Computing power information synchronization method and device, equipment, storage medium and product

Abstract

The application discloses a computing power information synchronization method and device, equipment, a storage medium and a product. The method comprises the following steps: an entry gateway of a distributed computing power network acquires to-be-updated computing power information, wherein the to-be-updated computing power information represents a state change of a computing power service; and the to-be-updated computing power information is synchronized to other entry gateways of the distributed computing power network. Compared with computing power node computing power information announcement in the distributed computing power network, the synchronization effect of the computing power information can be improved based on computing power information synchronization between a plurality of entry gateways of the distributed computing power network, thereby providing a more accurate basis for route decision-making and meeting the demand of computing power resource optimized allocation of the distributed computing power network.

Description

Technical Field

[0001] This application relates to the field of computing power networks, and in particular to a method, apparatus, device, storage medium, and product for synchronizing computing power information. Background Technology

[0002] A computational network is a network architecture that utilizes distributed computing power to accomplish specific tasks. Its core idea is to distribute large-scale computing demands among numerous participants, achieving efficient and rapid task completion through resource sharing and collaborative computing.

[0003] The distributed computing network is built upon the collaborative efforts of ingress gateways and egress gateways, forming a highly efficient and secure network architecture. The ingress gateway is a crucial component of the distributed system, acting as a bridge between clients and business services, responsible for receiving and processing requests from clients. The egress gateway primarily provides a unified exit point for system calls to third-party services and performs corresponding processing.

[0004] The ingress gateway needs to schedule the service request to the appropriate computing power node based on the client's service request. Therefore, each computing power node needs to notify the ingress gateway of its own computing power information in advance. The ingress gateway maintains a computing power status information table for each computing power node and makes routing decisions based on the maintained computing power status information table.

[0005] Because there are multiple ingress gateways and in order to reduce useless computing power information announcements in the network, related technologies often adopt a mechanism of periodic computing power information announcement and synchronization. That is, the information announcement and synchronization mechanism is from computing power nodes to egress gateways and then to ingress gateways. However, this periodic global update synchronization method will cause the computing power information of computing power nodes to be updated in a timely manner, affecting the routing decision effect. Summary of the Invention

[0006] In view of this, embodiments of this application provide a method, apparatus, device, storage medium, and product for synchronizing computing power information, aiming to improve the synchronization effect of computing power information and provide a more accurate basis for routing decisions.

[0007] The technical solution of this application embodiment is implemented as follows:

[0008] In a first aspect, embodiments of this application provide a method for synchronizing computing power information, applied to the entry gateway of a distributed computing power network, the method comprising:

[0009] Obtain the computing power information to be updated, wherein the computing power information to be updated represents the status change of computing power service;

[0010] The computing power information to be updated is synchronized to other entry gateways of the distributed computing power network.

[0011] In the above scheme, obtaining the computing power information to be updated includes:

[0012] Receive computing power information to be updated from computing power nodes based on a trigger-based mechanism; and / or,

[0013] The computing power information to be updated is generated based on an active mechanism.

[0014] In the above scheme, the computing power information to be updated sent by the receiving computing power node based on a trigger mechanism includes:

[0015] The system receives and sends updated computing power information generated by the computing power nodes based on the comparison results of the status change values ​​of computing power services with the set threshold.

[0016] In the above scheme, the generation of computing power information to be updated based on the active mechanism includes:

[0017] Receive a service request, the service request carrying the computing power resource usage and computing power resource usage duration;

[0018] The service requests are routed and scheduled based on the current computing power status information table.

[0019] The computing power information to be updated is generated based on the computing power resource usage and the computing power resource usage duration.

[0020] In the above scheme, the generation of computing power information to be updated based on the active mechanism includes:

[0021] Receive service requests;

[0022] Based on the business type to which the business request belongs, predict the computing resource consumption and computing resource consumption duration of the business request;

[0023] The service requests are routed and scheduled based on the current computing power status information table.

[0024] The computing power information to be updated is generated based on the computing power resource usage and the computing power resource usage duration.

[0025] In the above scheme, generating the computing power information to be updated based on the computing power resource occupancy and the computing power resource occupancy duration includes:

[0026] Based on the effective time of the routing schedule and the amount of computing resources occupied, generate the first computing power information to be updated corresponding to the resource occupancy;

[0027] Based on the duration of the computing power resource occupation, the time is calculated to determine the completion of the service request, and then the second computing power information to be updated corresponding to the resource release is generated.

[0028] In the above scheme, generating the computing power information to be updated based on the computing power resource occupancy and the computing power resource occupancy duration includes:

[0029] Based on the effective time of the routing schedule and the amount of computing power resources occupied, a first computing power information to be updated corresponding to the resource occupation is generated, and the first computing power information to be updated carries timing information indicating the duration of computing power resource occupation.

[0030] In the above scheme, synchronizing the computing power information to be updated to other entry gateways of the distributed computing power network includes one or more of the following:

[0031] Based on the boundary gateway protocol peer relationship between the ingress gateway and other ingress gateways of the distributed computing power network, the computing power information to be updated is announced globally.

