WO2017067484A1 - Virtualization data center scheduling system and method - Google Patents

Abstract

The present invention discloses a virtualization data center scheduling system and method. The system comprises: a service definition and orchestration module, a data center management platform, and a network controller, wherein defined interfaces of a service type, a resource template, and a reliability requirement level are provided, a service orchestration policy is received to generate a service function chain, and a resource of a data center is allocated according to the defined service type and resource template to generate a service node. Upon determination of an anomaly of the service node, a resource of the data center is scheduled according to the service reliability requirement level to restore the service node, and a data packet forwarding path of the service node is decided. In this way, when an anomaly of a service node occurs, the service node can be restored according to a service type, resource template, and service reliability requirement level defined by a user, thereby ensuring normal operation of the service node in a service function chain, and improving the service of the data center to be highly reliable.

Description

Virtualized data center scheduling system and method Technical field

The present invention relates to the field of cloud computing and data center virtualization technologies, and more particularly to a virtualized data center scheduling system and method.

Background technique

At present, with the rapid development of emerging applications such as cloud computing and big data, the demand for data centers is also increasing. Among them, data center virtualization is the basic trend of service-oriented data centers, and virtualized data centers have unified servers. Operational flexibility, power savings, space savings, disaster recovery, high availability and many other advantages.

In the prior art, with the development and promotion of a Software Defined Network (SDN) and a Network Function Virtualization (NFV) architecture and related technologies, a Service Function Chaining (Service Function Chaining, Abbreviation: SFC) is a set of service nodes with a specific service path, which is an ideal solution for providing services to virtualized data centers.

However, with the prior art, since the service function chain is complex in composition, its reliability depends on the reliability of each service function node. Therefore, in order to ensure high reliability service in the data center, how to maintain the service node in the service function chain Reliability is a key issue that must be addressed.

Summary of the invention

In order to solve the above technical problem, an embodiment of the present invention provides a virtualized data center scheduling system and method, which can maintain normal operation of a service node in a service function chain, thereby providing a high reliability service for a data center.

In a first aspect, an embodiment of the present invention provides a virtualized data center scheduling system, including:

Service definition and orchestration modules, data center management platforms, and network controllers;

The service definition and orchestration module is configured to provide a service interface, a resource template, and a reliability requirement level definition interface, and receive a service orchestration policy generation service function chain, where the service reliability requirement level includes an abnormal alarm, and a single service node automatically recovers Differentiated service active and standby nodes and equal service active and standby nodes;

The data center management platform is configured to allocate a resource node of the data center according to the defined service type and the resource template, and after determining that the service node is abnormal, scheduling resource recovery of the data center according to a service reliability requirement level. The service node;

The network controller is configured to formulate a packet forwarding path of the service node.

In a second aspect, an embodiment of the present invention provides a virtualized data center scheduling method, where the method includes:

Providing a service interface type, a resource template, and a reliability requirement level definition interface, and receiving a service orchestration policy generation service function chain, where the service reliability requirement level includes an abnormal alarm, a single service node automatic recovery, a differentiated service active standby node, and an equalization Serving the active and standby nodes;

Allocating, according to the defined service type and the resource template, the resource generation service node of the data center, after determining that the service node is abnormal, scheduling the resource of the data center according to the service reliability requirement level to restore the service node;

A packet forwarding path of the service node is formulated.

Compared with the prior art, a virtualized data center scheduling system and method provided by an embodiment of the present invention includes: a service definition and orchestration module, a data center management platform, and a network controller, and provides services through a service definition and orchestration module. The definition interface of the type, the resource template, and the reliability requirement level, and the service service orchestration policy generation service function chain, the service reliability requirement level includes an abnormal alarm, a single service node automatic recovery, a differentiated service active standby node, and an equal service active/standby a node, the data center management platform allocates a resource node of the data center according to the defined service type and the resource template, and after determining that the service node is abnormal, scheduling the resource of the data center according to the service reliability requirement level to restore the service. a node, the network controller formulates a packet forwarding path of the service node, so that when the service node is abnormal, the service node can be restored according to the service type, resource template, and service reliability requirement level defined by the user, thereby ensuring Service section of the functional service chain Normal operation, high reliability and improved service data center.

Other features and advantages of the invention will be set forth in the description which follows, The objectives and other advantages of the invention may be realized and obtained by means of the structure particularly pointed in the appended claims.

DRAWINGS

The drawings are used to provide a further understanding of the technical solutions of the present invention, and constitute a part of the specification, which together with the embodiments of the present application are used to explain the technical solutions of the present invention, and do not constitute a limitation of the technical solutions of the present invention.

