CN111147278A

CN111147278A - Device for executing computer instruction

Info

Publication number: CN111147278A
Application number: CN201911177485.0A
Authority: CN
Inventors: 夏军
Original assignee: Anhui University
Current assignee: Anhui University
Priority date: 2019-11-20
Filing date: 2019-11-20
Publication date: 2020-05-12

Abstract

The invention discloses a device of a computer instruction execution method, which relates to the technical field of computer instructions. The apparatus of the computer instruction execution method supports upgrading of various component modules of an industry gateway VNF via a VNFM. In the upgrading process, the device of the computer instruction execution method supports rollback through the VNFM to a pre-upgrading state if the upgrading process is terminated through the VNFM and the upgrading process fails.

Description

Device for executing computer instruction

Technical Field

The invention relates to the technical field of computer instructions, in particular to a device for executing a computer instruction.

Background

Existing computer communication systems all employ a conventional CT architecture. The existing computer communication system has complex equipment installation, which results in long new construction and capacity expansion period; the server of each product can only be used by itself, and hardware resources can not be shared among or few service products, so that the investment of operators is very large; modules of part of service products seriously depend on special hardware of manufacturers and lack flexibility; version upgrading operation is complex, service needs to be interrupted in the upgrading process, and user experience and income of operators are influenced; the complexity of network networking and system architecture leads to very high operation and maintenance costs for operators.

Disclosure of Invention

The technical problem to be solved by the invention is to provide a device of a computer instruction execution method, which is used for completing the virtualization and clouding of the whole computer communication system equipment, realizing software and hardware decoupling by virtualization, realizing the sharing of software and hardware resources and the automatic expansion of system resources along with service capacity by NFV clouding, and supporting the quick innovation and the quick online of services.

In order to achieve the above purpose, the present invention provides the following technical solutions:

NFV, also called network function virtualization, is a method for reducing the cost of expensive network devices by using general-purpose hardware such as x86 and virtualization technology to carry very many functional software processes. NFV generally refers to the process of integrating network resources and network functions into a single software-based management entity, i.e., a virtual network, where different services or control logics can share the same physical infrastructure through virtualization technology.

The device of the computer instruction execution method comprises a signaling front-end processor, wherein the signaling front-end processor is connected with a charging module and an operation maintenance module, the charging module is connected with a service processing module and a GDB, the service processing module is connected with an SMPP interface module and a DCACHE, and the operation maintenance module is connected with the GDB, the charging module, the DCACHE and a QAS; the signaling front-end processor is responsible for a module for processing SS7 signaling and is responsible for message butt joint of the computer communication center and the core network; the service processing module is a computer communication service processing logic module; the charging module is a ticket sorting module; the SMPP interface module is an SMPP protocol interface machine; the operation maintenance module is an operation maintenance platform; QAS is an inquiry analysis module, providing computer communication inquiry and statistical data; the GDB is responsible for storing user account opening data and the like; the DCACHE is a distributed computer communication logic memory database and is used for storing cache data of computer communication, such as computer communication to be sent and timing queues.

Aiming at a service processing module VNFC, the device of the computer instruction execution method supports the following elastic expansion model by a system: the method comprises the following steps that automatic elastic expansion and contraction are carried out, a computer communication center VNF management module (VNFM) triggers elastic expansion and contraction of a business processor module according to a rule of VNFD, and the triggering rule comprises the following steps: the service processing module uses the resource use condition of the virtual machine, the event condition received from the service processing module and the like; the method comprises the steps that a VNF instance of a computer communication center monitors the state of a VNFC of a service processor according to the scaling of the VNF, and elastic scaling of a service processing module is carried out by sending a clear request for increasing or decreasing the VNFC instance of the service processor to a VNFM; and according to the request expansion of the management module, manually triggering the expansion or sending a request to the NFVO through a certain interface by the OSS/BSS, and triggering the elastic expansion of the service processor according to the rule of the VNFD.