[0032] Based on the fact that the ingress gateway and other ingress gateways of the distributed computing power network are in the same service type in the first gateway announcement domain, the computing power information to be updated corresponding to the service type is announced in the first gateway announcement domain;

[0033] Based on the second gateway announcement domain, which is geographically divided between the ingress gateway and other ingress gateways of the distributed computing power network, the computing power information to be updated is announced within the second gateway announcement domain.

[0034] In the above scheme, the entry gateway synchronizes the computing power information to be updated to other entry gateways of the distributed computing power network based on a multicast mechanism.

[0035] In the above scheme, the entry gateway synchronizes the computing power information to be updated to other entry gateways of the distributed computing power network based on a multicast mechanism, including one or more of the following:

[0036] The entry gateway, based on a multicast mechanism, synchronizes the computing power information to be updated to all entry gateways of the distributed computing power network.

[0037] The entry gateway, based on a multicast mechanism, synchronizes the computing power information to be updated to some entry gateways of the distributed computing power network.

[0038] The method in the above scheme further includes:

[0039] Update the locally maintained computing power status information table based on the computing power information to be updated locally and / or the computing power information to be updated received from other ingress gateways.

[0040] Secondly, embodiments of this application provide a computing power information synchronization device, applied to the entry gateway of a distributed computing power network, the device comprising:

[0041] The acquisition module is used to acquire computing power information to be updated, wherein the computing power information to be updated represents the status change of computing power service;

[0042] The synchronization module is used to synchronize the computing power information to be updated to other entry gateways of the distributed computing power network.

[0043] Thirdly, embodiments of this application provide a network device, which is an entry gateway for a distributed computing network. The network device includes a processor and a memory for storing a computer program that can run on the processor. The processor is used to execute the steps of the method described in the first aspect of embodiments of this application when running the computer program.

[0044] Fourthly, embodiments of this application provide a computer storage medium storing a computer program, which, when executed by a processor, implements the steps of the method described in the first aspect of embodiments of this application.

[0045] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in the first aspect of embodiments of this application.

[0046] The technical solution provided in this application embodiment involves an entry gateway of a distributed computing power network acquiring computing power information to be updated, whereby the computing power information represents changes in the status of computing power services; and synchronizing the computing power information to be updated to other entry gateways of the distributed computing power network. Compared to computing power nodes announcing computing power information within the distributed computing power network, this method improves the synchronization effect of computing power information by synchronizing computing power information among multiple entry gateways of the distributed computing power network, thereby providing a more accurate basis for routing decisions and meeting the needs of optimized allocation of computing power resources in the distributed computing power network. Attached Figure Description

[0047] Figure 1 This is a flowchart illustrating the computing power information synchronization method according to an embodiment of this application;

[0048] Figure 2 This is a schematic diagram illustrating the process of synchronizing computing power information to be updated by the entry gateway based on a trigger mechanism of computing power nodes in an embodiment of this application.

[0049] Figure 3 This is a schematic diagram illustrating the principle of establishing a BGP-peer relationship between ingress gateways in an embodiment of this application.

[0050] Figure 4 This is a schematic diagram illustrating the principle of the same advertising domain divided by geographical location for the entry gateway in an embodiment of this application;

[0051] Figure 5 This is a schematic diagram illustrating the process of the entry gateway synchronizing computing power information to be updated based on an active mechanism in an embodiment of this application.

[0052] Figure 6 This is a schematic diagram of a multicast message according to an embodiment of this application;

[0053] Figure 7 This is a schematic diagram of a multicast message according to another embodiment of this application;

[0054] Figure 8 This is a schematic diagram of the computing power information synchronization device according to an embodiment of this application;

[0055] Figure 9 This is a schematic diagram of the network device according to an embodiment of this application. Detailed Implementation

[0056] The present application will now be described in further detail with reference to the accompanying drawings and embodiments.

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.

[0058] In a distributed computing network, there are multiple entry gateways. When an entry gateway processes a scheduling service request, it will cause the status of several computing nodes to change. If the preset notification period has not yet arrived, the entry gateway will still maintain the old status information. If other entry gateways also process the same type of service request at the same time or close to the same time, the service errors will be concentrated on a few computing nodes due to untimely information synchronization, and the service requirements cannot be met. Therefore, it is necessary to consider timely synchronization of computing node status change information.

[0059] Furthermore, when several computing nodes complete the previous batch of services, the status of the computing nodes changes; before the notification period is reached, the status changes of the computing nodes cannot be synchronized to the ingress gateway in a timely manner, causing the ingress gateway to be unaware that the current load has been released, and thus unable to make scientific and reasonable routing decisions.

[0060] The embodiments of this application aim to change the traditional, inefficient, periodic, network-wide announcement method of computing power information synchronization, and provide a real-time enhanced computing power information synchronization mechanism to improve the synchronization effect of computing power information, thereby providing a more accurate basis for routing decisions.

