeneous computing resources. Therefore, the embodiment of the present application provides a method for starting and configuring pooled heterogeneous computing resources, which self-defines a communication protocol, and requests an IP address while reporting information of the heterogeneous computing resources. Fig. 3 is a flowchart of a method for starting configuration of a pooled heterogeneous computing resource according to an embodiment of the present application; as shown in fig. 3, the method comprises the steps of:
s20: after the pooling heterogeneous computing resource is started, if the IP address needs to be allocated, the heterogeneous computing resource information and the IP address request information of the application server are broadcasted simultaneously.
And the application server receives the IP address request information and then allocates the IP address to the pooling heterogeneous computing resource. According to the method for reporting and combining the IP address request information and the heterogeneous computing resource information, the starting configuration performance is improved, namely, the original DHCP server and the original application server are combined. When configuration is started, the pooling heterogeneous computing resource broadcasts own heterogeneous computing resource information and simultaneously requests an IP address (if the IP address needs to be allocated), and the application server performs related application according to the heterogeneous computing resource information; in addition, since the pooled heterogeneous computing resources may have applied for an IP address or may be assigned a fixed IP address, the pooled heterogeneous computing resources do not necessarily need to apply for an IP address. Therefore, after the pooled heterogeneous computing resources are started, the fact that the IP addresses need to be allocated is judged, if the IP addresses do not need to be allocated, the heterogeneous computing resource information of the pooled heterogeneous computing resources is only broadcasted to the application server, and if the IP addresses need to be allocated, the heterogeneous computing resource information and the IP address request information of the pooled heterogeneous computing resources are broadcasted to the application server at the same time; the application server allocates an IP address to the pooled heterogeneous computing resources after receiving the IP address request message. When there is more than one application server, only one application server in the local area network can successfully allocate the IP address, but other application servers also need to record the heterogeneous computing resource information.
S21: and selecting one IP address returned by the application server as a target IP address.
If only one application server exists, the corresponding IP address is directly used as the target IP address. If a plurality of application servers exist, all the application servers can receive IP address request information broadcasted by the pooling heterogeneous computing resource, each application server can allocate an IP address and return the IP address to the pooling heterogeneous computing resource, and the pooling heterogeneous computing resource selects one IP address returned by the application server as a target IP address. The specific selection method is not limited, and the selection may be performed according to the order of receiving the IP addresses, and the general principle is to select the IP address received first as the target IP address.
S22: and adding a target IP address and the IP address of the corresponding application server into an option field in the DHCP Request message, and broadcasting the DHCP Request message to the application server.
So that the application server corresponding to the target IP address returns the DHCP ACK message. Adding a target IP address and a corresponding IP address of an application server into an option field in a DHCP Request message by using the pooling heterogeneous computing resource, judging whether the IP address in the option field is the same as the address of the application server after the application server receives the DHCP Request message, and only clearing a corresponding IP address allocation record without any processing by the application server if the IP address in the option field is not the same as the address of the application server; if the two messages are the same, the application server responds a DHCP ACK message to the pooling heterogeneous computing resource and can add the use lease information of the IP address in the option field. After receiving the DHCP ACK message, the pooled heterogeneous computing resources check whether the IP address allocated by the application server can be used. If the IP address can be used, the pooling heterogeneous computing resource successfully obtains the IP address and automatically starts a continuation process according to the use lease of the IP address; if the pooling heterogeneous computing resource finds that the allocated IP address is already used, the pooling heterogeneous computing resource can send a DHCP Decline message to the application server to inform the application server to disable the IP address, and then the pooling heterogeneous computing resource starts a new IP address application process. After the pooling heterogeneous computing resource successfully acquires the IP address, the IP address of the pooling heterogeneous computing resource can be released by sending a DHCP Release message at any time, and after receiving the DHCP Release message, the application server can recover and redistribute the corresponding IP address.