The device of the computer instruction execution method comprises three parts of SMSC signaling gateway clouding: a fixed virtual machine part of a signaling gateway foreground, an elastic virtual machine part of the signaling gateway foreground and a control part of the signaling gateway background; the signaling gateway background control part and the signaling gateway background operation maintenance part are accessed to observe the running condition of the signaling gateway, integrate the OMU function and are responsible for interacting with VNFM and DCF to complete the instantiation and service reporting functions; the fixed virtual machine part of the signaling gateway foreground is a necessary part of the signaling gateway virtual machine, is not telescopic and consists of two modules. The OMP is a master control module and is the master control logic of various service functions. The SLB thread is in charge of the elastic expansion function and interacts with an OMU of the SIU background control part to complete the SIU instantiation and elastic expansion functions. The SIPI module provides an external network interface, and comprises an external network element which is butted by using an SCTP (stream control transmission protocol) and a service processing module which is interacted by using a TCP (transmission control protocol); the elastic virtual machine part of the signaling gateway foreground is an essential part of the signaling gateway virtual machine and can be elastically stretched. And the VNFM judges whether elastic expansion is needed or not according to the indexes reported by the OMU. And after the expansion is finished, reporting the CMP service state to the DCF by the OMU.

The beneficial effect of adopting above technical scheme is: after the device of the computer instruction execution method adopts a VNF framework, the system is not distinguished to be upgraded or updated any more, and is processed in a unified manner. The system supports upgrading of individual component modules of an industry gateway VNF through a VNFM. In the upgrading process, if the upgrading process is failed due to the fact that the system supports the VNFM to terminate, the system supports rollback through the VNFM, and the system rolls back to the state before upgrading. For the computer communication center, the signaling front-end module and the service processing module can automatically or manually pop out a new virtual machine through KPI (key performance indicator) control of a VNFM (virtual network FM) to realize capacity expansion of the service and recycle virtual machine resources to realize capacity reduction of the service. For the computer communication gateway, the service processor and interface module can automatically or manually eject a new virtual machine through the KPI index control of the VNFM to realize the capacity expansion of the service and recover the virtual machine resources to realize the capacity reduction of the service. And in the process of popping up a new virtual machine, the business operation is not influenced. When the virtual machine resources are recovered, graceful offline is supported, that is, the virtual machine module to be recovered receives the offline preprocessing message sent by the VNFM first, then a new service request is not sent to the virtual machine module to be recovered, and the virtual machine module to be recovered formally offline after processing all current service requests.

Drawings

The following describes embodiments of the present invention in further detail with reference to the accompanying drawings.

FIG. 1 is a schematic diagram of a VNF architecture of a computer communication center of an apparatus for implementing a computer instruction according to the present invention;

FIG. 2 is a schematic diagram showing the operation of the apparatus for executing the computer instructions according to the present invention;

fig. 3 is a schematic diagram of the operation of an industry gateway.

Detailed Description

The preferred embodiments of the apparatus for implementing the method by computer instructions according to the present invention are described in detail below with reference to the accompanying drawings.

FIGS. 1, 2 and 3 show embodiments of an apparatus for performing a method according to the present invention by computer instructions:

as shown in fig. 1, the apparatus for executing the computer instruction includes a signaling front-end processor, the signaling front-end processor is connected with a charging module and an operation maintenance module, the charging module is connected with a service processing module and a GDB, the service processing module is connected with an SMPP interface module and a DCACHE, and the operation maintenance module is connected with the GDB, the charging module, the DCACHE and a QAS; the signaling front-end processor is responsible for a module for processing SS7 signaling and is responsible for message butt joint of the computer communication center and the core network; the service processing module is a computer communication service processing logic module; the charging module is a ticket sorting module; the SMPP interface module is an SMPP protocol interface machine; the operation maintenance module is an operation maintenance platform; QAS is an inquiry analysis module, providing computer communication inquiry and statistical data; the GDB is responsible for storing user account opening data and the like; the DCACHE is a distributed computer communication logic memory database and is used for storing cache data of computer communication, such as computer communication to be sent and timing queues.