[0061] This application provides a method for synchronizing computing power information, applied to the entry gateway of a distributed computing power network. It should be noted that the distributed computing power network includes multiple entry gateways and multiple exit gateways. Each entry gateway acts as a bridge between clients and service providers, responsible for receiving service requests from clients, processing them accordingly, and routing the service requests to the target service node. Each service node is connected to a corresponding exit gateway, and the entry gateways and exit gateways are connected via routing nodes, realizing a network system where entry gateways and exit gateways cooperate.

[0062] like Figure 1 As shown, the method includes:

[0063] Step 101: Obtain the computing power information to be updated, wherein the computing power information to be updated represents the status change of the computing power service.

[0064] Here, the computing power information to be updated can be computing power information at the node level or computing power information at the service level. In one application example, the computing power information at the node level may include: computing power node identifier, computing power resource change amount and timestamp information. The computing power information at the service level may include: service identifier, service performance change amount and timestamp information. This application embodiment does not limit this.

[0065] Step 102: Synchronize the computing power information to be updated to other entry gateways of the distributed computing power network.

[0066] Here, after the entry gateway obtains the computing power information to be updated, it synchronizes the computing power information to other entry gateways in the distributed computing power network. In this way, compared with the computing power nodes announcing computing power information within the distributed computing power network, the synchronization effect of computing power information can be improved based on the synchronization of computing power information among multiple entry gateways in the distributed computing power network. This provides a more accurate basis for routing decisions and meets the needs of optimized allocation of computing power resources in the distributed computing power network.

[0067] For example, obtaining the computing power information to be updated includes:

[0068] Receive computing power information to be updated from computing power nodes based on a trigger-based mechanism; and / or,

[0069] The computing power information to be updated is generated based on an active mechanism.

[0070] Here, computing nodes can send node-level computing power information to the ingress gateway based on a trigger-based mechanism. For example, a computing node can send the node-level computing power information to be updated to the egress gateway based on a trigger-based mechanism, and then the egress gateway can send it to the ingress gateway. In addition, the ingress gateway can also generate node-level and / or service-level computing power information based on a proactive mechanism.

[0071] It should be noted that both the computing power information to be updated under the trigger-based mechanism and the computing power information to be updated under the proactive mechanism have changed the traditional computing power information update mechanism based on the announcement period. This allows the entry gateway of the distributed computing power network to update computing power information more in real time, thereby providing a more reliable basis for the routing decision of computing power services and effectively avoiding the defects such as resource waste or service delay caused by the concentration of computing power.

[0072] For example, the computing power information to be updated sent by the receiving computing power node based on a triggered mechanism includes:

[0073] The system receives and sends updated computing power information generated by the computing power nodes based on the comparison results of the status change values ​​of computing power services with the set threshold.

[0074] Here, the change in the state of the computing power service refers to the change in computing resources. Computing resources may include, but are not limited to, CPU (Central Processing Unit) utilization, IOPS (Input / Output Operations Per Second), and other computing power parameters. For example, the change in the state of the computing power service can be obtained by comparing the computing resource characteristics of the computing power node before and after the update.

[0075] In one application example, the change in computing resources can be generated based on the difference between the previous update of the computing node and the current value representing the computing resources. For example, it could be a proportional value representing the rate of change of computing resources. The computing resource information to be updated is generated and sent based on the comparison between this proportional value and a set proportional threshold. Thus, unlike the periodic update mechanism based on a notification cycle, the updating of computing resource information can be achieved based on a trigger-based notification mechanism of the computing node, thereby increasing the update frequency of computing resource information. The computing resource information to be updated is the change in computing resources of the computing node.

[0076] For example, the generation of computing power information to be updated based on the proactive mechanism includes:

[0077] Receive a service request, the service request carrying the computing power resource usage and computing power resource usage duration;

[0078] The service requests are routed and scheduled based on the current computing power status information table.

[0079] The computing power information to be updated is generated based on the computing power resource usage and the computing power resource usage duration.

[0080] Here, the ingress gateway receives a service request from a client, which carries the amount and duration of computing power resource usage. The ingress gateway can route and schedule the service request based on the current computing power status information table. In addition, the ingress gateway also actively generates computing power information to be updated based on the amount and duration of computing power resource usage.

[0081] It should be noted that the ingress gateway can perform routing scheduling based on resource scheduling policies. In this case, the distributed computing power network focuses on how to effectively manage and allocate resources to ensure optimal resource utilization, that is, scheduling the target service node with the best resources to respond to business requests. Accordingly, the maintained computing power status information table includes computing power resource status information at the node level, and the actively generated computing power information to be updated is computing power information at the node level, which may include, for example, the node identifier, service performance change amount, and timestamp information of the target service node.

[0082] In other examples, the ingress gateway can also perform route scheduling based on service scheduling policies. Service scheduling focuses on optimizing task execution order and resource allocation to improve system performance and user experience. Accordingly, the maintained computing power status information table includes service performance information at the service granularity, and the actively generated computing power information to be updated is service-granularity computing power information, which may include, for example, the service identifier, service performance change amount, and timestamp information of the service.