1 is a functional block diagram of a virtualized data center scheduling system according to an embodiment of the present invention;

2 is a functional block diagram of a second embodiment of a virtualized data center scheduling system according to the present invention;

3 is a functional block diagram of a third embodiment of a virtualized data center scheduling system according to the present invention;

FIG. 4 is a schematic flowchart of Embodiment 1 of a method for scheduling a virtualized data center according to the present invention.

detailed description

The embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that, in the case of no conflict, the features in the embodiments and the embodiments in the present application may be arbitrarily combined with each other.

The steps illustrated in the flowchart of the figures may be executed in a computer system such as a set of computer executable instructions. Also, although logical sequences are shown in the flowcharts, in some cases the steps shown or described may be performed in a different order than the ones described herein.

The data center according to the embodiment of the present invention refers to transmitting, accelerating, displaying, calculating, and storing data information on the internet infrastructure based on virtualization, that is, multiple virtual machines can be run on one computer, or multiple Environments share the resources of this computer. Different virtual machines can run different operating systems and multiple applications on the same computer.

The system according to the embodiment of the present invention aims to solve the technical problem of how to maintain the normal operation of the service node of the service function chain to provide a highly reliable service for the data center in the prior art.

The technical solutions of the present invention will be described in detail below with specific embodiments. The following specific embodiments may be combined with each other, and the same or similar concepts or processes may not be described in some embodiments.

1 is a functional block diagram of a virtualized data center scheduling system according to an embodiment of the present invention. As shown in FIG. 1, the system includes: a service definition and orchestration module 10, a data center management platform 20, and a network controller 30;

The service definition and orchestration module 10 is configured to provide a service interface, a resource template, and a reliability requirement level definition interface, and receive a service orchestration policy generation service function chain, where the service reliability requirement level includes an abnormal alarm, and a single service node automatically Recovery, differentiated service primary and secondary nodes and equal service active and standby nodes;

Specifically, the service definition and orchestration module 10 accepts a tenant or a user to define an interface according to a service type, a resource template, and a reliability requirement level provided by the virtualized data center scheduling system, and customizes a service orchestration policy to determine a requirement. The service reliability requirement level generates a service function chain, wherein the service reliability requirement level includes: an abnormal alarm, a single service node automatic recovery, a differentiated service active standby node, and an equal service active standby node.

The data center management platform 20 is configured to allocate a resource generation service node of the data center according to the defined service type and the resource template, and after determining that the service node is abnormal, scheduling the resource of the data center according to the service reliability requirement level. Recovering the service node;

Specifically, the data center management platform 20 allocates a resource generating service node according to the defined service type and the resource template to provide a guarantee for the service node to process the service service, and after determining the abnormality of the service node, according to the service reliability. The demand level schedules resources of the data center to recover the service node.

The network controller 30 is configured to formulate a packet forwarding path of the serving node.

Specifically, the network of the virtualized data center scheduling system generally adopts a control plane and a forwarding plane separation scheme, and the network controller 30 uniformly controls the forwarding plane device to formulate a packet forwarding policy of the service node.

A virtualized data center scheduling system provided by an embodiment of the present invention includes: a service definition and orchestration module, a data center management platform, and a network controller, and provides a service type, a resource template, and a reliability requirement level through a service definition and orchestration module. The definition interface, the receiving service orchestration policy generates a service function chain, and the service reliability requirement level includes an abnormal alarm, a single service node automatic recovery, a differentiated service active standby node, and an equal service active standby node, and the data center management platform is defined according to the definition The service type and the resource template are allocated to the resource generation service node of the data center. After determining the abnormality of the service node, the resource of the data center is scheduled to be restored according to the service reliability requirement level, and the network controller formulates the service. The packet forwarding path of the node, so that when the service node is abnormal, the service node can be restored according to the service type, resource template, and service reliability requirement level defined by the user, thereby ensuring the normal operation of the service node of the functional service chain. , increased the number High reliability and service center.

2 is a functional block diagram of a second embodiment of a virtualized data center scheduling system according to the present invention. As shown in FIG. 2, the data center management platform 20 includes: a device resource management platform 210 and a virtualization service management platform 220. among them,

The device resource management platform 210 is configured to allocate, according to the defined service type and the resource template, the hardware device resource of the data center, manage and maintain the hardware device resource, provide performance metric data of the hardware device resource, and determine the Whether the service node has an exception;

The virtualization service management platform 220 is configured to allocate a virtualized resource generation service node of the data center according to the defined service type and the resource template, and after determining that the service node is abnormal, scheduling the service according to the service reliability requirement level. The virtualized resource of the data center recovers the service node

Specifically, the data center management platform 20 includes: a device resource management platform 210 and a virtualization service management platform 220, wherein the device resource management platform 210 is configured to manage and maintain data center hardware device resources, and provide hardware device resource performance. Metric data, such as hardware performance indicators such as storage, calculation, and network of the computer or server corresponding to the service node. According to the performance metric data, the running status of the service node can be detected to determine whether the service node is abnormal, and the hardware device resource is abnormal. A fault alarm is provided. The virtualization service management platform 220 is configured to allocate a virtualized resource generation service node of the data center according to the defined service type and the resource template, and after determining that the service node is abnormal, according to the service reliability requirement level, The service node provides virtualized resource management and scheduling based on hardware device resources, including computing, storage, and network virtualization resources to recover the service node.