Whether each VNFC of the computer communication center contains a different state is different, and for a service processing module that requires support of elastic scaling, separation of service logic and data states needs to be achieved, that is, stateless is achieved. The VNFC states are listed below:

VNFC	VNFC status
		Signaling front module	Stateless
Business processing module	Stateless
		Charging module	Has a state
SMPP interface module	Stateless
		Operation maintenance module	Has a state
Query analysis module	Has a state
		GDB module	Has a state
DCACHE	Has a state

As shown in fig. 2, for the traffic processing module VNFC, the system supports the following elastic scaling model:

automatic elastic expansion

The method comprises the following steps that a computer communication center VNF management module (VNFM) triggers elastic expansion and contraction of a business processor module according to a rule of VNFD, and the triggering rule comprises the following steps: the service processing module uses the resource use condition of the virtual machine, the event condition received from the service processing module and the like.

Scaling as required by VNF

The VNF instance of the computer communication center monitors the state of the VNFC, and elastic expansion and contraction of the service processing module are carried out by sending explicit request for increasing or decreasing the VNFC instance of the service processing machine to the VNFM.

Scaling on request of management module

The scaling is triggered manually (for example, by an operator in a network operation center) or the OSS/BSS sends a request to the NFVO via an interface to trigger the elastic scaling of the service handler according to the rules of the VNFD.

SMSC signaling gateway cloud divides into three parts: the virtual machine part is fixed on the signaling gateway foreground, the virtual machine part is elastic on the signaling gateway foreground, and the control part is background on the signaling gateway.

Signaling gateway background control part

And the signaling gateway background operation maintenance part is accessed to observe the running condition of the signaling gateway, integrates the OMU function, is responsible for interacting with the VNFM and the DCF and completes the instantiation and service reporting functions.

Front stage fixed virtual machine part of signaling gateway

The part is a necessary part of a signaling gateway virtual machine, is not telescopic and consists of two modules. The OMP is a master control module and is the master control logic of various service functions. The SLB thread is in charge of the elastic expansion function and interacts with an OMU of the SIU background control part to complete the SIU instantiation and elastic expansion functions. The SIPI module provides an external network interface, and comprises an external network element which is butted by using an SCTP protocol and a service processing module which interacts by using a TCP protocol.

Signaling gateway foreground elastic virtual machine part

The part is an essential part of the signaling gateway virtual machine and can be elastically stretched. And the VNFM judges whether elastic expansion is needed or not according to the indexes reported by the OMU. And after the expansion is finished, reporting the CMP service state to the DCF by the OMU.

The traditional architecture computer communication center service processor is combined with a local database, cache short messages stored in the local database are regularly issued to queues and other data which do not support the dynamic expansion of service processing nodes. A DCACHE mechanism is adopted to store cache data in a cloud scheme of a database, and meanwhile, a DRM (data resource management) module is designed at each service node to provide access interfaces for data addition, deletion, search, modification and the like for upper-layer services, so that the data storage requirements under the condition of dynamic expansion of the service nodes are met.

After the VNF framework is adopted, the system does not distinguish upgrading and updating and is processed in a unified mode. The system supports upgrading individual component modules of a computer communications center VNF through a VNFM. In the upgrading process, if the upgrading process is failed due to the fact that the system supports the VNFM to terminate, the system supports rollback through the VNFM, and the system rolls back to the state before upgrading.

The vssc supports COTS-Ready. The vssc has no dependency at all on the infrastructure hardware, allowing it to run on any suitable hardware. All VNFCs of the computer communication center can adopt virtualization software such as KVM and VMware EXSi. And a service processing module of the VNF of the computer communication center supports elastic expansion and contraction. The computer communications center provides comprehensive policy management functionality by the NVFO based on VNF-based policy management. Determining elastic expansion of the business processing module based on the following strategies: a TPS-based resiliency policy; an elasticity strategy based on CPU occupancy rate; an elastic strategy based on memory occupancy rate; time-based resiliency policies, etc.; and (5) migrating the operation. The resource allocation is dynamically changed by adopting a mechanism of closing the virtual machine resources on the current hardware and then restarting the virtual machine resources on the other hardware, so that the realization of disaster tolerance of the internal module of the VNF of the computer communication center is ensured.