[0083] For example, the generation of computing power information to be updated based on the proactive mechanism includes:

[0084] Receive service requests;

[0085] Based on the business type to which the business request belongs, predict the computing resource consumption and computing resource consumption duration of the business request;

[0086] The service requests are routed and scheduled based on the current computing power status information table.

[0087] The computing power information to be updated is generated based on the computing power resource usage and the computing power resource usage duration.

[0088] It should be noted that in one application example, the service request does not include the computing resource usage and duration of computing resource usage. The ingress gateway can predict the computing resource usage and duration of computing resource usage for the service request based on its service type. For example, it can predict the computing resource usage and duration of computing resource usage for the same service based on historical response data of the service request. In this way, after the ingress gateway routes and schedules the service request based on the current computing power status information table, it can also proactively generate computing power information to be updated based on the predicted computing resource usage and duration of computing resource usage.

[0089] For example, generating the computing power information to be updated based on the computing power resource occupancy and the computing power resource occupancy duration includes:

[0090] Based on the effective time of the routing schedule and the amount of computing resources occupied, generate the first computing power information to be updated corresponding to the resource occupancy;

[0091] Based on the duration of the computing power resource occupation, the time is calculated to determine when the service request is completed, and then the corresponding second computing power information to be updated is released.

[0092] Understandably, the ingress gateway can generate first computing power information to be updated, representing resource occupancy, based on the effective time of routing scheduling and the amount of computing power resources occupied. It can also generate second computing power information to be updated, representing resource release, by timing the duration of the computing power resource occupation based on a timer and determining that the service request has been completed (i.e., the timer duration has reached the duration of the computing power resource occupation).

[0093] In one embodiment, generating the computing power information to be updated based on the computing power resource occupancy and the computing power resource occupancy duration includes:

[0094] Based on the effective time of the routing schedule and the amount of computing power resources occupied, a first computing power information to be updated corresponding to the resource occupation is generated, and the first computing power information to be updated carries timing information indicating the duration of computing power resource occupation.

[0095] Understandably, after the entry gateway synchronizes the first computing power information to be updated corresponding to resource occupancy to other entry gateways, each entry gateway can determine the completion of the service request based on timed information (e.g., a timer), and then determine the release of resources and complete the update of the computing power information status table. In this way, the amount of computing power information synchronized between entry gateways can be reduced.

[0096] For example, synchronizing the computing power information to be updated to other entry gateways of the distributed computing power network includes one or more of the following:

[0097] Based on the Border Gateway Protocol peer relationship between the ingress gateway and other ingress gateways of the distributed computing power network, the computing power information to be updated is announced globally.

[0098] Based on the fact that the ingress gateway and other ingress gateways of the distributed computing power network are in the same service type in the first gateway announcement domain, the computing power information to be updated corresponding to the service type is announced in the first gateway announcement domain;

[0099] Based on the second gateway announcement domain, which is geographically divided between the ingress gateway and other ingress gateways of the distributed computing power network, the computing power information to be updated is announced within the second gateway announcement domain.

[0100] In one application example, BGP-peer relationships can be established between the entry gateways of the distributed computing power network, thereby enabling the global announcement of computing power information to be updated among the entry gateways of the distributed computing power network and synchronizing the computing power status information tables on each entry gateway side.

[0101] In another application example, in order to limit the scope of invalid announcements, the ingress gateway of the distributed computing power network can be divided into gateway announcement domains based on service type or geographical location. Then, the computing power information to be updated can be announced within the gateway announcement domain based on the gateway announcement domain identifier.

[0102] For example, the entry gateway synchronizes the computing power information to be updated to other entry gateways of the distributed computing power network based on a multicast mechanism.

[0103] Here, the entry gateway can synchronize the computing power information to be updated to other entry gateways of the distributed computing power network based on the multicast mechanism. Since the computing power information to be synchronized is the same, bandwidth and information overhead can be saved.

[0104] For example, the ingress gateway synchronizes the computing power information to be updated to other ingress gateways of the distributed computing power network based on a multicast mechanism, including one or more of the following:

[0105] The entry gateway, based on a multicast mechanism, synchronizes the computing power information to be updated to all entry gateways of the distributed computing power network.

[0106] The entry gateway, based on a multicast mechanism, synchronizes the computing power information to be updated to some entry gateways of the distributed computing power network.

[0107] It is understandable that all ingress gateways of the distributed computing power network can be added to a multicast group to achieve multicast across the entire network; or, the ingress gateways corresponding to the aforementioned gateway announcement domains can be added to the multicast group to achieve multicast within the gateway announcement domains.

[0108] Exemplarily, the method further includes:

[0109] Update the locally maintained computing power status information table based on the computing power information to be updated locally and / or the computing power information to be updated received from other ingress gateways.

[0110] Here, the ingress gateway can update its locally maintained computing power status information table based on the local computing power information to be updated and / or the computing power information to be updated received from other ingress gateways. For example, for a node-level computing power status information table, it determines whether the service node identifier of the computing power information to be updated belongs to the table. If so, it updates the service performance status and timestamp information corresponding to the service node in the table; otherwise, it adds the service node and its corresponding service performance status and timestamp information to the table. The update of the service-level computing power status information table can refer to the above process and will not be elaborated further here.