Further, FIG. 3 is a functional block diagram of a third embodiment of a virtualized data center scheduling system according to the present invention. As shown in FIG. 3, the system further includes: a service detecting module 40;

The service detection module 40 is configured to detect, when the service service chain has a service primary node and a service standby node, consistency of service status of the service primary node and the service standby node, and timeliness of receiving data packets, Determining whether the service primary node and the service standby node are abnormal.

Specifically, the service detecting module 40 records a mapping relationship between the service master node and the service standby node when the service function node has a service master node and a service standby node, and the mapping relationship may be a service master node. Corresponding to a service standby node, which may also be a corresponding multiple service standby node, the service master node and the service standby node together form a service set, and detect the consistency of service status of the active and standby service nodes of the service set. That is, whether the data processed by the service master node and the service standby node are consistent, so as to determine whether the service master node and the service standby node service need to be synchronized, and also detect the timeliness of receiving the data packet. Determining whether the service master node and the service standby node are abnormal according to whether the time of receiving the data packet of the service master node or the service standby node is out of time. For example, the service detection module 40 receives the service master at time T1. Node data packet X receives the service standby node data packet X at time T2, only when the time difference between T1 and T2 is less than the limit When the value of the service node and the master node serving standby normal, abnormal and vice versa. It is determined whether there is a delay, so that it can be determined whether the service primary node and the service standby node are abnormal.

Further, after determining that the service master node is abnormal, the service detecting module 40 is further configured to:

Re-determine the service master-slave relationship, set the original service standby node as the service master node, and set the original service master node as the service standby node.

Specifically, when the service detecting module 40 determines that the service primary node is abnormal, in order to restore the service as soon as possible, the service The service detection module 40 is further configured to re-determine the relationship between the service master node and the service standby node. The original service standby node may be the service master node, and the original service master node is the service standby node to ensure uninterrupted service.

Further, based on the foregoing embodiment, as shown in FIG. 3, the system further includes: a service backup module 50;

The service backup module 50 is configured to: if the service primary node and the service standby node have different service states, determine that the service standby node needs to synchronize with the service primary node, and synchronize the memory of the service primary node. And the image file to the service standby node.

Specifically, when the service detection module 40 detects that the service status of the service primary node and the service standby node is inconsistent, it is determined that the service standby node needs to synchronize with the service primary node, and the service detection module 40 triggers synchronization. The request to the service backup module 50, the service backup module 50 synchronizes the primary node memory and the image file to the service standby node in an incremental manner, so that the service active and standby nodes synchronize to work non-periodically, saving data center resources and minimizing It affects the processing power of the service master node.

Further, on the basis of the foregoing embodiment, the service backup module 50 is further configured to create a snapshot storage according to a preset time interval. If a remote disaster recovery data center exists, the service standby node is synchronized according to a preset time interval. Mirror the file to the remote disaster recovery data center.

Specifically, the service backup module 50 is further configured to create a snapshot storage according to a certain time interval, and the interval time may be specifically set according to the situation, and the image file of the standby node is synchronously synchronized to the disaster recovery data center to ensure data. The data security and reliability of the center. If there is a disaster recovery data center, the disaster recovery data center can perform health status monitoring and function switching on the data center. When a system stops working due to an accident (such as a fire, an earthquake, etc.), The entire system can be switched to another location so that the system functions can continue to function normally.

Further, on the basis of the foregoing embodiment, the virtualization service management platform 220 is configured to, after determining that the service node is abnormal, scheduling the virtualized resource of the data center to restore the service node according to a service reliability requirement level. ,Refers to:

The virtualization service management platform 220 is configured to generate an alarm prompt after determining that the service node is abnormal, when the service reliability requirement level is abnormal.

When the service reliability requirement level is automatically restored by the single service node, after determining that the service node is abnormal, the service node is restored according to the single service node recovery policy;

Generating a service main section for the service node when the service reliability requirement level is a differentiated or equal service primary and secondary node After determining that the service standby node is abnormal, the point and the service standby node recover the service standby node according to the single service node recovery policy, and complete synchronization with the service primary node, and after determining that the service primary node is abnormal, switch The original service standby node is the service primary node, and the original service primary node is the service standby node, and the original service primary node is restored according to the single service node recovery policy.

