WO2016054906A1 - 计费方法和设备、接入设备、服务质量控制方法和设备 - Google Patents

计费方法和设备、接入设备、服务质量控制方法和设备 Download PDF

Info

Publication number
WO2016054906A1
WO2016054906A1 PCT/CN2015/076985 CN2015076985W WO2016054906A1 WO 2016054906 A1 WO2016054906 A1 WO 2016054906A1 CN 2015076985 W CN2015076985 W CN 2015076985W WO 2016054906 A1 WO2016054906 A1 WO 2016054906A1
Authority
WO
WIPO (PCT)
Prior art keywords
host
message
backup machine
response
input
Prior art date
Application number
PCT/CN2015/076985
Other languages
English (en)
French (fr)
Inventor
朱建
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP15848459.2A priority Critical patent/EP3193482B1/en
Publication of WO2016054906A1 publication Critical patent/WO2016054906A1/zh
Priority to US15/481,962 priority patent/US10567507B2/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1095—Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/14—Error detection or correction of the data by redundancy in operation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/14—Error detection or correction of the data by redundancy in operation
    • G06F11/1402—Saving, restoring, recovering or retrying
    • G06F11/1415—Saving, restoring, recovering or retrying at system level
    • G06F11/1438—Restarting or rejuvenating
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/14—Error detection or correction of the data by redundancy in operation
    • G06F11/1402—Saving, restoring, recovering or retrying
    • G06F11/1446—Point-in-time backing up or restoration of persistent data
    • G06F11/1456—Hardware arrangements for backup
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/14—Error detection or correction of the data by redundancy in operation
    • G06F11/1479—Generic software techniques for error detection or fault masking
    • G06F11/1482—Generic software techniques for error detection or fault masking by means of middleware or OS functionality
    • G06F11/1484—Generic software techniques for error detection or fault masking by means of middleware or OS functionality involving virtual machines
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44—Arrangements for executing specific programs
    • G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533—Hypervisors; Virtual machine monitors
    • G06F9/45558—Hypervisor-specific management and integration aspects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/54—Interprogram communication
    • G06F9/542—Event management; Broadcasting; Multicasting; Notifications
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • H04L67/61—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources taking into account QoS or priority requirements
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16—Error detection or correction of the data by redundancy in hardware
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44—Arrangements for executing specific programs
    • G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533—Hypervisors; Virtual machine monitors
    • G06F9/45558—Hypervisor-specific management and integration aspects
    • G06F2009/45591—Monitoring or debugging support
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • H04L67/10015—Access to distributed or replicated servers, e.g. using brokers

Definitions

  • the present invention relates to the field of communications technologies, and in particular, to a message processing method, apparatus, and message processing system on a cloud computing platform.
  • Telecom services such as call services and billing services
  • have extremely high reliability requirements reaching five 9 (99.999%), far greater than enterprise-level reliability.
  • the reliability of cloud computing systems is about 95% to 98%, which is far lower than the reliability requirements of telecommunication services, but cloudization is a general trend. How to maintain its telecommunications after telecom services migrate to the cloud. Level reliability is a technical problem that needs to be solved.
  • the prior art provides a two-machine cloud solution, in which dual machines (hosts and backup machines) are respectively deployed in the cloud, and the host performs external requests, and synchronizes with the backup machines after performing external requests, and the synchronization process includes data. Synchronization, status synchronization.
  • the embodiment of the invention provides a message processing method, device and message processing system for improving backup reliability in a dual-machine backup scenario.
  • an embodiment of the present invention provides a message processing method on a cloud computing platform, including: receiving an input request message, sending the request message to a host deployed on the cloud computing platform, and a backup machine on which the same application runs on the backup machine;
  • the receiving, by the receiving host and the backup machine, multiple input response messages, and sending one of the input response messages to the external information includes:
  • the method further includes: updating the health of the host and the backup machine according to the received input response message, the host and the backup machine The health is calculated based on the number of input response messages sent by the respective responses that carry the correct response.
  • the method further includes:
  • Receiving an output response message sending the output response message to the host and the backup machine.
  • the first possible implementation manner of the first aspect, the second possible implementation manner of the first aspect, and the third possible implementation manner of the first aspect, in a fourth possible implementation manner further includes:
  • an embodiment of the present invention provides a message processing apparatus on a cloud computing platform, including:
  • An input message processing unit configured to receive an input request message, send the request message to a host and a backup machine deployed on the cloud computing platform, where the host and the backup machine are used to implement dual-system backup, the host and the backup Run the same application on the machine;
  • the input response processing unit is configured to receive a plurality of input response messages returned by the host and the backup machine, and send one of the input response messages to the outside.
  • the input response processing unit receives multiple input response messages returned by the host and the backup machine, and sending one of the input response messages to the external device includes:
  • the input response processing unit receives the plurality of input response messages returned by the host and the backup machine, and sends the first arriving response message outward;
  • the input response processing unit receives a plurality of input response messages returned by the host and the backup machine, where the input response message carries a response result, and determines a first type of input response message corresponding to the response result with the highest number of repetitions, and sends the An input response message in the first type of input response message; or
  • the input response processing unit receives a plurality of input response messages returned by the host and the backup machine, where the input response message carries a response result, and when the response results of the multiple input response messages are different, the predefined host is selected. Or the backup machine's input response message is sent out; or
  • the input response processing unit receives a plurality of input response messages returned by the host and the backup machine, and sends an input response message returned by the most healthy virtual machine in the host and the backup machine.
  • the method further includes an output message processing unit, configured to receive multiple output request messages sent by the host and the backup device, to Send one of the output request messages outside;
  • an output response processing unit configured to receive an output response message, and send the output response message to the host and the backup machine.
  • the method further includes:
  • a storage message processing unit configured to receive multiple storage access requests sent by the host and the backup machine, Sending one of the storage access requests to the I-tier storage device shared by the host and the backup machine;
  • a storage response processing unit configured to receive a storage access response returned by the layer I storage device, and send the storage access response to the host and the backup machine.
  • the embodiment of the present invention further provides a message processing system on a cloud computing platform, including a host and a backup machine deployed on the cloud computing platform, and a message processing apparatus according to the second aspect.