[0111] The present application will be further described in detail below with reference to application examples.

[0112] Application Example 1

[0113] This application example involves a scenario where computing nodes send computing power information to be updated based on a trigger-based mechanism.

[0114] Here, changes in the state of a computing node can be caused by either a new business request taking over or a previous business request completing. For example, as... Figure 2 As shown, after the task of computing power node 2 (Server2) is completed, if the status change reaches the threshold, it will actively notify the computing power information to be updated. For example, it will notify the change of computing power information of computing power node 2 to the ingress gateway 2 through the egress gateway 2. When the ingress gateway 2 obtains the change of computing power information of computing power node 2 (i.e., the computing power information to be updated), it will notify the change of computing power information of computing power node 2 to other ingress gateways such as ingress gateway 1 and ingress gateway 3.

[0115] For example, the computing power information to be updated may include: node identifier, node location information, and the change in the node's computing power resources. It is understood that the computing power information to be updated may also include timestamp information representing the time corresponding to the change in computing power resources.

[0116] Here, computing nodes can be configured with different thresholds for different services, as shown in the following example:

[0117] 1) Based on the business popularity, for example, for hot business, due to its large number of business requests, a lower threshold can be set, such as a change of more than 10%, to trigger an announcement to the ingress gateway; for ordinary business, due to its smaller volume of business, a higher threshold can be set, such as a change of more than 30%, to trigger an announcement.

[0118] 2) Based on the number of service deployment nodes, if a certain service has a large number of deployment nodes and the ingress gateway has a wide range of nodes to choose from, a higher threshold can be set, triggering a notification to the ingress gateway when the current node status changes significantly; if a certain service has a small number of deployment nodes and the ingress gateway has a small range of nodes to choose from, a lower threshold can be set, requiring a notification to the ingress gateway even when the current node status changes slightly.

[0119] In one application example, computing power information can be advertised globally among the entry gateways of a distributed computing power network, as follows:

[0120] 1) Establish BGP-peer relationships between ingress gateways (e.g.) Figure 3 As shown), the entire domain will be notified of changes in computing power. This information may include: the identifier of the computing power node, the location information of the computing power node, the amount of change in the computing power resources of the computing power node, and the timestamp information. For example, the information may be (Computing power node A, newly occupied resources of 20%, start time t1).

[0121] 2) After other entry gateways receive the change in computing power node status information, they compare it with the local computing power node status information table. If there is corresponding node information, they update the current computing power node information; if there is no current node information, they add the current node information.

[0122] 3) If both ingress gateway 1 and ingress gateway 2 simultaneously schedule service requests to computing power node A, then generate (Computing power node A, newly occupied resources 20%, start time t1, source ingress gateway 1 IP) and (Computing power node A, newly occupied resources 10%, start time t2, source ingress gateway 2 IP) respectively, and send them to other ingress gateways to distinguish the newly occupied resources and start time.

[0123] In another application example, after the ingress gateway processes multiple user requests, it needs to notify other ingress gateway nodes of the latest status changes, filter the received computing power information, and forward it separately. In order to limit the scope of invalid announcements, the ingress gateway can announce computing power information based on the gateway announcement domain.

[0124] For example, you can divide the ingress gateway announcement domains for the same service based on service deployment information, and set the announcement relationships between ingress gateways, specifically including:

[0125] 1) The ingress gateway receives multiple computing power status information from the egress gateway; such as (service A, performance information; service B, performance information), that is, it receives computing power information at the service granularity.

[0126] 2) Based on the announcement domains between ingress gateways, the ingress gateway queries and determines the corresponding ingress gateway announcement list according to different services. For example, for service A, ingress gateways 2 and 3; for service B, ingress gateways 3 and 4. It can be understood that the ingress gateway side is configured with the ingress gateway announcement list corresponding to the service, and thus can determine the corresponding ingress gateway announcement list based on the service identifier.

[0127] 3) The entry gateway announces the service status information according to the selected entry gateway; for example, only announce the information of service A to entry gateways 2 and 3; including service identification information, computing node location information, the locking status of newly added business resources of computing nodes, and announcement domain identification information.

[0128] For example, the notification range between nearby entry gateways can be divided according to geographical location, and the notification relationship between entry gateways can be set, specifically including:

[0129] 1) The ingress gateway receives multiple computing power status messages from the egress gateway;

[0130] 2) Based on the announcement domains between ingress gateways, the ingress gateway queries to determine the corresponding ingress gateway announcement list. For example, ingress gateways 2, 3, and 4 belong to the same announcement domain as ingress gateway 1, defined by geographical location (e.g., ...). Figure 4 (as shown);

[0131] 3) The entry gateway announces service status information based on the selected entry gateway; for example, only entry gateways 2, 3, and 4 are notified; this includes service identification information, computing node location information, the locking status of newly added business resources on the computing node, and announcement domain identification information. It can be understood that this announcement domain identification information is used to identify the gateway's announcement domain. In this way, the updating and synchronization of computing power information within the gateway's announcement domain can be achieved.