Specifically, the virtualization service management platform 220 is configured to detect the service node by using the hardware performance metric data of the service node when the service reliability requirement level is abnormal, and generate an alarm prompt when the service is unavailable. An exception that can be handled manually by a system administrator or related user.

The virtualization service management platform 220 is configured to perform a restart on the service node according to the hardware device resource performance metric data provided by the device resource management platform 210 when the service reliability requirement level is automatically restored by the single service node. Migration or rebuild operation.

The virtualization service management platform 220 is configured to generate a service primary node and a service standby node for the service node when the service reliability requirement level is a differentiated or equal service primary standby node, after determining that the service standby node is abnormal Performing a restart, migration, or reconstruction operation on the service node according to the hardware device resource performance metric data provided by the device resource management platform 210 to restore the service standby node, and completing synchronization to the service primary node, in determining the After the service master node is abnormal, the original service standby node is the service master node, and the original service master node is the service standby node, and the service node is restarted and migrated according to the hardware device resource performance metric data provided by the device resource management platform 210. Or the rebuild operation restores the original service master node.

As described above, the single service node recovery strategy includes:

If only the service provided by the service node is abnormal, and the corresponding virtual computing, network connection, and storage resources are normal, the service node is restarted;

If the virtual computing or network resource corresponding to the service node is abnormal, performing a migration operation according to the storage resource image of the service node to restore a service function of the service node;

If the service node storage resource is abnormal, causing the service node image file to use abnormality, if there is a snapshot resource, attempting to reconstruct the service node with the latest snapshot of the service node, or when there is no snapshot resource, then virtualizing The service management platform reconstructs the service node according to the virtual resource template of the service node.

Specifically, the performance metric data may reflect a specific abnormality, such as a storage exception, a calculation abnormality, or a network abnormality. When the service reliability requirement level is automatically restored for the single service node, if only the service node is abnormal, and the virtual computing, network, and storage resources of the computing node are normal, the virtualization service management platform 220 performs the restart. The service node, if the virtual node of the compute node or the network resource of the service node is abnormal, according to the shared storage image, the virtualization service management platform 220 performs a migration operation and attempts to restore the service function. If the storage node stores the resource abnormally, the If the image file usage is abnormal, the virtualization service management platform 220 attempts to reconstruct the service node with the latest snapshot of the service node. If there is no snapshot resource, the virtualization service management platform 220 reconstructs the service node according to its virtual resource template.

Further, the virtualization service management platform is configured to: when the service reliability requirement level is a difference or an equal service primary standby node, after the service primary node and the service standby node are generated for the service node, further configured to:

When the service reliability requirement level is a differentiated service primary and secondary node, different virtual resources are allocated to the service primary node and the service standby node, and the virtual resources are distributed on different hardware devices;

When the service reliability requirement level is an equal service primary and secondary node, the same virtual resource is allocated to the service primary node and the service standby node, and the virtual resources are distributed on different hardware devices.

Specifically, when the service reliability requirement level is a differentiated service active and standby node, the virtualization service management platform 220 allocates different virtual resources, which are different from the service primary node and the service standby node, and the service standby node The virtual resources are thinned relative to the virtual resources of the service master node and distributed on different hardware devices. In order to achieve the effect of saving resources, in order to detect the abnormality of the main service, the switching service can be automatically automated to the standby node, and the service is not interrupted.

Specifically, when the service reliability requirement level is an equal service active standby node, the virtualization service management platform 220 allocates the same virtual resource of the service primary node and the service standby node, and is distributed in different hardware. On the device, in order to detect the main service exception, because of the same virtual resources, the switch between the active and standby services can be more rapid, and the impact on the entire service chain is minimized.

Further, on the basis of the foregoing embodiment, when the service reliability requirement level is a differentiated or equal service active standby node, the virtualization service management platform switches the service after determining that the service primary node is abnormal. The service master node means:

The network controller is requested to change a service port chain, and the service master node is replaced by a service standby node to process a service service.

Specifically, after determining that the service primary node is abnormal, when the serving primary node is a serving standby node, a request is sent to the network controller to change a service port chain, and the service standby node is replaced by the service standby node. Dealing with the service business can make the service uninterrupted and ensure the normal operation of the entire service function chain.

Further, on the basis of the foregoing embodiment, before the virtualized service management platform replaces the service master node as a service standby node to process the service service, the method further includes:

When the service reliability requirement level is the differentiated service primary and secondary nodes, the service standby node is online expanded with the virtualized resources equal to the service primary node.