As shown in fig. 3, the industry gateway adopts a VNF-internal Load Balancer mode, and a processor & interface VNFC of the computer communication a is used for distributing CMPP messages.

Whether each VNFC of the industry gateway contains a different state is different, and for a service processing module that requires support of elastic scaling, separation of service logic and data state needs to be achieved, that is, stateless implementation is achieved. The VNFC states are listed below:

aiming at an A module and an M module of an industry gateway and a service processing module VNFC of a log server, the system supports the following elastic expansion model:

automatic elastic expansion, an industry gateway VNF management module (VNFM) triggers the elastic expansion of a business processor module according to the rule of VNFD, and the triggering rule comprises the following steps: the service processing module uses the resource usage condition of the virtual machine, the queue state of the service processing machine, the processing capacity of the service processing machine, the event condition received from the service processing module and the like.

And scaling according to the requirement of the VNF, monitoring the state of the VNFC by the VNF instance of the industry gateway, and performing elastic scaling of the service processing module by sending a clear request for increasing or decreasing the VNFC instance of the service processor to the VNFM.

According to the request scaling of the management module, scaling is triggered manually (for example, through an operator of a network operation center) or the OSS/BSS sends a request to the NFVO through an interface to trigger the service handler to scale elastically according to the rule of the VNFD.

The gateway service processor and the local database of the traditional architecture industry are combined, cache short messages stored in the local database are stored, and data such as queues and the like are issued at regular time to not support dynamic expansion of service processing nodes. A DCACHE mechanism is adopted to store cache data in a cloud scheme of a database, and meanwhile, a DRM (data resource management) module is designed at each service node to provide access interfaces for data addition, deletion, search, modification and the like for upper-layer services, so that the data storage requirements under the condition of dynamic expansion of the service nodes are met.

After the VNF framework is adopted, the system does not distinguish upgrading and updating and is processed in a unified mode. The system supports upgrading of individual component modules of an industry gateway VNF through a VNFM. In the upgrading process, if the upgrading process is failed due to the fact that the system supports the VNFM to terminate, the system supports rollback through the VNFM, and the system rolls back to the state before upgrading.

Industry gateway VNF characteristics:

hardware dependency, vSMGM supports COTS-Ready. vsmmg has no dependency at all on the infrastructure hardware, allowing it to run on any suitable hardware.

Virtual layer dependency, all VNFCs of an industry gateway may employ virtualization software such as KVM, VMware EXSi, and the like.

And elastic support, wherein the business processing modules of A and M of the VNF support elastic expansion.

VNF policy management, VNF-based policy management of the industry gateway, and comprehensive policy management functions provided by the NVFO. Determining elastic expansion of the M business processing module based on the following strategies:

TPS-based resiliency policy

Elastic strategy based on CPU occupancy rate

Elastic strategy based on memory occupancy rate

Time-based resiliency policies and the like

Migration operations

Shielding strategy for EC MT service processing success rate

Process exception masking policy

The resource allocation is dynamically changed by adopting a mechanism of closing the virtual machine resources on the current hardware and then restarting the virtual machine resources on the other hardware, so that the disaster tolerance of an internal module of the industry gateway VNF is ensured.

Reliability, the internal components of the VNF of the industry gateway mainly adopt 2 ways to ensure the Reliability of the service.

The double computers (1+1) comprise a charging module, an operation maintenance system, a real-time report system and the like. The dual-machine management of the two virtual machines is realized through dual-machine software, and the two virtual machines are deployed on the same physical machine. The two virtual machines are not on the same physical machine through the exclusive function of the cloud management software. And multiple nodes (N +1), such as the service processing modules of A and M, a log server and the like. By realizing a multi-node deployment mode, the parallel processing of the business process is realized. If one node is abnormal, the node deployed in the N +1 mode can migrate the service to other nodes for continuous processing.

The resource pool system is composed of a storage network, a service network, a management network and an external gateway, wherein:

a storage network: and deploying the storage equipment and the storage switch, wherein the storage equipment is proposed to be deployed by adopting an IP SAN, and the storage switch adopts IP switching equipment.