[0132] Application Example 2

[0133] This application example involves a scenario where the ingress gateway generates computing power information to be updated based on an active mechanism.

[0134] In one application example, a business request can carry the estimated computing resource usage and duration. Based on this business request, the ingress gateway can proactively update the computing power status information table and synchronize it to other ingress gateways. Specifically, this includes:

[0135] 1) The ingress gateway receives a user request (i.e., a business request) and carries the estimated computing resource usage and computing resource usage duration;

[0136] 2) After the ingress gateway completes the scheduling, it actively updates the local computing power status information table. For example, if computing power node A is available, it can be adjusted from 100% to (computing power node A newly occupies 20%, occupancy time).

[0137] 3) The ingress gateway announces the status information of the computing power nodes to other ingress gateways via BGP / multicast, and carries a timer to indicate the latest status information of the computing power nodes within this period;

[0138] 4) Each entry gateway adjusts its local computing power status information table; when the preset occupancy time expires, each entry gateway actively adjusts the status information of the corresponding computing power node A.

[0139] In another application example, after the ingress gateway processes multiple user service requests, it can locally maintain information on the estimated additional resources and time occupied by computing power nodes, proactively update the computing power status information table, and synchronize it to other ingress gateways, such as... Figure 5 As shown, it can specifically include:

[0140] 1) The entry gateway estimates the occupancy and occupancy time based on historical information;

[0141] Here, the ingress gateway 2 receives a user request, which carries the target service requested. The ingress gateway 2 determines the computing power resource usage and computing power resource usage duration based on a prediction mechanism. For example, a timer is introduced to represent the required usage duration of the computing power node service resources. This usage duration can be inferred by the ingress gateway based on the historical usage duration of the same service.

[0142] 2) Maintain the local timer and update the computing node information;

[0143] During the preset timer period, the ingress gateway 2 locally and proactively adjusts the computing power status information table; for example, computing power node A can be adjusted from 100% to (computing power service A newly occupies 20%, occupancy time).

[0144] 3) Notify other ingress gateways of computing power status information, including time information;

[0145] Ingress Gateway 2 announces the status information of computing power nodes to other ingress gateways via BGP / multicast, and carries a timer to indicate the latest status information of the computing power nodes within this period;

[0146] 4) Update the computing node information after the timer expires.

[0147] When the preset timer expires, each ingress gateway actively adjusts the status information of the corresponding computing node A.

[0148] For example, when an ingress gateway announces computing power information to other ingress gateways, it announces the same computing power information. Therefore, to save bandwidth and information overhead, information announcement between ingress gateways can be implemented based on a multicast mechanism, which may specifically include:

[0149] 1) Add all ingress gateways to a multicast group, with the source IP address (source ingress gateway address) in the multicast packet header, and add all ingress gateways to the destination node;

[0150] 2) The source ingress gateway copies and sends multiple copies according to ingress gateway 2 and ingress gateway 3; the computing power node information includes computing power node identification information, computing power node location information, the locking status of newly added service resources of the computing power node, and timestamp information. For example, this multicast message is as follows: Figure 6 As shown, the payload carries computing power node information (i.e., computing power information to be updated).

[0151] In other application examples, an announcement domain can be considered. Based on the announcement domain, the relevant ingress gateways are added to different multicast groups. The computing power information includes computing node identification information, computing node location information, the locking status of newly added service resources on the computing node, and announcement domain identification information (used to identify different announcement domains). For example, the multicast message is as follows: Figure 7 As shown, the multicast packet header carries the source IP address (source ingress gateway address) and adds the ingress gateway within the advertised domain to the destination node.

[0152] Understandably, multicast messaging can effectively improve the efficiency of transmitting computing power information.

[0153] In order to implement the method of the embodiments of this application, the embodiments of this application also provide a computing power information synchronization device, which corresponds to the computing power information synchronization method described above. The steps in the embodiments of the computing power information synchronization method described above are also fully applicable to the embodiments of this computing power information synchronization device.

[0154] like Figure 8 As shown, the computing power information synchronization device includes an acquisition module 801 and a synchronization module 802. The acquisition module 801 is used to acquire computing power information to be updated, which represents the status change of computing power service; the synchronization module 802 is used to synchronize the computing power information to be updated to other entry gateways of the distributed computing power network.

[0155] For example, the acquisition module 801 is specifically used for:

[0156] Receive computing power information to be updated from computing power nodes based on a trigger-based mechanism; and / or,

[0157] The computing power information to be updated is generated based on an active mechanism.

[0158] For example, receiving computing power information to be updated from a computing power node based on a trigger mechanism includes:

[0159] The system receives and sends updated computing power information generated by the computing power nodes based on the comparison results of the status change values ​​of computing power services with the set threshold.

[0160] For example, the generation of computing power information to be updated based on the proactive mechanism includes:

[0161] Receive a service request, the service request carrying the computing power resource usage and computing power resource usage duration;

[0162] The service requests are routed and scheduled based on the current computing power status information table.

[0163] The computing power information to be updated is generated based on the computing power resource usage and the computing power resource usage duration.