Specifically, when the service reliability requirement level is a differentiated service primary and secondary node, the resources allocated by the virtualization service management platform to the service primary node and the service standby node are different, and an abnormality occurs in the service primary node. The service primary node is switched to be a service standby node. Therefore, the service standby node needs to be expanded online, so that the service standby node and the abnormal service primary node have equal virtualized resources, so that the service is prepared. The node has the ability to handle the business that was previously processed by the abnormal service master node.

Further, on the basis of the foregoing embodiment, after the determining that the service primary node is abnormal, the virtualized service management platform recovers the service primary node according to the single service node recovery policy, and further includes:

When the service reliability requirement level is a differentiated service primary and secondary node, the virtualized resource equal to the original serving standby node is obtained when the service primary node is restored.

Specifically, if the service reliability requirement level is the differentiated service primary and secondary nodes, when the original service primary node is restored according to the single service node recovery policy, the virtualized resources allocated according to the original service standby node virtual resource specifications are used to enable the service to be processed. The service service of the original service standby node.

The network controller 30 formulates the data forwarding policy of the service primary node and the service standby node when the service reliability requirement level is a difference or an equal service primary and secondary node, and the service primary node and the service standby node receive The request of the upstream service node sends the data processed by the service master node and the service standby node to the downstream service node, and forwards the data to the service detection module, that is, in the process of performing the service, the service master node and the service standby node both Data processing is performed so that when the service master node is abnormal, the service standby node can continue to serve, thereby ensuring the normal operation of the service service function chain.

Further, FIG. 4 is a schematic flowchart of Embodiment 1 of a method for scheduling a virtualized data center according to an embodiment of the present invention. As shown in FIG. 4, the method includes:

S101. The interface for providing a service type, a resource template, and a reliability requirement level, and receiving a service orchestration policy to generate a service function chain, where the service reliability requirement level includes an abnormal alarm, a single service node automatic recovery, and a differentiated service active and standby node. And equal service primary and secondary nodes;

S102. Allocate a resource of the data center according to the defined service type and the resource template to generate a service node. After the service node is abnormal, scheduling the resource of the data center according to the service reliability requirement level to restore the service node;

S103. Formulate a packet forwarding path of the service node.

A virtualized data center scheduling method provided by the embodiment of the present invention includes: providing a service interface type, a resource template, and a reliability requirement level definition interface, and receiving a service orchestration policy generation service function chain, where the service reliability requirement level includes An abnormal alarm, a single service node automatic recovery, a differentiated service primary standby node, and an equal service active standby node, and a data center resource generation service node is allocated according to the defined service type and resource template, and after determining the service node abnormality, according to The service reliability requirement level schedules the resources of the data center to recover the service node, and formulates a data packet forwarding path of the service node, so that when an abnormality occurs in the service node, the service type, resource template, and The service reliability requirement level is used to restore the service node, thereby ensuring the normal operation of the service node of the functional service chain and improving the high reliability service of the data center.

Further, based on the foregoing embodiment, the foregoing S102 includes:

Allocating hardware device resources of the data center according to the defined service type and the resource template, managing and maintaining the hardware device resource, providing performance metric data of the hardware device resource, and determining whether the service node is abnormal;

And assigning a service node to the virtualized resource of the data center according to the defined service type and the resource template, and after determining that the service node is abnormal, scheduling the virtualized resource of the data center to restore the service node according to the service reliability requirement level.

The method provided by the embodiment of the present invention can be used to implement the foregoing device embodiment, and the implementation principle and the technical effect are similar, and details are not described herein again.

Further, on the basis of the foregoing embodiment, after the foregoing S102 generates the service node by using the virtualized resource of the data center according to the defined service type and the resource template, the method further includes:

When the service service node has a service primary node and a service standby node, the mapping relationship between the service primary node and the service standby node is recorded, and the service status of the service primary node and the service standby node is detected. The timeliness of the data packet determines whether the service master node and the service standby node are abnormal.

The method provided by the embodiment of the present invention can be used to implement the foregoing device embodiment, and the implementation principle and the technical effect are similar, and details are not described herein again.

Further, on the basis of the foregoing embodiment, after determining that the service master node is abnormal, the method further includes:

Re-determine the service master-slave relationship, set the original service standby node as the service master node, and set the original service master node as the service standby node.

The method provided by the embodiment of the present invention can be used to implement the foregoing device embodiment, and the implementation principle and the technical effect are similar, and details are not described herein again.

Further, after the foregoing detecting the consistency of the service status of the service master node and the service standby node, the method further includes:

If the service master node and the service standby node have different service states, determine that the service standby node needs to synchronize with the service master node, and synchronize the memory and image files of the service master node to the service standby node.

The method provided by the embodiment of the present invention can be used to implement the foregoing device embodiment, and the implementation principle and the technical effect are similar, and details are not described herein again.