FIG. 4 is a flowchart illustrating a method for implementing boot configuration by interacting a pooled heterogeneous computing resource with a server according to an embodiment of the present application; as shown in fig. 4, the method includes the steps of: s30: broadcasting heterogeneous computing resource information and IP address request information; s31: performing heterogeneous computing resource information reply and allocating an IP address; s32: a heterogeneous computing request; s33: and returning the heterogeneous computation. Fig. 4 is only one example provided in this embodiment, and does not limit other solutions. As can be seen from comparing fig. 4 and fig. 1, the scheme provided by the embodiment of the present application is simpler than the current scheme. In this embodiment, all application servers start a customized version of DHCP service, the pooled heterogeneous computing resources broadcast heterogeneous computing resource information and IP address request information, all relevant application servers simultaneously record the heterogeneous computing resource information and return an IP address to the pooled heterogeneous computing resources, and after the heterogeneous computing resource information successfully obtains the IP address, heterogeneous computing can be performed through the application servers.
According to the starting configuration method of the pooling heterogeneous computing resources, after the pooling heterogeneous computing resources are started, the pooling heterogeneous computing resources firstly judge whether the pooling heterogeneous computing resources need to allocate IP addresses, and if the pooling heterogeneous computing resources need to allocate the IP addresses, heterogeneous computing resource information and IP address request information of the pooling heterogeneous computing resources are broadcasted to an application server at the same time; according to the scheme, the original DHCP server and the original application server are combined, one application server is adopted to complete corresponding functions, and the application server receives IP address request information and then allocates IP addresses to the pooling heterogeneous computing resources. The method comprises the steps that a plurality of common application servers are provided, a plurality of returned IP addresses are provided, one IP address returned by the application server is selected by the pooling heterogeneous computing resource to serve as a target IP address, then the target IP address and the IP address of the corresponding application server are added into an option field in a DHCP Request message by the pooling heterogeneous computing resource, and the DHCP Request message is broadcasted to the application server, so that the application server corresponding to the target IP address can return a DHCP ACK message. If the corresponding target IP address is not occupied, the pooling heterogeneous computing resource receives the DHCP ACK message and represents that the target IP address is successfully obtained, and the starting configuration is completed. Compared with the method that after the IP address is obtained, the heterogeneous computing resource information of the application server is broadcasted and notified, the pooling heterogeneous computing resource simultaneously broadcasts the heterogeneous computing resource information and the IP address request information, redundant system bandwidth cannot be occupied by single broadcast, starting time of the pooling heterogeneous computing resource is shortened, and starting efficiency of the pooling heterogeneous computing resource is greatly improved.
If a plurality of application servers exist in the local area network, the pooling heterogeneous computing resource receives IP addresses returned by the application servers, only one of the IP addresses returned by the application servers needs to be selected as a target IP address, the application server corresponding to the target IP address receives the broadcasted DHCP Request message and then processes the message according to a normal flow, and the other application servers except the application server corresponding to the target IP address need to clear the IP address distribution records of the application servers after receiving the broadcasted DHCP Request message so as to continuously distribute and use the IP addresses. However, the application server will record complete heterogeneous computing resource information no matter whether the IP address in the DHCP Request message is the IP address corresponding to itself or not.
The target IP address is obtained successfully from the pooling heterogeneous computing resource, and after the target IP address is used up, the target IP address is possibly not needed any more, at the moment, the corresponding target IP address needs to be released, and a DHCP Release message can be sent to the corresponding application server; after receiving the DHCP Release message, the application server recovers the corresponding target IP address and redistributes the target IP address.
In practical application, when a plurality of application servers generally exist in a local area network, the pooled heterogeneous computing resources receive the IP addresses returned by the application servers, one of the IP addresses returned by the application servers needs to be selected as a target IP address, and the IP address returned by the application server which is received first in the selection process is used as the target IP address. Specifically, the IP address requesting process is as follows, the pooled heterogeneous computing resources send a DHCP Discover message in a broadcast manner, all application servers can receive the DHCP Discover message sent by the pooled heterogeneous computing resources, and all application servers send a DHCP Offer message to the pooled heterogeneous computing resources in response. The "round (Client) IP Address" field in the DHCP Offer message is an IP Address that the application server can provide to the pooled heterogeneous computing resources for use, and the application server will put its own IP Address in the "option" field to distinguish different application servers with the pooled heterogeneous computing resources, and the application server will have a record of the allocated IP Address after sending out the message. The pooled heterogeneous computing resources can only process one DHCP Offer message, and the general principle is to process the DHCP Offer message received firstly, namely, the IP address returned by the application server received firstly is selected as the target IP address, so that the processing efficiency is higher.