Service network: the system consists of a computing node and a service switch, wherein the computing node is provided with a standard X86 server and is deployed with computing virtualization software and virtual switch software.

A management network: a management node and a management switch are deployed.

The resource pool comprises hardware resources, VM, Hypervisor and a resource pool management VIM.

Hardware resources

The hardware resources are divided into three parts, namely computing resources, storage resources and network resources, the computing resources refer to a local general physical server, and the general physical server (COTS) comprises a CPU, a memory, a local disk, a network card and the like, and may also comprise accelerated hardware (such as hardware encryption and decryption, packet switching, packet forwarding acceleration). The storage resource refers to an IP-SAN disk array externally connected for storage. The network resources refer to network communication connection equipment such as switches and routers.

Virtual resources

The Virtual resources of each service module are mainly embodied in the form of Virtual machines (Virtual machines). A virtual machine has virtual computing resources such as virtual cpu (vcpu), virtual storage resources such as virtual memory, virtual disk, and virtual network resources such as virtual network card. The virtual machine can have different specifications, the specification of the virtual machine is described by a resource template, and the specification of the virtual machine can be configured and managed.

The virtual machines are provided by the Hypervisor on the universal physical server in the hardware resources, and the Hypervisor separates the universal physical server from the upper software application, so that a plurality of virtual machines with different operating systems can run on the same physical server, and the hardware resources are utilized to the maximum extent, namely the hardware resources of one physical server can be shared by a plurality of virtual machines. The Hypervisor can interact with the cloud management system to realize the functions of creating, deleting and other operations of the virtual machine, fault management, performance management and the like.

Resource pool management system

VIM: the system is responsible for virtualized infrastructure management, and has the main function of realizing management and monitoring of resources of the whole infrastructure layer.

The VNFM provides management of VNF instances based on resource pools, lifecycle management of VNF instances/functionality of resource utilization

The NFVO provides network element function virtualization coordination based on the resource pool, and is responsible for providing views of hardware resources and virtual resources, monitoring, performance statistics and fault management of the hardware resources and the virtual resources, controlling the VNFM to realize management of a VNF software package, and creation, update, termination and elastic expansion of a VNF instance.

The virtualized network element function VNF provides each service function based on the resource pool, and the VNF can be deployed on one or more virtual machines VM to meet the network element function requirements defined by each service technical requirement.

Network element module configuration scaling capability and operation mode

And the number of servers operated in the resource pool can be increased or reduced according to the actual service requirement. The resource scheduling function of the resource pool can monitor the load of each host in the cluster, and if the load of the host exceeds the migration threshold of the preset strategy, load balancing is performed, and the virtual machine on the host with higher load is migrated to the host with lower load. The user can place the appointed multiple virtual machines on the same host according to the host priority scheduling policy. The user may also schedule the virtual machine manually. When the resource utilization rate in the cluster is low, the virtual machines running on some hosts are selected to be intensively migrated to several hosts, so that the hosts keep higher loads; and the idle host is manually or automatically set to be in a shutdown state, so that the energy consumption of the cluster is reduced.

Meanwhile, the system also supports off-line and on-line adjustment of virtual machine resources:

offline adjustment of virtual machine resources:

the system supports increasing or decreasing the number of virtual machine CPUs when the virtual machine is shut down.

The system supports adjusting the memory resources of the virtual machine when the virtual machine is closed.

The system supports adjusting the network card resources of the virtual machine when the virtual machine is closed.

The system supports increasing or decreasing of virtual machine hard disk resources when the virtual machine is closed.

Adjusting virtual machine resources on line:

the system supports the dynamic increase of the number of virtual machine CPUs during the running of the virtual machine.

The system supports dynamic increase of the memory amount of the virtual machine when the virtual machine runs.

The system supports that the number of the network cards of the virtual machine is dynamically increased or decreased when the virtual machine runs.

The system supports dynamic increase or decrease of the number of virtual machine hard disks when the virtual machine runs.