  • the embodiment of the present invention After receiving an input request message addressed to the host and the backup machine, the input request message is sent to all hosts and backup machines, and the host and the backup are received. After the machine returns an input response message, an input response message is sent out.
  • the embodiment of the present invention sends an input request to all hosts and backup machines for execution, and the host and the backup machine perform the same input request, so data synchronization and state synchronization are not required between the host and the backup machine. Thereby improving the stability of the dual-machine backup and further improving the switching speed of the dual-machine.
  • FIG. 1 is a schematic diagram of deployment of a host and a backup machine in an embodiment of the present invention
  • FIG. 2 is a flowchart of a message processing method according to Embodiment 1 of the present invention.
  • FIG. 3 is a flowchart of a message processing method according to Embodiment 2 of the present invention.
  • FIG. 5 is a flowchart of a message processing method according to Embodiment 4 of the present invention.
  • FIG. 6 is a schematic structural diagram of a message processing apparatus according to Embodiment 5 of the present invention.
  • FIG. 7 is a schematic structural diagram of a message processing apparatus according to Embodiment 6 of the present invention.
  • FIG. 8 is a schematic structural diagram of a message processing system according to Embodiment 7 of the present invention.
  • FIG. 1 is a schematic diagram of deployment of a host and a backup machine in an embodiment of the present invention.
  • the host and the backup machine are virtual machines (also referred to as virtualized containers) deployed on the cloud computing platform, and the same application is run on the host and the backup machine, and the application (App1) is run in the virtualized container 1 as The host, virtualized container 2 runs an image of the application (App1) as a backup machine.
  • the number of backup machines can be flexibly increased according to requirements.
  • the concept of the host and the backup machine is relative. All virtual machines can be called hosts, and all virtual machines can be called backup machines.
  • the virtualization container 1 and the virtualization container 2 use network resources, storage resources, and computing (CPU) resources of the cloud computing IaaS layer (referred to as the I layer).
  • the virtualization container 1 and the virtualization container 2 may share the same network resource and storage resource, and may also use independent network resources and storage resources.
  • a message processing device also understood as an interface adaptation layer is also included between the application and the IaaS layer.
  • FIG. 2 is a flowchart of a message processing method on a cloud computing platform according to Embodiment 1 of the present invention.
  • a message processing method on a cloud computing platform mainly includes:
  • Step 101 Receive an input request message, send the request message to a host and a backup machine deployed on a cloud computing platform, and run the same application on the host and the backup machine.
  • the message processing apparatus receives an input request message, such as an accounting request, externally ("external” refers to the host and the backup machine), and copies and sends the request message to the host and the backup machine, the host and the backup machine.
  • the input request message is independently executed and the corresponding input response message is returned.
  • the received input request message may be sent to the host and/or the backup machine, or may be sent to the message processing device, that is, the destination address of the input request message may be the host and/or the backup machine, or may be the implementation.
  • the message processing device can receive the input request message.
  • the host and the backup machine can have the same address and can also have mutually independent addresses.
  • Step 102 Receive multiple input response messages returned by the host and the backup machine, and send one of the input response messages to the outside.
  • the message processing apparatus may identify the repeated input response message and send one of the input response messages to the outside.
  • the embodiment of the present invention can receive multiple input response messages returned by the host and the backup machine, and send the first arriving response message to improve the message sending efficiency. After the first arriving response message is sent out, the message that is duplicated with the first returned input response message can be quickly identified and discarded according to the message characteristics of the input response message (eg, session identifier).
  • the message processing device can also identify the input response message sent by the host and the backup machine by means of complete message matching.
  • the embodiment of the present invention may further receive a plurality of input response messages returned by the host and the backup machine, where the input response message carries a response result, and the embodiment of the present invention determines the first type of input response message corresponding to the response result with the highest number of repetitions, An input response message in the first type of input response message is sent out, thereby improving the correctness of the response message transmission.
  • the response result may be included in the message body of the response message, and may also be directly the response message itself.
  • the embodiment of the present invention may further receive a plurality of input response messages returned by the host and the backup machine, where the input response message carries a response result, and when determining that the response results of the multiple input response messages are different, selecting a predefined
  • the input response message of the host or the backup machine is sent out, that is, it is sent out according to the input response message of a predetermined virtual machine.
  • the message processing apparatus in order to enable the message processing apparatus to receive all input request messages addressed to the host or the backup machine, the message processing apparatus may be configured as a necessary routing node.
  • the message processing device can intercept the input request message sent to the host and the backup machine and intercept the multiple input response messages returned by the host and the backup machine.
  • the message processing apparatus may identify the repeated input response message according to the message feature, and may further Features to match input request messages and input response messages.
  • a globally unique message feature for example, a session identifier session-id
  • the message processing apparatus may identify the repeated input response message according to the message feature, and may further Features to match input request messages and input response messages.
  • the specific description of the message features may be applied to other embodiments of the first embodiment of the present invention.
  • the message processing device may add some associated information to the input request message to be sent, for example,
  • the input request message sent to each virtual machine adds a virtual machine identifier (such as a process ID of the virtual machine, a MAC address of the virtual machine) and a serial number (the serial number may correspond to an input request message), and then receives the message.
  • a virtual machine identifier such as a process ID of the virtual machine, a MAC address of the virtual machine
  • serial number may correspond to an input request message
  • the message processing apparatus may identify the repeated input response message according to the serial number in the associated information, according to the The virtual machine identifier in the associated message identifies an input response message returned by a different virtual machine, and may also match the input request message and the input response message according to the associated information. Since the association information is added to the interface layer of the I layer, the message processing apparatus may not carry the association information when transmitting the input request message to the host and the backup machine.
  • Step 103 Rebuild the host and the backup machine when an abnormality occurs in the response result of the multiple input response messages.
  • the response result of the multiple input response messages is abnormal, for example, the response results of the multiple input response messages are different, or only the response messages with different partial response results are present in the multiple input response messages, that is, The response result of the other part of the input response message is the same, indicating that some of the virtual machines in the host and the backup machine may be faulty, and the host and the backup machine are rebuilt at this time.