[0164] For example, the generation of computing power information to be updated based on the proactive mechanism includes:

[0165] Receive service requests;

[0166] Based on the business type to which the business request belongs, predict the computing resource consumption and computing resource consumption duration of the business request;

[0167] The service requests are routed and scheduled based on the current computing power status information table.

[0168] The computing power information to be updated is generated based on the computing power resource usage and the computing power resource usage duration.

[0169] For example, generating the computing power information to be updated based on the computing power resource occupancy and the computing power resource occupancy duration includes:

[0170] Based on the effective time of the routing schedule and the amount of computing resources occupied, generate the first computing power information to be updated corresponding to the resource occupancy;

[0171] Based on the duration of the computing power resource occupation, the time is calculated to determine the completion of the service request, and then the second computing power information to be updated corresponding to the resource release is generated.

[0172] For example, generating the computing power information to be updated based on the computing power resource occupancy and the computing power resource occupancy duration includes:

[0173] Based on the effective time of the routing schedule and the amount of computing power resources occupied, a first computing power information to be updated corresponding to the resource occupation is generated, and the first computing power information to be updated carries timing information indicating the duration of computing power resource occupation.

[0174] For example, the synchronization module 802 synchronizes the computing power information to be updated to other entry gateways of the distributed computing power network, including one or more of the following:

[0175] Based on the boundary gateway protocol peer relationship between the ingress gateway and other ingress gateways of the distributed computing power network, the computing power information to be updated is announced globally.

[0176] Based on the fact that the ingress gateway and other ingress gateways of the distributed computing power network are in the same service type in the first gateway announcement domain, the computing power information to be updated corresponding to the service type is announced in the first gateway announcement domain;

[0177] Based on the second gateway announcement domain, which is geographically divided between the ingress gateway and other ingress gateways of the distributed computing power network, the computing power information to be updated is announced within the second gateway announcement domain.

[0178] For example, the synchronization module 802 synchronizes the computing power information to be updated to other entry gateways of the distributed computing power network based on a multicast mechanism.

[0179] For example, the synchronization module 802 synchronizes the computing power information to be updated to other entry gateways of the distributed computing power network based on a multicast mechanism, including one or more of the following:

[0180] The entry gateway, based on a multicast mechanism, synchronizes the computing power information to be updated to all entry gateways of the distributed computing power network.

[0181] The entry gateway, based on a multicast mechanism, synchronizes the computing power information to be updated to some entry gateways of the distributed computing power network.

[0182] For example, the computing power information synchronization device further includes: an update module 803, used to update the locally maintained computing power status information table based on the local computing power information to be updated and / or the computing power information to be updated received from other ingress gateways.

[0183] In practical applications, the acquisition module 801, synchronization module 802, and update module 803 can be implemented by the processor in the ingress gateway. Of course, the processor needs to run the computer program in memory to implement its functions.

[0184] It should be noted that the computing power information synchronization device provided in the above embodiments is only illustrated by the division of the above-described program modules when performing computing power information synchronization. In practical applications, the above processing can be assigned to different program modules as needed, that is, the internal structure of the device can be divided into different program modules to complete all or part of the processing described above. In addition, the computing power information synchronization device and the computing power information synchronization method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiments, which will not be repeated here.

[0185] Based on the hardware implementation of the above program modules, and in order to implement the method of the embodiments of this application, the embodiments of this application also provide a network device. Figure 9 This is only an exemplary structure of the network device, not the entire structure; implementation is possible as needed. Figure 9 The structure shown may be part or all of the structure.

[0186] like Figure 9 As shown, the network device 900 provided in this embodiment includes at least one processor 901, a memory 902, a user interface 903, and at least one network interface 904. The various components in the network device 900 are coupled together via a bus system 905. It can be understood that the bus system 905 is used to implement communication between these components. In addition to a data bus, the bus system 905 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in… Figure 9 The general labeled all buses as Bus System 905.

[0187] The user interface 903 may include a monitor, keyboard, mouse, trackball, click wheel, buttons, touchpad, or touch screen.

[0188] The memory 902 in this embodiment is used to store various types of data to support the operation of the network device. Examples of such data include any computer program used to operate on the network device.

[0189] The computing power information synchronization method disclosed in this application can be applied to or implemented by the processor 901. The processor 901 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the computing power information synchronization method can be completed by the integrated logic circuits in the hardware of the processor 901 or by instructions in software form. The processor 901 can be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 901 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor, etc. The steps of the method disclosed in the embodiments of this application can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software modules can be located in a storage medium, specifically in the memory 902. The processor 901 reads the information in the memory 902 and, in conjunction with its hardware, completes the steps of the computing power information synchronization method provided in the embodiments of this application.

[0190] In an exemplary embodiment, network device 900 can be an entry gateway for a distributed computing network. Network device 900 can be implemented by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers (MCUs), microprocessors, or other electronic components to perform the aforementioned method.

[0191] It is understood that memory 902 can be volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), ferromagnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM); magnetic surface memory can be disk storage or magnetic tape storage. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), Sync Link Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM).The memories described in the embodiments of this application are intended to include, but are not limited to, these and any other suitable types of memories.