Further, on the basis of the foregoing embodiment, after determining that the service standby node needs to synchronize with the service master node, and synchronizing the memory and the image file of the service master node to the service standby node, the method further includes:

The snapshot storage is created at the preset interval. If the remote disaster recovery data center exists, the image of the standby node is synchronized to the remote disaster recovery data center.

The method provided by the embodiment of the present invention can be used to implement the foregoing device embodiment, and the implementation principle and the technical effect are similar, and details are not described herein again.

Further, based on the foregoing embodiment, the foregoing S102 is specifically:

When the service reliability requirement level is an abnormal alarm, the virtualization service management platform generates an alarm prompt after determining that the service node is abnormal;

When the service reliability requirement level is automatically restored by the single service node, the virtualized service management platform recovers the service node according to the single service node recovery policy after determining that the service node is abnormal;

When the service reliability requirement level is a differentiated or equal service primary and secondary node, the virtualization service management platform generates a service primary node and a service standby node for the service node, and after determining that the service standby node is abnormal, according to the single The service node recovery policy restores the service standby node, and completes synchronization with the service primary node. After determining that the service primary node is abnormal, the service primary node is switched, and the service primary node is restored according to a single service node recovery policy. .

The method provided by the embodiment of the present invention can be used to implement the foregoing device embodiment, and the implementation principle and the technical effect are similar, and details are not described herein again.

Specifically, the single service node recovery policy includes:

If only the service provided by the service node is abnormal, and the corresponding virtual computing, network connection, and storage resources are normal, the service node is restarted;

If the virtual computing or network resource corresponding to the service node is abnormal, performing a migration operation according to the storage resource image of the service node to restore a service function of the service node;

If the service node storage resource is abnormal, causing the service node image file to use abnormality, if there is a snapshot resource, attempting to reconstruct the service node with the latest snapshot of the service node, or when there is no snapshot resource, then virtualizing The service management platform reconstructs the service node according to the virtual resource template of the service node.

The method provided by the embodiment of the present invention can be used to implement the foregoing device embodiment, and the implementation principle and the technical effect are similar, and details are not described herein again.

Further, on the basis of the foregoing embodiment, when the service reliability requirement level is a difference or an equal service primary and secondary node, generating a service primary node and a service standby node for the service node, including:

When the service reliability requirement level is a differentiated service primary and secondary node, different virtual resources are allocated to the service primary node and the service standby node, and the virtual resources are distributed on different hardware devices;

When the service reliability requirement level is an equal service primary and secondary node, the same virtual resource is allocated to the service primary node and the service standby node, and the virtual resources are distributed on different hardware devices.

The method provided by the embodiment of the present invention can be used to implement the foregoing device embodiment, and the implementation principle and the technical effect are similar, and details are not described herein again.

Further, on the basis of the foregoing embodiment, when the service reliability requirement level is a differentiated or equal service primary and secondary node, after determining that the service primary node is abnormal, switching the service primary node includes:

The network controller is requested to change a service port chain, and the service master node is replaced by a service standby node to process a service service.

The method provided by the embodiment of the present invention can be used to implement the foregoing device embodiment, and the implementation principle and the technical effect are similar, and details are not described herein again.

Further, on the basis of the foregoing embodiment, before the service primary node is replaced by the serving standby node to process the service service, the method further includes:

When the service reliability requirement level is the differentiated service primary and secondary nodes, the service standby node is online expanded with the virtualized resources equal to the service primary node.

The method provided by the embodiment of the present invention can be used to implement the foregoing device embodiment, and the implementation principle and the technical effect are similar, and details are not described herein again.

Further, on the basis of the foregoing embodiment, after determining that the service primary node is abnormal, after recovering the service primary node according to the single service node recovery policy, the method further includes:

When the service reliability requirement level is a differentiated service primary and secondary node, the virtualized resource equal to the original service standby node is obtained when the service primary node is restored.

The method provided by the embodiment of the present invention can be used to implement the foregoing device embodiment, and the implementation principle and the technical effect are similar, and details are not described herein again.

While the embodiments of the present invention have been described above, the described embodiments are merely for the purpose of understanding the invention and are not intended to limit the invention. Any modification and variation in the form and details of the embodiments may be made by those skilled in the art without departing from the spirit and scope of the invention. The scope defined by the appended claims shall prevail.