  • the rebuild host and the backup machine can rebuild the host and the backup machine according to pre-defined rules (such as reinstalling the application, restoring data, etc.), so that the host and the backup machine are restored.
  • the virtual machine corresponding to the first type of input response message is reconstructed according to the virtual machine corresponding to the first type of input response message.
  • the host or the backup machine can restore the normality of the host or backup machine other than the virtual machine corresponding to the first type of input response message, and reduce the number of virtual machine reconstructions, thereby indirectly improving the reliability of the backup.
  • the host and the backup machine in the embodiment of the present invention may respectively correspond to the health degree (or the credibility), and the health degree is determined according to the proportion of the total number of returned input response messages and the total number of returned input response messages. .
  • the embodiment of the present invention determines whether the input response message is correct, and updates the health of the corresponding host and the backup machine according to the judgment result (the health of the input response message is correct) And, when the input response message is sent out, the input response message returned by the highly healthy virtual machine is sent. Among them, determine whether the input response message is correct The process may be based on the principle that the minority obeys the majority, for example, counting the number of times the response result appears in each input response message, determining the input response message corresponding to the most frequently occurring response result as correct, and determining the other response result as an error.
  • the specific description of the health of the host and the backup machine in the embodiment of the present invention may be applied to other embodiments of the first embodiment of the present invention.
  • the host or the backup machine whose health is lower than the threshold is reconstructed. For example, rebuilding the virtual machine according to a predefined rule or rebuilding a host or backup machine whose health is below a threshold based on the most healthy virtual machine.
  • FIG. 3 is a flowchart of a message processing method on a cloud computing platform according to Embodiment 2 of the present invention.
  • a message processing method on a cloud computing platform mainly includes:
  • Step 201 Receive multiple output request messages sent by the host and the backup machine, and send one of the output request messages to the outside.
  • the host running the application and the backup device send an output request message, for example, sending a location request message to the terminal.
  • the embodiment of the present invention sends one of the output request messages, for example, the most The output request message received first.
  • Embodiments of the present invention may identify duplicate output request messages based on message characteristics of the output request message, for example, based on the session identification.
  • the host and the backup machine may add a sequence number of the message to the output request message sent to the message processing device, and the message processing device quickly identifies the duplicate output request message according to the sequence number.
  • the message processing device in order for the message processing device to receive the output request message sent by all the hosts or the backup machine, the message processing device can be configured as a necessary routing node, and the message processing device can intercept all hosts and backups. Output request message sent by the machine.
  • Step 202 Receive an output response message, and send the output response message to the host and the backup machine.
  • the embodiment of the present invention sends the received location response message to the host and the backup machine after receiving an externally returned output response message, such as a location response message.
  • the host and the backup machine perform data processing according to the received output response message.
  • the embodiment of the present invention may further store the processing result after performing data processing, that is, the host and the backup machine send a storage access request to the storage device of the layer I.
  • the message processing apparatus may send an output response message to the host and the backup machine according to the locally maintained routing table, or broadcast the output response message to the host and the backup machine.
  • the output request message in the second embodiment of the present invention may be an output request message actively triggered by the host and the backup machine, and may also be an output request returned by the host and the backup machine according to the input request message received in the first embodiment.
  • the message, at this time, the input response message returned by the host and the backup machine is the output request message.
  • the message processing method in the first embodiment further performs the step 202 of receiving an output response message after performing the step 102 of transmitting one of the input response messages.
  • Step 203 Receive a plurality of storage access requests sent by the host and the backup machine, and send one of the storage access requests to the I-level storage device shared by the host and the backup machine.
  • the sharing when the host and the backup machine share the same I-level storage device, when receiving a plurality of storage access requests sent by the host and the backup machine (for example, according to a storage access request returned by the output response message), the sharing is performed.
  • the I-tier storage device sends one of the storage access requests, for example, the first storage access request that was sent.
  • the message processing device can also send a storage access request for the most healthy virtual machine.
  • the message processing apparatus if the storage access request carries the message feature, the message processing apparatus identifies the repeated storage access request according to the message feature.
  • the host and the backup machine may further increase the sequence number of the access request in the issued storage access request, and the message processing device quickly identifies the repeated storage access request according to the serial number.
  • the above serial number can be added to the interface layer of the I layer to facilitate identification by the message processing apparatus.
  • the host and the backup device do not share the same I-level storage device, that is, the host and the backup device are configured with independent storage devices, the host and the backup device respectively send storage access to the corresponding storage device.
  • the storage access response is requested and received, and the message processing device does not process the storage access request.
  • Step 204 Receive a storage access response returned by the layer I storage device, and send the storage access response to the host and the backup machine.
  • the embodiment of the present invention After receiving the storage access response returned by the I-layer storage device, the embodiment of the present invention matches the corresponding storage access request according to the message feature or the serial number in the storage access response, and then copies and sends the storage access response to the host and the backup machine. So that the host and backup machine can get the storage access result.
  • the embodiment of the present invention After receiving the output request message sent by the host and the backup machine, sending one of the output request messages, receiving After outputting a response message to the external device, the output response message is sent to the host and the backup machine.
  • the embodiment of the present invention sends an output response message to all hosts and backup machines, and the host and the backup machine perform the same output response message, so that data synchronization and state synchronization are not required between the host and the backup machine. Therefore, the stability of the dual-machine backup is improved, and the switching speed of the dual-machine is further improved.
  • the embodiment of the present invention further sends one of the shared I-level storage devices after receiving the storage access request, and sends the received storage access response to all the hosts and the backup machine, thereby implementing the shared storage device. access.
  • the host and the backup device in the embodiment of the present invention can share the I-layer storage device, which reduces the redundancy of the storage resources in the dual-system backup scenario.
  • FIG. 4 is a flowchart of a message processing method on a cloud computing platform according to Embodiment 3 of the present invention.
  • the application eg, billing application
  • the convergent billing part (CBP) in the billing application is taken as an example to illustrate that the virtualized container in which the billing application is located shares the storage resource, and the network resource and the storage resource of the virtualized container are independently deployed.
  • the consolidated billing part is deployed on both backup machines 1 and 2.
  • the message processing device After the host and backup machine are deployed, the message processing device stores the routing table locally for subsequent message transmission.