[0192] In an exemplary embodiment, this application also provides a computer storage medium, specifically a computer-readable storage medium, such as a memory 902 storing a computer program, which can be executed by a processor 901 of a network device to complete the steps described in the method of this application embodiment. The computer-readable storage medium can be a ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface memory, optical disc, or CD-ROM, etc.

[0193] In an exemplary embodiment, this application also provides a computer program product, including a computer program that can be executed by a processor 901 of a network device 900 to perform the steps described in the method of this application embodiment.

[0194] It should be noted that terms such as "first" and "second" are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0195] Furthermore, the technical solutions described in the embodiments of this application can be combined arbitrarily without conflict.

[0196] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for synchronizing computing power information, characterized in that, An entry gateway applied to a distributed computing power network, the method comprising: Obtain the computing power information to be updated, wherein the computing power information to be updated represents the status change of computing power service; The computing power information to be updated is synchronized to other entry gateways of the distributed computing power network; The step of obtaining the computing power information to be updated includes: Receive service requests; The service requests are routed and scheduled based on the current computing power status information table. The computing power information to be updated is generated based on the computing power resource usage and the computing power resource usage duration; the computing power resource usage and the computing power resource usage duration are carried by the service request or predicted based on the service type to which the service request belongs; The step of generating the computing power information to be updated based on the computing power resource occupancy and the computing power resource occupancy duration includes: Based on the effective time of the routing schedule and the amount of computing resources occupied, a first set of computing power information to be updated corresponding to the resource occupation is generated; based on the duration of computing power resource occupation, a timer is performed to determine that the service request has been completed, and then a second set of computing power information to be updated corresponding to the resource release is generated; or... Based on the effective time of the routing schedule and the amount of computing power resources occupied, a first computing power information to be updated corresponding to the resource occupation is generated, and the first computing power information to be updated carries timing information indicating the duration of computing power resource occupation.

2. The method according to claim 1, characterized in that, The process of obtaining the computing power information to be updated also includes: Receive computing power information to be updated from computing power nodes based on a trigger mechanism.

3. The method according to claim 2, characterized in that, The computing power information to be updated sent by the receiving computing power node based on a triggered mechanism includes: The system receives and sends updated computing power information generated by the computing power nodes based on the comparison results of the status change values ​​of computing power services with the set threshold.

4. The method according to claim 1, characterized in that, The step of synchronizing the computing power information to be updated to other entry gateways of the distributed computing power network includes one or more of the following: Based on the boundary gateway protocol peer relationship between the ingress gateway and other ingress gateways of the distributed computing power network, the computing power information to be updated is announced globally. Based on the fact that the ingress gateway and other ingress gateways of the distributed computing power network are in the same service type in the first gateway announcement domain, the computing power information to be updated corresponding to the service type is announced in the first gateway announcement domain; Based on the second gateway announcement domain, which is geographically divided between the ingress gateway and other ingress gateways of the distributed computing power network, the computing power information to be updated is announced within the second gateway announcement domain.

5. The method according to claim 1, characterized in that, The entry gateway uses a multicast mechanism to synchronize the computing power information to be updated to other entry gateways of the distributed computing power network.

6. The method according to claim 5, characterized in that, The entry gateway, based on a multicast mechanism, synchronizes the computing power information to be updated to other entry gateways of the distributed computing power network, including one or more of the following: The entry gateway, based on a multicast mechanism, synchronizes the computing power information to be updated to all entry gateways of the distributed computing power network. The entry gateway, based on a multicast mechanism, synchronizes the computing power information to be updated to some entry gateways of the distributed computing power network.

7. The method according to claim 1, characterized in that, The method further includes: Update the locally maintained computing power status information table based on the computing power information to be updated locally and / or the computing power information to be updated received from other ingress gateways.

8. A computing power information synchronization device, characterized in that, An ingress gateway for a distributed computing network, the device comprising: The acquisition module is used to acquire computing power information to be updated, wherein the computing power information to be updated represents the status change of computing power service; The synchronization module is used to synchronize the computing power information to be updated to other entry gateways of the distributed computing power network; Specifically, the acquisition module is used to: receive a service request; route the service request based on the current computing power status information table; generate first computing power information to be updated corresponding to resource occupancy based on the effective time of the route scheduling and the computing power resource occupancy; determine the completion of the service request based on the computing power resource occupancy duration, and then generate second computing power information to be updated corresponding to resource release; or, generate first computing power information to be updated corresponding to resource occupancy based on the effective time of the route scheduling and the computing power resource occupancy, wherein the first computing power information to be updated carries timing information indicating the computing power resource occupancy duration; the computing power resource occupancy and the computing power resource occupancy duration are carried by the service request or predicted based on the service type to which the service request belongs.

9. A network device, characterized in that, The network device serves as the entry gateway for the distributed computing network. The network device includes a processor and a memory for storing computer programs that can run on the processor. The processor, when running a computer program, performs the steps of the method according to any one of claims 1 to 7.

10. A computer storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images