In actual application, the option field of the DHCP ACK message returned by the application server may include the use lease information of the target IP address, so that the pooled heterogeneous computing resources automatically start a renewal process according to the use lease of the IP address. In specific implementation, after the pooled heterogeneous computing resource receives the DHCP ACK message, the target IP address allocated by the application server may be already used, and the pooled heterogeneous computing resource cannot use the IP address any more, and needs to re-allocate an IP address, and send a DHCP Decline message to the corresponding application server to notify the application server to disable the corresponding target IP address, so as to prevent the application server from continuously allocating the IP address and causing the IP address of the pooled heterogeneous computing resource to be unusable.
In particular implementations, since the pooled heterogeneous computing resources may have applied for an IP address or may be assigned a fixed IP address, the pooled heterogeneous computing resources do not necessarily need to apply for an IP address. Therefore, in the solution provided in this embodiment, if the IP address of the pooled heterogeneous computing resource has been applied for or is a fixed IP address, it is determined that it is not necessary to allocate an IP address. After the pooling heterogeneous computing resource is started, if the IP address does not need to be allocated, only the self heterogeneous computing resource information is broadcasted to the application server, and if the IP address needs to be allocated, the self heterogeneous computing resource information and the IP address request information are broadcasted to the application server at the same time; the application server allocates an IP address to the pooled heterogeneous computing resources after receiving the IP address request message.
The broadcast request initiated by the pooled heterogeneous computing resources comprises: the type and model of the pooled heterogeneous computing resources, whether to allocate an IP address, the current state, and a list of supported heterogeneous applications. And when the IP address needs to be allocated, the IP address allocation identification position is true. The message format is as follows:
table 1 message format table for pooled heterogeneous computing resources
Figure BDA0003795454040000101
Options (variable) and above are standard DHCP protocol headers, and a custom optimization part protocol is added behind the standard protocol headers. Wherein: htype represents a hardware type, model represents a Model, isIP represents whether an IP address is allocated or not, status represents the running state of the current heterogeneous computing resource, sappN represents the number of supported heterogeneous applications, and SappId represents an application ID specifically supported.
In the foregoing embodiment, a detailed description is given to a start-up configuration method for pooled heterogeneous computing resources, and the present application also provides an embodiment corresponding to a start-up configuration apparatus for pooled heterogeneous computing resources. It should be noted that the present application describes the embodiments of the apparatus portion from two perspectives, one from the perspective of the function module and the other from the perspective of the hardware.
Based on the angle of the function module, this embodiment provides a starting configuration device of pooled heterogeneous computing resources, and fig. 5 is a structural diagram of the starting configuration device of pooled heterogeneous computing resources provided in this embodiment of the present application, as shown in fig. 5, the device includes:
the first broadcasting module 10 is configured to, after the pooled heterogeneous computing resources are started, broadcast heterogeneous computing resource information and IP address request information of the first broadcasting module to the application server at the same time if the first broadcasting module needs to allocate an IP address; the application server receives the IP address request information and then allocates an IP address to the pooling heterogeneous computing resource;
the selection module 11 is used for selecting an IP address returned by the application server as a target IP address;
the second broadcasting module 12 is configured to add the target IP address and the IP address of the corresponding application server into the option field in the DHCP Request message, and broadcast the DHCP Request message to the application server, so that the application server corresponding to the target IP address returns a DHCP ACK message.
Since the embodiments of the apparatus portion and the method portion correspond to each other, please refer to the description of the embodiments of the method portion for the embodiments of the apparatus portion, which is not repeated here.
As a preferred embodiment, the apparatus for starting configuration of a pooled heterogeneous computing resource further comprises: the first sending module is used for sending a DHCP Release message to a corresponding application server if the target IP address needs to be released by the pooling heterogeneous computing resource after the pooling heterogeneous computing resource successfully obtains the target IP address; and after receiving the DHCP Release message, the application server recovers the corresponding target IP address.