  • Step 301 Receive an external charging request message.
  • the message requesting device receives the message feature, and the message feature may be a session identifier (session-id) and a request request number (request number) in this embodiment.
  • the charging request message sent by different external devices may be distinguished by the identifier of the external device (for example, the user equipment identifier).
  • the destination address of the external charging request message may be a host and/or a backup machine, and may also be the message processing apparatus in this embodiment, and the message processing apparatus may receive the external charging request message.
  • Step 302 Send a charging request message to the host and the backup machine.
  • the message processing device sends the charging request message to the host and the backup machine 1, 2 according to the local routing table.
  • Step 303 The host and the backup machine process the charging request message, and return an accounting response message.
  • the host and the backup machine 1 and 2 independently perform the charging process according to the charging request, and return an accounting response message, and the message processing apparatus performs step 304.
  • step 306 after the processing of the charging request message, the host and the backup machine 1 and 2 need to write data to the I-layer storage device, that is, step 306 is performed.
  • Step 304 Receive an accounting response message sent by the host and the backup device, and send a first type of charging response message.
  • the message processing device may identify, by using a session identifier (session-id) and a request number, a charging response message returned by different virtual machines for the same charging request message.
  • session-id session-id
  • request number a charging response message returned by different virtual machines for the same charging request message.
  • the host or the backup machine may perform different charging results in the charging response message.
  • the message processing device obtains the first type corresponding to the charging result with the highest number of repetitions.
  • the charging response message is then selected from the first type of charging response message to be sent out.
  • Step 305 Rebuild a host or a backup machine other than the virtual machine corresponding to the first type of charging response message according to the virtual machine corresponding to the first type of charging response message.
  • the message processing device reconstructs a virtual machine (which may be a host or a backup machine) other than the virtual machine corresponding to the first type of the charging response message according to the virtual machine corresponding to the first type of the charging response message, so that the first type of charging response message is corresponding.
  • the host or backup machine other than the virtual machine is restored to normal, thereby improving the reliability of the dual-machine backup.
  • the embodiment of the present invention rebuilds the virtual machine when the virtual machine has a significant running abnormality (for example, the response message cannot be returned), for example, rebuilding the unreturnable response message according to the virtual machine corresponding to the first type of charging response message. virtual machine.
  • Step 306 The host and the backup machine send a data write request to the layer I storage device.
  • the host and backup machines 1, 2 also need to perform charging information storage to the I-tier storage device, and send a storage access request to the I-tier storage device.
  • the message processing apparatus can intercept all storage access requests, and the storage access request is specifically a data write request.
  • the data in the data write request sent by the host and the backup machines 1, 2 may be inconsistent.
  • Step 307 Receive a plurality of data write requests sent by the host and the backup machine, and send a first type of data write request to the I layer storage device.
  • the message message in the data write request is characterized by a keyword in the data write request (corresponding to different data to be written).
  • the message processing device After receiving the data write request, the message processing device identifies the data write request sent by the host and the backup machine 1, 2 to the layer I storage device by the keyword in the data write request.
  • the data write request carries the data to be written
  • the embodiment of the present invention determines the first type of data write request corresponding to the data to be written with the highest number of repetitions from the plurality of data write requests having the same keyword. And sending a data write request in the first type of data write request to the I layer storage device, thereby improving the correctness of data writing.
  • the layer I storage device writes the billing data according to the received data write request and returns a data write response.
  • Step 308 Receive a data write response returned by the I layer storage device.
  • the message processing device receives the data write response returned by the layer I storage device.
  • Step 309 Send the data write response to the host and the backup machine.
  • the message processing device After receiving the data write response corresponding to the data write request, the message processing device matches the corresponding data write request according to the keyword, and then sends and receives the received data to the host and the backup machine 1, 2 according to the internally stored routing table. The data is written to the response to complete the data writing process.
  • FIG. 5 is a flowchart of a message processing method on a cloud computing platform according to Embodiment 4 of the present invention.
  • an application (such as a call processing application) runs in a virtualized container that takes over all of the application's resource access requests, including network resources, disk resources, and CPU resources.
  • a service control part (SCP) in a call processing application is taken as an example to illustrate that a service control part is deployed on both the host and the backup machines 1 and 2.
  • Step 501 Receive an externally reported call control request.
  • the message processing device receives the call control request reported by the external Service Switching Point (SSP).
  • SSP Service Switching Point
  • Step 502 Send a call control request to the host and the backup machine.
  • the message processing device sends a call control request to all according to the internally stored routing table. Host and backup machine.
  • the host and the backup machine independently perform call control, return call control response, and call identifier (dialog id) of the Transaction Capabilities Application Part (TCAP) layer in the call control response (CAPEL Application Part, CAP) protocol.
  • TCAP Transaction Capabilities Application Part
  • CAPEL Application Part, CAP call control response
  • Step 503 Receive a call control response sent by the host and the backup machine.
  • the message processing device can receive the call control responses returned by all the hosts and backup machines 1, 2.
  • Step 504 Send a call control response returned by the most healthy virtual machine.
  • the message processing device stores health data of each host and the backup machine, and the health degree data is calculated according to the historical call control response, for example, the returned call control according to the returned call control response correct times The proportion of the total number of responses is determined.
  • the message processing apparatus transmits the call control response returned by the most healthy virtual machine, distinguishes the repeated call control response according to the session identifier of the TCAP layer of the CAP protocol in the call control response, and discards, thereby completing the call control process.
  • Step 505 Update the health of the corresponding host and the backup machine according to the received call control response.
  • the embodiment of the present invention performs correctness judgment on the call control responses returned by all the hosts and the backup machine, that is, counts the number of repeated call control results in all call control responses.
  • the call control response corresponding to the most frequently repeated response result is determined to be the correct call control response (if all the call control responses in the call control response are different, all call control responses can be determined as the wrong call)
  • the control response is determined, and the call control response corresponding to the other response results is determined as an erroneous call control response, thereby updating the health of the host and the backup machine corresponding to the call control response, so as to determine the highest health when transmitting the call control response.