For the computer communication center, the signaling front-end module and the service processing module can automatically or manually pop out a new virtual machine through KPI (key performance indicator) control of a VNFM (virtual network function) to realize capacity expansion of the service and recover virtual machine resources to realize capacity reduction of the service.

For the computer communication gateway, the service processor and interface module can automatically or manually eject a new virtual machine through the KPI index control of the VNFM to realize the capacity expansion of the service and recover the virtual machine resources to realize the capacity reduction of the service.

And in the process of popping up a new virtual machine, the business operation is not influenced. When the virtual machine resources are recovered, graceful offline is supported, that is, the virtual machine module to be recovered receives the offline preprocessing message sent by the VNFM first, then a new service request is not sent to the virtual machine module to be recovered, and the virtual machine module to be recovered formally offline after processing all current service requests.

The above is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, many variations and modifications can be made without departing from the inventive concept of the present invention, which falls into the protection scope of the present invention.

Claims

1. An apparatus for performing a method in accordance with computer instructions, comprising: the device of the computer instruction execution method comprises a signaling front-end processor, wherein the signaling front-end processor is connected with a charging module and an operation maintenance module, the charging module is connected with a service processing module and a GDB, the service processing module is connected with an SMPP interface module and a DCACHE, and the operation maintenance module is connected with the GDB, the charging module, the DCACHE and a QAS; the signaling front-end processor is responsible for a module for processing SS7 signaling and is responsible for message butt joint of the computer communication center and the core network; the service processing module is a computer communication service processing logic module; the charging module is a ticket sorting module; the SMPP interface module is an SMPP protocol interface machine; the operation maintenance module is an operation maintenance platform; QAS is an inquiry analysis module, providing computer communication inquiry and statistical data; the GDB is responsible for storing user account opening data and the like; the DCACHE is a distributed computer communication logic memory database and is used for storing cache data of computer communication, such as computer communication to be sent and timing queues.

2. The apparatus of claim 1, wherein the computer instructions perform a method further characterized by: aiming at a service processing module VNFC, the device of the computer instruction execution method supports the following elastic expansion model by a system: the method comprises the following steps that automatic elastic expansion and contraction are carried out, a computer communication center VNF management module (VNFM) triggers elastic expansion and contraction of a business processor module according to a rule of VNFD, and the triggering rule comprises the following steps: the service processing module uses the resource use condition of the virtual machine, the event condition received from the service processing module and the like; the method comprises the steps that a VNF instance of a computer communication center monitors the state of a VNFC of a service processor according to the scaling of the VNF, and elastic scaling of a service processing module is carried out by sending a clear request for increasing or decreasing the VNFC instance of the service processor to a VNFM; and according to the request expansion of the management module, manually triggering the expansion or sending a request to the NFVO through a certain interface by the OSS/BSS, and triggering the elastic expansion of the service processor according to the rule of the VNFD.

3. The apparatus of claim 1, wherein the computer instructions perform a method further characterized by: the device of the computer instruction execution method is provided with three parts of SMSC signaling gateway clouding: a fixed virtual machine part of a signaling gateway foreground, an elastic virtual machine part of the signaling gateway foreground and a control part of the signaling gateway background; the signaling gateway background control part and the signaling gateway background operation maintenance part are accessed to observe the running condition of the signaling gateway, integrate the OMU function and are responsible for interacting with VNFM and DCF to complete the instantiation and service reporting functions; the fixed virtual machine part of the signaling gateway foreground is a necessary part of the signaling gateway virtual machine, is not telescopic and consists of two modules. The OMP is a master control module and is the master control logic of various service functions. The SLB thread is in charge of the elastic expansion function and interacts with an OMU of the SIU background control part to complete the SIU instantiation and elastic expansion functions. The SIPI module provides an external network interface, and comprises an external network element which is butted by using an SCTP (stream control transmission protocol) and a service processing module which is interacted by using a TCP (transmission control protocol); the elastic virtual machine part of the signaling gateway foreground is an essential part of the signaling gateway virtual machine and can be elastically stretched. And the VNFM judges whether elastic expansion is needed or not according to the indexes reported by the OMU. And after the expansion is finished, reporting the CMP service state to the DCF by the OMU.