And the second sending module is used for sending a DHCP Decline message to the corresponding application server to inform the application server to disable the corresponding target IP address if the target IP address allocated by the application server is used after the pooling heterogeneous computing resource receives the DHCP ACK message.
The startup configuration device for pooled heterogeneous computing resources provided by this embodiment corresponds to the above method, and therefore has the same advantages as the above method.
Based on a hardware angle, this embodiment provides another starting configuration device for pooled heterogeneous computing resources, and fig. 6 is a structural diagram of the starting configuration device for pooled heterogeneous computing resources provided in another embodiment of this application, and as shown in fig. 6, the starting configuration device for pooled heterogeneous computing resources includes: a memory 20 for storing a computer program;
a processor 21 for implementing the steps of the method for start-up configuration of pooled heterogeneous computing resources as mentioned in the above embodiments when executing the computer program.
The processor 21 may include one or more processing cores, such as a 4-core processor, an 8-core processor, and the like. The processor 21 may be implemented in at least one hardware form of DSP, FPGA, programmable Logic Array (PLA). The processor 21 may also include a main processor and a coprocessor, where the main processor is a processor, also called a CPU, for processing data in an awake state; a coprocessor is a low power processor for processing data in a standby state. In some embodiments, the processor 21 may be integrated with a GPU, which is responsible for rendering and drawing the content that the display screen needs to display. In some embodiments, the processor 21 may further include an Artificial Intelligence (AI) processor for processing computational operations related to machine learning.
The memory 20 may include one or more computer-readable storage media, which may be non-transitory. Memory 20 may also include high speed random access memory, as well as non-volatile memory, such as one or more magnetic disk storage devices, flash memory storage devices. In this embodiment, the memory 20 is at least used for storing the following computer program 201, wherein after being loaded and executed by the processor 21, the computer program is capable of implementing the relevant steps of the startup configuration method for pooled heterogeneous computing resources disclosed in any of the foregoing embodiments. In addition, the resources stored in the memory 20 may also include an operating system 202, data 203, and the like, and the storage manner may be a transient storage manner or a permanent storage manner. Operating system 202 may include, among others, windows, unix, linux, and the like. Data 203 may include, but is not limited to, data related to a boot configuration method of pooled heterogeneous computing resources, and the like.
In some embodiments, the means for configuring the pool of heterogeneous computing resources may further include a display 22, an input/output interface 23, a communication interface 24, a power supply 25, and a communication bus 26.
Those skilled in the art will appreciate that the architecture shown in FIG. 6 does not constitute a limitation on the means for initiating configuration of pooled heterogeneous computing resources and may include more or fewer components than those shown.
The starting configuration device for the pooled heterogeneous computing resources provided by the embodiment of the application comprises a memory and a processor, wherein when the processor executes a program stored in the memory, the following method can be realized: a method for starting and configuring heterogeneous computing resources in a pool.
The startup configuration device for pooled heterogeneous computing resources provided by this embodiment corresponds to the above method, and therefore has the same advantages as the above method.
Finally, the application also provides a corresponding embodiment of the computer readable storage medium. The computer-readable storage medium has stored thereon a computer program which, when being executed by a processor, carries out the steps as set forth in the above-mentioned method embodiments.
It is to be understood that if the method in the above embodiments is implemented in the form of software functional units and sold or used as a stand-alone product, it can be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be substantially or partially implemented in the form of a software product, which is stored in a storage medium and performs all or part of the steps of the methods described in the embodiments of the present application, or all or part of the technical solution. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The computer-readable storage medium provided by the embodiment corresponds to the method, and therefore has the same beneficial effects as the method.
The foregoing describes in detail a method, an apparatus, and a medium for startup configuration of pooled heterogeneous computing resources provided by the present application. The embodiments are described in a progressive mode in the specification, the emphasis of each embodiment is on the difference from the other embodiments, and the same and similar parts among the embodiments can be referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description. It should be noted that, for those skilled in the art, without departing from the principle of the present application, the present application can also make several improvements and modifications, and those improvements and modifications also fall into the protection scope of the claims of the present application.
It should also be noted that, in this specification, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, 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 phrases "comprising a component of' 8230; \8230;" does not exclude the presence of additional like elements in a process, method, article, or apparatus that comprises the same element.