  • Virtual machine
  • the host or backup machine does not return a call control response, it can be determined that the virtual machine returns an incorrect call control response and updates the health of the virtual machine.
  • Step 506 Rebuild a host or a backup machine whose health is lower than a threshold.
  • the virtual machine is rebuilt. For example, rebuilding the virtual machine according to a predefined rule or rebuilding a host or backup machine whose health is below a threshold based on the most healthy virtual machine.
  • the host and the backup machine simultaneously perform call requests, and the dual-machine backup improves the stability of the call control process.
  • embodiments of the present invention can reduce the cost of telecommunication hardware through the scale of low-cost virtual machines.
  • FIG. 6 is a schematic structural diagram of a message processing apparatus according to Embodiment 5 of the present invention.
  • the message processing apparatus provided by the embodiment of the present invention mainly includes:
  • the input request processing unit 610 is configured to receive an input request message, send the request message to a host and a backup machine deployed on the cloud computing platform, where the host and the backup machine run the same application, the host and the backup The machine is used to implement dual-machine backup;
  • the input response processing unit 620 is configured to receive a plurality of input response messages returned by the host and the backup machine, and send one of the input response messages to the outside.
  • the input response processing unit 620 receives a plurality of input response messages returned by the host and the backup machine, and the sending of one of the input response messages to the external device may include:
  • the input response processing unit 620 receives the plurality of input response messages returned by the host and the backup machine, and sends the first arriving response message outward; or
  • the input response processing unit 620 receives a plurality of input response messages returned by the host and the backup machine, where the input response message carries a response result, and determines a first type of input response message corresponding to the response result with the highest number of repetitions, and sends the An input response message in the first type of input response message; or
  • the input response processing unit 620 receives a plurality of input response messages returned by the host and the backup machine, where the input response message carries a response result, and when the response results of the multiple input response messages are different, the predefined one is selected.
  • the input response message of the host or backup machine is sent out; or
  • the input response processing unit 620 receives a plurality of input response messages returned by the host and the backup machine, and sends an input response message returned by the most healthy virtual machine in the host and the backup machine.
  • step 102 For the specific processing of the input response processing unit 620, refer to the related description of step 102 in the first embodiment, and details are not described herein again.
  • the message processing apparatus provided by the embodiment of the present invention further includes:
  • the virtual machine reconstruction unit 630 is configured to rebuild the host and the backup machine when an abnormality occurs in the response result of the multiple input response messages, or rebuild the health when the health of the host or the backup machine is lower than a threshold A host or backup machine whose health is below the threshold.
  • the abnormal situation includes: the response results of the multiple input response messages are different, or there are only some response messages with different response results in the multiple input response messages, that is, the response results of the other partial input response messages are the same, and the virtual machine is reconstructed.
  • the unit 630 to rebuild the virtual machine refer to the description of step 103 in the first embodiment, and details are not described herein again.
  • the health update unit 640 is configured to update the health of the host and the backup machine according to the received input response message, and the health of the host and the backup machine is based on the number of input response messages respectively sent to carry the correct response result. Calculated.
  • the health update unit 640 may be specifically calculated according to the total amount of the input response message sent by the virtual machine and the input response message carrying the correct response result.
  • the output request processing unit 650 is configured to receive a plurality of output request messages sent by the host and the backup machine, and send one of the output request messages to the outside; the output response processing unit 660 is configured to receive the output response message to the host and the backup machine. Sending the output response message.
  • the specific processing of the output request processing unit 650 and the output response processing unit 660 for the output request message and the output response message are described in the foregoing description of the steps 201-202 in the second embodiment, and details are not described herein again.
  • the storage request processing unit 670 is configured to receive a plurality of storage access requests sent by the host and the backup machine, and send one of the storage access requests to the I-level storage device shared by the host and the backup machine; and the storage response processing unit 680 is configured to receive And storing, by the I-tier storage device, a storage access response, and sending the storage access response to the host and the backup machine.
  • the storing response processing unit 680 sends one of the storage access requests to the I-level storage device shared by the host and the backup machine, and the storage response processing unit 680 sends the first-storage storage to the I-level storage device shared by the host and the backup machine. Access request.
  • the specific processing procedure of the storage request processing unit 670 and the storage response processing unit 680 for the storage access request and the storage access response is described in the related description of the steps 203-204 in the second embodiment, and details are not described herein again.
  • the message processing apparatus in the embodiment of the present invention can enable the host and the backup machine to share the I-layer storage device, which reduces the redundancy of the storage resources in the dual-system backup scenario.
  • FIG. 7 is a schematic structural diagram of a message processing apparatus according to Embodiment 6 of the present invention.
  • the message processing apparatus provided by the embodiment of the present invention mainly includes:
  • a processor 710 for storing computer execution instructions, a processor 710 connected to the memory 720 via the bus 730, and the processor 710 when the device is running. Executing the computer to execute an instruction to cause the apparatus to perform the following method:
  • Receiving an input request message sending the request message to a host and a backup machine deployed on the cloud computing platform, where the host and the backup machine run the same application;
  • FIG. 8 is a schematic structural diagram of a message processing system according to Embodiment 7 of the present invention.
  • the message processing system provided by the embodiment of the present invention mainly includes: a host 810, a backup machine 820, and a message processing device 830.
  • the specific structure and function of the message processing device 830 are the same as those of the message processing device provided in the foregoing fifth and sixth embodiments. For details, refer to the foregoing embodiment, and details are not described herein again.
  • the message processing system provided by the embodiment of the present invention sends an input request message to all hosts and backup machines for execution, and the host and the backup machine perform the same input request message, so that data synchronization and state synchronization are not required between the host and the backup machine. Thereby improving the stability of the dual-machine backup and further improving the switching speed of the dual-machine.
  • aspects of the present invention, or possible implementations of various aspects may be embodied as a system, method, or computer program product.
  • aspects of the invention, or possible implementations of various aspects may be in the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, etc.), or a combination of software and hardware aspects, They are collectively referred to herein as "circuits," “modules,” or “systems.”
  • aspects of the invention, or possible implementations of various aspects may take the form of a computer program product, which is a computer readable program code stored in a computer readable medium.
  • the computer readable medium can be a computer readable signal medium or a computer readable storage medium.