Claims (24)

A virtualized data center scheduling system includes: a service definition and orchestration module, a data center management platform, and a network controller; The service definition and orchestration module is configured to provide a service interface, a resource template, and a reliability requirement level definition interface, and receive a service orchestration policy generation service function chain, where the service reliability requirement level includes an abnormal alarm, and a single service node automatically recovers Differentiated service active and standby nodes and equal service active and standby nodes; The data center management platform is configured to allocate a resource node of the data center according to the defined service type and the resource template, and after determining that the service node is abnormal, scheduling resource recovery of the data center according to a service reliability requirement level. The service node; The network controller is configured to formulate a packet forwarding path of the service node. The system of claim 1, wherein the data center management platform comprises: a device resource management platform and a virtualization service management platform; The device resource management platform is configured to allocate hardware device resources of the data center according to the defined service type and the resource template, manage and maintain the hardware device resources, provide performance metric data of the hardware device resources, and determine the service. Whether the node has an exception; The virtualization service management platform is configured to allocate a virtualized resource generation service node of the data center according to the defined service type and the resource template, and after determining that the service node is abnormal, scheduling the data center according to a service reliability requirement level. The virtualized resource restores the service node. The system of claim 2, wherein the system further comprises: a service detection module; The service detection module is configured to detect, when the service service chain has a service primary node and a service standby node, the consistency of the service status of the service primary node and the service standby node, and the timeliness of receiving the data packet, and determine Whether the service primary node and the service standby node are abnormal. The system according to claim 3, wherein the service detecting module further sets the following after determining that the service master node is abnormal: Re-determine the service master-slave relationship, set the original service standby node as the service master node, and set the original service master node as the service standby node. The system of claim 4, wherein the system further comprises: a service backup module; The service backup module is configured to: if the service primary node and the service standby node have different service states, determine that the service standby node needs to synchronize with the service primary node, and synchronize the memory of the service primary node and Mirror the file to the service standby node. The system of claim 5, wherein the service backup module is further configured to create a snapshot storage according to a preset time interval, and if there is a remote disaster recovery data center, the mirror of the service standby node is synchronized according to a preset time interval. Files to a remote disaster recovery data center. The system according to claim 6, wherein the virtualization service management platform is configured to, after determining that the service node is abnormal, scheduling the virtualized resource of the data center to restore the service node according to a service reliability requirement level, Refers to: The virtualization service management platform is configured to generate an alarm prompt after determining that the service node is abnormal, when the service reliability requirement level is abnormal. When the service reliability requirement level is automatically restored by the single service node, after determining that the service node is abnormal, the service node is restored according to the single service node recovery policy; And when the service reliability requirement level is a differentiated or equal service primary and secondary node, generating a service primary node and a service standby node for the service node, and after determining that the service standby node is abnormal, recovering according to the single service node recovery policy. Serving the standby node, and completing synchronization with the service primary node. After determining that the service primary node is abnormal, the original service standby node is switched as the service primary node, and the original service primary node is the service standby node, and the recovery is performed according to the single service node recovery policy. The original service master node. The system of claim 7 wherein said single serving node recovery policy comprises: If only the service provided by the service node is abnormal, and the corresponding virtual computing, network connection, and storage resources are normal, the service node is restarted; If the virtual computing or network resource corresponding to the service node is abnormal, performing a migration operation according to the storage resource image of the service node to restore a service function of the service node; If the service node storage resource is abnormal, causing the service node image file to use abnormality, if there is a snapshot resource, attempting to reconstruct the service node with the latest snapshot of the service node, or when there is no snapshot resource, then virtualizing The service management platform reconstructs the service node according to the virtual resource template of the service node. The system according to claim 7, wherein the virtualization service management platform is configured to generate a service primary node and a service standby node for the service node when the service reliability requirement level is a difference or an equal service primary standby node, Refers to: When the service reliability requirement level is a differentiated service primary and secondary node, different virtual resources are allocated to the service primary node and the service standby node, and the virtual resources are distributed on different hardware devices; When the service reliability requirement level is an equal service primary and secondary node, the same virtual resource is allocated to the service primary node and the service standby node, and the virtual resources are distributed on different hardware devices. The system according to claim 9, wherein the virtualization service management platform is configured to switch after determining that the service primary node is abnormal when the service reliability requirement level is a differentiated or equal service primary standby node. The service master node refers to: The network controller is requested to change a service port chain, and the service master node is replaced by a service standby node to process a service service. The system of claim 10, wherein the virtualization service management platform is configured to replace the service main section Before the point is the service standby node to handle the service business, it also includes: When the service reliability requirement level is the differentiated service primary and secondary nodes, the service standby node is online expanded with the virtualized resources equal to the service primary node. The system according to claim 11, wherein the virtualization service management platform is configured to restore the service master node according to a single service node recovery policy after determining that the service master node is abnormal, and further configured to: When the service reliability requirement level is a differentiated service