Claims (10)

1. A method for startup configuration of pooled heterogeneous computing resources, comprising:
after the pooling heterogeneous computing resource is started, if the IP address needs to be allocated, broadcasting own heterogeneous computing resource information and IP address request information to the application server at the same time; the application server receives the IP address request information and then allocates an IP address to the pooling heterogeneous computing resource;
selecting an IP address returned by the application server as a target IP address;
and adding the target IP address and the corresponding IP address of the application server into an option field in a DHCP Request message, and broadcasting the DHCP Request message to the application server so that the application server corresponding to the target IP address returns a DHCP ACK message.
2. The method for starting and configuring pooling heterogeneous computing resources of claim 1, wherein the other application servers except the application server corresponding to the target IP address clear their own IP address allocation records after receiving the broadcasted DHCP Request message.
3. The method for startup configuration of pooled heterogeneous computing resources according to claim 2, further comprising, after said pooled heterogeneous computing resources successfully obtain said target IP address:
if the target IP address needs to be released by the pooling heterogeneous computing resource, sending a DHCP Release message to the corresponding application server; and after receiving the DHCP Release message, the application server recovers the corresponding target IP address.
4. The method for startup configuration of pooled heterogeneous computing resources according to claim 3, wherein said selecting an IP address returned by said application server as a target IP address comprises:
and selecting the IP address returned by the application server which is received firstly as the target IP address.
5. The method for starting configuration of pooled heterogeneous computing resources according to claim 4, further comprising, after the pooled heterogeneous computing resources receive the DHCP ACK packet:
if the target IP address allocated by the application server is used, sending a DHCP Decline message to the corresponding application server to inform the application server to disable the corresponding target IP address.
6. The method for startup configuration of pooled heterogeneous computing resources according to claim 1, wherein an option field of the DHCP ACK packet returned by the application server includes use lease information of the target IP address.
7. The method for startup configuration of pooled heterogeneous computing resources according to claim 1, wherein if the IP address of the pooled heterogeneous computing resource has been applied for or is a fixed IP address for allocation, it is determined that it is not necessary to allocate an IP address.
8. An apparatus for boot configuration of pooled heterogeneous computing resources, comprising:
the first broadcasting module is used for broadcasting self heterogeneous computing resource information and IP address request information to the application server simultaneously if the first broadcasting module needs to allocate an IP address after the start of the pooled heterogeneous computing resources; the application server receives the IP address request information and then allocates an IP address to the pooling heterogeneous computing resource;
the selection module is used for selecting one IP address returned by the application server as a target IP address;
and the second broadcasting module is used for adding the target IP address and the corresponding IP address of the application server into an option field in a DHCP Request message, and broadcasting the DHCP Request message to the application server so as to facilitate the application server corresponding to the target IP address to return a DHCP ACK message.
9. An apparatus for boot configuration of pooled heterogeneous computing resources comprising a memory for storing a computer program;
a processor for implementing the steps of the method for startup configuration of pooled heterogeneous computing resources according to any one of claims 1 to 7 when said computer program is executed.
10. A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method for startup configuration of a pooled heterogeneous computing resource according to any one of claims 1 to 7.
CN202210967888.0A 2022-08-12 2022-08-12 Starting configuration method, device and medium for pooled heterogeneous computing resources Active CN115296994B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210967888.0A CN115296994B (en) 2022-08-12 2022-08-12 Starting configuration method, device and medium for pooled heterogeneous computing resources