  • the computer readable storage medium includes, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing, such as random access memory (RAM), read only memory (ROM), Erase programmable read-only memory (EPROM or flash memory), optical fiber, portable read-only memory (CD-ROM).
  • the processor in the computer reads the computer readable program code stored in the computer readable medium such that the processor is capable of performing the various functional steps specified in each step of the flowchart, or a combination of steps; A device that functions as specified in each block, or combination of blocks.
  • the computer readable program code can execute entirely on the user's computer, partly on the user's computer, as a separate software package, partly on the user's computer and partly on the remote computer, or entirely on the remote computer or server.
  • the functions noted in the various steps in the flowcharts or in the blocks in the block diagrams may not occur in the order noted. For example, two steps, or two blocks, shown in succession may be executed substantially concurrently or the blocks may be executed in the reverse order.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Quality & Reliability (AREA)
  • Multimedia (AREA)
  • Hardware Redundancy (AREA)

Abstract

本发明提供了一种云计算平台上消息处理方法、装置及消息处理系统。在本发明提供的云计算平台上的消息处理方法中，在接收到发往主机和备份机的输入请求消息后，发送输入请求消息到所有的主机和备份机，在接收到主机和备份机返回的输入响应消息后，向外发送一个输入响应消息。与现有技术相比，本发明实施例将输入请求发送到所有的主机和备份机执行，主机和备份机执行同样的输入请求，因而主机和备份机之间不需要进行数据同步和状态同步，从而提高了双机备份的稳定性，进一步提高了双机切换速度。

Description

计费方法和设备、接入设备、服务质量控制方法和设备
本申请要求于2014年10月08日提交中国专利局、申请号为201410528631.0发明名称为“消息处理方法、装置及消息处理系统”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及通信技术领域，具体而言涉及一种云计算平台上的消息处理方法、装置及消息处理系统。
背景技术
电信业务，例如呼叫业务、计费业务对可靠性要求极高，达到5个9(99.999％)，远大于企业级可靠性。根据目前公开的材料看，云计算系统的可靠性大约在95％～98％，远低于电信业务的可靠性要求，但云化是大趋势，如何在电信业务迁移到云后继续保持其电信级可靠性是需要解决的技术问题。
对软件业务来说，提高可靠性的一个重要手段就是热双机。现有技术提供了双机云化的解决方案，将双机(主机和备份机)分别部署在云端，由主机执行外部请求，在执行外部请求之后和备份机之间进行同步，同步过程包括数据同步、状态同步。
但是，现有技术的提供的双机云化方案在需要同步的数据量较大时，双机备份的可靠性降低，导致双机切换延时明显，进而降低业务的可靠性。
发明内容
本发明实施例提供了一种在双机备份场景下提高备份可靠性的消息处理方法、装置及消息处理系统。
第一方面，本发明实施例提供一种云计算平台上的消息处理方法，包括：接收输入请求消息，发送所述请求消息到部署在所述云计算平台上的主机和 备份机，所述主机和备份机上运行相同的应用程序；
接收所述主机和备份机返回的多个输入响应消息，向外发送其中一个输入响应消息。
结合第一方面，在第一种可能的实现方式中，所述接收主机和备份机返回的多个输入响应消息，向外发送其中一个输入响应消息包括：
接收所述主机和备份机返回的多个输入响应消息，向外发送最先到达的响应消息；或者
接收所述主机和备份机返回的多个输入响应消息，所述输入响应消息中携带响应结果，确定重复次数最多的响应结果对应的第一类输入响应消息，向外发送所述第一类输入响应消息中的一个输入响应消息；或者
接收所述主机和备份机返回的多个输入响应消息，所述输入响应消息中携带响应结果，在所述多个输入响应消息的响应结果均不相同时，选择预先定义的主机或备份机的输入响应消息向外发送；或者
接收所述主机和备份机返回的多个输入响应消息，发送所述主机和备份机中健康度最高的虚拟机返回的输入响应消息。
结合第一方面的第一种可能的实现方式，在第二种可能的实现方式中，还包括：根据接收到的输入响应消息更新所述主机和备份机的健康度，所述主机和备份机的健康度根据各自发送的携带正确的响应结果的输入响应消息的数量计算得到。
结合第一方面、第一方面的第一种可能的实现方式或第一方面的第二种可能的实现方式，在第三种可能的实现方式中，还包括：
接收主机和备份机发送的多个输出请求消息，向外发送其中一个输出请求消息；
接收输出响应消息，向所述主机和备份机发送所述输出响应消息。
结合第一方面、第一方面的第一种可能的实现方式、第一方面的第二种可能的实现方式、第一方面的第三种可能的实现方式，在第四种可能的实现方式中，所述方法还包括：