primary and secondary node, the virtualized resource equal to the original service standby node is obtained when the service primary node is restored. A virtualized data center scheduling method includes: Providing a service interface type, a resource template, and a reliability requirement level definition interface, and receiving a service orchestration policy generation service function chain, where the service reliability requirement level includes an abnormal alarm, a single service node automatic recovery, a differentiated service active standby node, and an equalization Serving the active and standby nodes; Allocating, according to the defined service type and the resource template, the resource generation service node of the data center, after determining that the service node is abnormal, scheduling the resource of the data center according to the service reliability requirement level to restore the service node; A packet forwarding path of the service node is formulated. The method according to claim 13, wherein the resource generation service node according to the defined service type and resource template allocation data center, after determining that the service node is abnormal, scheduling the data according to a service reliability requirement level The central resource restores the service node, including: Allocating hardware device resources of the data center according to the defined service type and the resource template, managing and maintaining the hardware device resource, providing performance metric data of the hardware device resource, and determining whether the service node is abnormal; And assigning a service node to the virtualized resource of the data center according to the defined service type and the resource template, and after determining that the service node is abnormal, scheduling the virtualized resource of the data center to restore the service node according to the service reliability requirement level. The method of claim 14, wherein after the service node is allocated according to the defined service type and the resource template to allocate the virtualized resource of the data center, the method further includes: When the service service node has a service primary node and a service standby node, the mapping relationship between the service primary node and the service standby node is recorded, and the service status of the service primary node and the service standby node is detected. The timeliness of the data packet determines whether the service master node and the service standby node are abnormal. The method according to claim 15, wherein after determining that the service master node is abnormal, the method further comprises: Re-determine the service master-slave relationship, set the original service standby node as the service master node, and set the original service master node as the service standby node. The method according to claim 16, wherein said detecting said service master node and said service standby node service state After the consistency of the state, it also includes: If the service master node and the service standby node have different service states, determine that the service standby node needs to synchronize with the service master node, and synchronize the memory and image files of the service master node to the service standby node. The method of claim 17, wherein the determining that the service standby node needs to synchronize with the service master node, after synchronizing the memory and image files of the service master node to the service standby node, further includes: The snapshot storage is created at the preset interval. If the remote disaster recovery data center exists, the image of the standby node is synchronized to the remote disaster recovery data center. The method of claim 18, wherein, after determining that the service node is abnormal, scheduling the virtualized resource of the data center to restore the service node according to a service reliability requirement level comprises: When the service reliability requirement level is an abnormal alarm, the virtualization service management platform generates an alarm prompt after determining that the service node is abnormal; When the service reliability requirement level is automatically restored by the single service node, the virtualized service management platform recovers the service node according to the single service node recovery policy after determining that the service node is abnormal; When the service reliability requirement level is a differentiated or equal service primary and secondary node, the virtualization service management platform generates a service primary node and a service standby node for the service node, and after determining that the service standby node is abnormal, according to the single The service node recovery policy restores the service standby node, and completes synchronization with the service primary node. After determining that the service primary node is abnormal, the service primary node is switched, and the service primary node is restored according to a single service node recovery policy. . The method of claim 19, wherein the single serving node recovery policy comprises: If only the service provided by the service node is abnormal, and the corresponding virtual computing, network connection, and storage resources are normal, the service node is restarted; If the virtual computing or network resource corresponding to the service node is abnormal, performing a migration operation according to the storage resource image of the service node to restore a service function of the service node; If the service node storage resource is abnormal, causing the service node image file to use abnormality, if there is a snapshot resource, attempting to reconstruct the service node with the latest snapshot of the service node, or when there is no snapshot resource, then virtualizing The service management platform reconstructs the service node according to the virtual resource template of the service node. The method according to claim 19, wherein when the service reliability requirement level is a difference or an equal service primary node, generating a service master node and a service standby node for the service node, including: When the service reliability requirement level is a differentiated service primary and secondary node, different virtual resources are allocated to the service primary node and the service standby node, and the virtual resources are distributed on different hardware devices; When the service reliability requirement level is an equal service primary and secondary node, the same virtual resource is allocated to the service primary node and the service standby node, and the virtual resources are distributed on different hardware devices. The method of claim 21, wherein, when the service reliability requirement level is a differentiated or equal service primary and secondary node, after determining that the service primary node is abnormal, switching the service primary node comprises: The network controller is requested to change a service port chain, and the service master node is replaced by a service standby node to process a service service. The method according to claim 22, further comprising: before replacing the service master node as a service standby node to process service services, further comprising: When the service reliability requirement level is the differentiated service primary and secondary nodes, the service standby node is online expanded with the virtualized resources equal to the service primary node. The method of claim 23, wherein when the service reliability requirement level is a differentiated or equal service primary and backup node, recovering the service master node according to the single service node recovery policy, further comprising: The virtualized resource restored by the service primary node is equal to the virtualized resource of the original serving standby node.

Images