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210967888.0A CN115296994B (en) 2022-08-12 2022-08-12 Starting configuration method, device and medium for pooled heterogeneous computing resources

Publications (2)

Publication Number Publication Date
CN115296994A true CN115296994A (en) 2022-11-04
CN115296994B CN115296994B (en) 2023-07-14

Family

ID=83828442

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210967888.0A Active CN115296994B (en) 2022-08-12 2022-08-12 Starting configuration method, device and medium for pooled heterogeneous computing resources

Country Status (1)

Country Link
CN (1) CN115296994B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115964167A (en) * 2022-12-16 2023-04-14 摩尔线程智能科技(北京)有限责任公司 Resource pooling method, apparatus, device, medium, and product for heterogeneous computing platforms

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108429824A (en) * 2018-05-31 2018-08-21 新华三技术有限公司 A kind of address distribution method and device
CN109862134A (en) * 2019-03-18 2019-06-07 中国联合网络通信集团有限公司 A kind of the lease time configuration method and system of IP address

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108429824A (en) * 2018-05-31 2018-08-21 新华三技术有限公司 A kind of address distribution method and device
CN109862134A (en) * 2019-03-18 2019-06-07 中国联合网络通信集团有限公司 A kind of the lease time configuration method and system of IP address

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115964167A (en) * 2022-12-16 2023-04-14 摩尔线程智能科技(北京)有限责任公司 Resource pooling method, apparatus, device, medium, and product for heterogeneous computing platforms
CN115964167B (en) * 2022-12-16 2023-09-01 摩尔线程智能科技(北京)有限责任公司 Resource pooling method, device, equipment, medium and product of heterogeneous computing platform

Also Published As

Publication number Publication date
CN115296994B (en) 2023-07-14

Similar Documents

Publication Publication Date Title
US8762999B2 (en) Guest-initiated resource allocation request based on comparison of host hardware information and projected workload requirement
CN108293041B (en) Distributed system, resource container allocation method, resource manager and application controller
WO2015196931A1 (en) Disk io-based virtual resource allocation method and device
US20220004410A1 (en) Method For Deploying Virtual Machine And Container, And Related Apparatus
US20130173804A1 (en) System and Method for Providing Effective Resource Reusability Management in a Virtualized Environment
CN111314799A (en) Terminal system architecture, communication system, communication method, and storage medium
JP2013513174A (en) Method and system for managing virtual machine storage space and physical hosts
CN111857956B (en) Virtual machine starting method and equipment
CN103810015A (en) Virtual machine establishing method and equipment
CN113312142A (en) Virtualization processing system, method, device and equipment
US20200272526A1 (en) Methods and systems for automated scaling of computing clusters
CN115296994A (en) Starting configuration method, device and medium for pooling heterogeneous computing resources
CN109600439B (en) PaaS platform and deployment method thereof based on microservice
CN111124437A (en) Bare computer resource configuration information acquisition method, system and equipment based on OpenStack
CN110636149B (en) Remote access method, device, router and storage medium
US10592221B2 (en) Parallel distribution of application services to virtual nodes
CN115576654B (en) Request processing method, device, equipment and storage medium
CN113691465B (en) Data transmission method, intelligent network card, computing device and storage medium
CN111580935A (en) Network communication method, device, equipment and storage medium
CN113076171B (en) Virtual machine installation method and device
CN111274022A (en) Server resource allocation method and system
CN112671561B (en) Network card configuration method and equipment of cloud host
CN116668372B (en) Flow control method and related device
CN110795201B (en) Management method and device for servers in cloud platform
CN114257505A (en) Server node configuration method, device, equipment and storage medium

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
GR01 Patent grant
GR01 Patent grant