接收主机和备份机发送的多个存储访问请求，向所述主机和备份机共享的I层存储设备发送其中一个存储访问请求；
接收所述I层存储设备返回的存储访问响应，向所述主机和备份机发送 所述存储访问响应。
第二方面，本发明实施例提供一种云计算平台上的消息处理装置，其包括：
输入消息处理单元，用于接收输入请求消息，发送所述请求消息到部署在所述云计算平台上的主机和备份机，所述主机和备份机用于实现双机备份，所述主机和备份机上运行相同的应用程序；
输入响应处理单元，用于接收所述主机和备份机返回的多个输入响应消息，向外发送其中一个输入响应消息。
结合第二方面，在第一种可能的实现方式中，所述输入响应处理单元接收主机和备份机返回的多个输入响应消息，向外发送其中一个输入响应消息包括：
输入响应处理单元接收所述主机和备份机返回的多个输入响应消息，向外发送最先到达的响应消息；或者
输入响应处理单元接收所述主机和备份机返回的多个输入响应消息，所述输入响应消息中携带响应结果，确定重复次数最多的响应结果对应的第一类输入响应消息，向外发送所述第一类输入响应消息中的一个输入响应消息；或者
输入响应处理单元接收所述主机和备份机返回的多个输入响应消息，所述输入响应消息中携带响应结果，在所述多个输入响应消息的响应结果均不相同时，选择预先定义的主机或备份机的输入响应消息向外发送；或者
输入响应处理单元接收所述主机和备份机返回的多个输入响应消息，发送主机和备份机中健康度最高的虚拟机返回的输入响应消息。
结合第二方面、第二方面的第一种可能的实现方式，在第二种可能的实现方式中，还包括输出消息处理单元，用于接收主机和备份机发送的多个输出请求消息，向外发送其中一个输出请求消息；
输出响应处理单元，用于接收输出响应消息，向所述主机和备份机发送所述输出响应消息。
结合第二方面、第二方面的第一种可能的实现方式或第二方面的第二种可能的实现方式，在第三种可能的实现方式中，还包括：
存储消息处理单元，用于接收主机和备份机发送的多个存储访问请求， 向所述主机和备份机共享的I层存储设备发送其中一个存储访问请求；
存储响应处理单元，用于接收所述I层存储设备返回的存储访问响应，向所述主机和备份机发送所述存储访问响应。
第三方面，本发明实施例还提供一种云计算平台上的消息处理系统，包括部署在所述云计算平台上的主机和备份机，还包括如第二方面所述的消息处理装置。
在本发明实施例提供的云计算平台上的消息处理方法中，在接收到发往主机和备份机的输入请求消息后，发送输入请求消息到所有的主机和备份机，在接收到主机和备份机返回的输入响应消息后，向外发送一个输入响应消息。与现有技术相比，本发明实施例将输入请求发送到所有的主机和备份机执行，主机和备份机执行同样的输入请求，因而主机和备份机之间不需要进行数据同步和状态同步，从而提高了双机备份的稳定性，进一步提高了双机切换速度。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。其中：
图1是本发明实施例中主机和备份机的部署示意图；
图2是本发明实施例一提供的消息处理方法的流程图；
图3是本发明实施例二提供的消息处理方法的流程图；
图4是本发明实施例三提供的消息处理方法的流程图；
图5是本发明实施例四提供的消息处理方法的流程图；
图6是本发明实施例五提供的消息处理装置的结构示意图；
图7是本发明实施例六提供的消息处理装置的结构示意图；
图8是本发明实施例七提供的消息处理系统的结构示意图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行 清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性的劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
本发明提供一种云计算平台上的消息处理方法、装置及消息处理系统，参见图1，图1是本发明实施例中主机和备份机的部署示意图。
如图所示，主机和备份机为部署在云计算平台上的虚拟机(也可以称为虚拟化容器)，主机和备份机上运行相同的应用，虚拟化容器1内运行应用程序(App1)作为主机，虚拟化容器2内运行应用程序(App1)的镜像作为备份机。其中，备份机的数量可以根据需要灵活增加，主机和备份机的概念是相对的，可以将所有的虚拟机称为主机，也可以将所有的虚拟机称为备份机。
虚拟化容器1和虚拟化容器2使用云计算IaaS层(简称I层)的网络资源、存储资源和计算(CPU)资源。其中，虚拟化容器1和虚拟化容器2可以共用相同的网络资源和存储资源，也可以使用独立的网络资源和存储资源。应用程序和IaaS层之间还包括消息处理装置(也可理解为接口适配层)。
参见图2，图2是本发明实施例一提供的云计算平台上的消息处理方法的流程图。
如图所示，一种云计算平台上的消息处理方法主要包括：
步骤101、接收输入请求消息，发送所述请求消息到部署在云计算平台上的主机和备份机，所述主机和备份机上运行相同的应用程序。
本发明实施例中消息处理装置在接收到外部(“外部”指的是主机和备份机以外)的输入请求消息，例如计费请求，复制发送该请求消息到主机和备份机，主机和备份机各自独立执行该输入请求消息，并返回对应的输入响应消息。
其中，接收到的输入请求消息可以是发往主机和/或备份机的，还可以是发往消息处理装置的，即输入请求消息目的地址可以为主机和/或备份机，还可以为本实施例中的消息处理装置，消息处理装置均可接收到该输入请求消息。主机和备份机可以具有相同的地址，还可以具有相互独立的地址。
步骤102、接收主机和备份机返回的多个输入响应消息，向外发送其中一个输入响应消息。
在本发明实施例中，消息处理装置在接收主机和备份机返回的多个输入响应消息后，可以识别出重复的输入响应消息，向外发送其中一个输入响应消息。
具体的，本发明实施例可以接收主机和备份机返回的多个输入响应消息，向外发送最先到达的响应消息，以提高消息发送效率。在向外发送最先到达的响应消息后，还可以根据输入响应消息的消息特征(例如会话标识)快速识别与最先返回的输入响应消息重复的消息并丢弃。消息处理装置还可以通过完全消息匹配的方式来识别主机和备份机发送的输入响应消息。
本发明实施例还可以接收主机和备份机返回的多个输入响应消息，所述输入响应消息中携带响应结果，本发明实施例确定重复次数最多的响应结果对应的第一类输入响应消息，向外发送所述第一类输入响应消息中的一个输入响应消息，进而提高响应消息发送的正确性。需要说明的是，响应结果可以包含在响应消息的消息体中，还可以直接为该响应消息本身。
本发明实施例还可以接收主机和备份机返回的多个输入响应消息，所述输入响应消息中携带响应结果，在判断所述多个输入响应消息的响应结果均不相同时，选择预先定义的主机或备份机的输入响应消息向外发送，即以预先规定的一个虚拟机的输入响应消息为准向外发送。
其中，本发明实施例中处理主机和备份机返回的多个输入响应消息的详细过程可以适用于本发明实施例一以下的其他实施例。
在本发明实施例中,为使得消息处理装置能接收到所有发往主机或备份机的输入请求消息，可以将消息处理装置配置做成必经的路由节点。消息处理装置可以拦截发往主机和备份机的输入请求消息以及拦截主机和备份机返回的多个输入响应消息。
在本发明实施例中，若接收到的输入请求消息中携带了全局唯一的消息特征(例如会话标识session-id)，消息处理装置可以根据消息特征来识别重复的输入响应消息，同时可以根据消息特征来匹配输入请求消息和输入响应消息。其中，消息特征的具体描述可以适用于本发明实施例一以下的其他实施例。
若消息处理装置接收到的输入请求消息中不携带全局唯一的消息标识，则消息处理装置可以在待发送的输入请求消息中增加一些关联信息，例如在 发送给每个虚拟机的输入请求消息中增加接收该消息的虚拟机标识(例如虚拟机的进程ID、虚拟机的MAC地址)和序列号(该序列号可以与一条输入请求消息对应)，然后发送修改的输入请求消息到主机和备份机。本发明实施例可以在I层的接口层增加关联信息，以便于消息处理装置进行识别。对应的，由于在发送的输入请求消息中增加了关联信息，因而返回的输入响应消息中同样携带该关联信息，消息处理装置可以根据该关联信息中的序列号识别重复的输入响应消息，根据该关联消息中的虚拟机标识识别不同的虚拟机返回的输入响应消息，还可以根据该关联信息匹配输入请求消息和输入响应消息。由于上述关联信息增加到I层的接口层，因而消息处理装置在向主机和备份机发送输入请求消息时，可以不携带该关联信息。
步骤103、在所述多个输入响应消息的响应结果出现异常时，重建所述主机和备份机。
在本发明实施例中，若多个输入响应消息的响应结果出现异常时，例如多个输入响应消息的响应结果均不相同或多个输入响应消息中只存在部分响应结果不同的响应消息，即另一部分输入响应消息的响应结果相同，则说明主机和备份机中的部分虚拟机可能存在故障，此时重建主机和备份机。其中，重建主机和备份机可以根据预先定义的规则(例如重新安装应用程序、恢复数据等)重建主机和备份机，以便于主机和备份机恢复正常。
此外，本发明实施例还可以在上述确定重复次数最多的响应结果对应的第一类输入响应消息时，根据第一类输入响应消息对应的虚拟机重建第一类输入响应消息对应的虚拟机以外的主机或备份机，以便于第一类输入响应消息对应的虚拟机以外的主机或备份机恢复正常，同时减少了虚拟机重建的数量，间接提高了备份的可靠性。
进一步的，在本发明实施例中的主机和备份机还可以分别对应健康度(或可信度)，健康度根据返回的输入响应消息正确的次数占返回的输入响应消息的总次数的比例确定。
本发明实施例在接收到主机或备份机返回的输入响应消息后，判断输入响应消息是否正确，并根据判断结果更新对应的主机和备份机的健康度(输入响应消息正确，则健康度增加)，进而在向外发送输入响应消息时，发送健康度高的虚拟机返回的输入响应消息。其中，判断输入响应消息是否正确 的过程可以根据少数服从多数的原则，例如统计每条输入响应消息中响应结果出现的次数，将出现次数最多的响应结果对应的输入响应消息判断为正确，而将其他的响应结果判断为错误。其中，本发明实施例中主机和备份机的健康度的具体描述可以适用于本发明实施例一以下的其他实施例。此外，本发明实施例还在所述主机或备份机的健康度低于阈值时，重建所述健康度低于阈值的主机或备份机。例如，根据预先定义的规则重建该虚拟机或根据健康度最高的虚拟机重建健康度低于阈值的主机或备份机。
参见图3，图3是本发明实施例二提供的云计算平台上的消息处理方法的流程图。
如图所示，一种云计算平台上的消息处理方法主要包括：
步骤201、接收主机和备份机发送的多个输出请求消息，向外发送其中一个输出请求消息。
具体的，运行应用程序的主机和备份机会向外发送输出请求消息，例如向终端发送位置请求消息，本发明实施例在接收到输出请求消息后，向外发送其中一个输出请求消息，例如发送最先收到的输出请求消息。
本发明实施例可以根据输出请求消息的消息特征来识别重复的输出请求消息，例如根据会话标识来识别。此外，主机和备份机可以在发送给消息处理装置的输出请求消息中增加该消息的序列号，消息处理装置根据该序列号来快速识别重复的输出请求消息。
在本发明实施例中,为使得消息处理装置能接收到所有主机或备份机发送的输出请求消息，可以将消息处理装置配置做成必经的路由节点，消息处理装置可以拦截所有的主机和备份机发送的输出请求消息。
步骤202、接收输出响应消息，向所述主机和备份机发送所述输出响应消息。
具体的，本发明实施例在接收到外部返回的输出响应消息，例如位置响应消息，则向主机和备份机发送接收到的位置响应消息。主机和备份机则根据接收到的输出响应消息进行数据处理。本发明实施例还可以在进行数据处理后将处理结果进行存储，即主机和备份机向I层的存储设备发送存储访问请求。
其中，消息处理装置可以根据本地保持的路由表向主机和备份机发送输出响应消息，或者向主机和备份机广播该输出响应消息。
需要说明的是，本发明实施例二中的输出请求消息可以为主机和备份机主动触发的输出请求消息，还可以为主机和备份机根据实施例一中接收到的输入请求消息返回的输出请求消息，此时主机和备份机返回的输入响应消息即为输出请求消息。实施例一中的消息处理方法在执行向外发送其中一个输入响应消息的步骤102后，还执行接收输出响应消息的步骤202。
步骤203、接收主机和备份机发送的多个存储访问请求，向所述主机和备份机共享的I层存储设备发送其中一个存储访问请求。
在本发明实施例中，主机和备份机共享相同的I层存储设备，则在接收到主机和备份机发送的多个存储访问请求(例如根据输出响应消息返回的存储访问请求)时，向共享的I层存储设备发送其中一个存储访问请求，例如发送最先到达的存储访问请求。此外，消息处理装置同样可以发送健康度最高的虚拟机的存储访问请求。
在本发明实施例中，若存储访问请求中携带消息特征，则消息处理装置根据该消息特征来识别重复的存储访问请求。此外，主机和备份机还可以在发出的存储访问请求中增加该访问请求的序列号，消息处理装置根据该序列号来快速识别重复的存储访问请求。上述序列号可以增加到I层的接口层，以便于消息处理装置进行识别。
在本发明实施例中，若主机和备份机不共享相同的I层存储设备，即主机和备份机配置的了各自独立的存储设备，此时主机和备份机各自向对应的存储设备发送存储访问请求和接收存储访问响应，消息处理装置不对存储访问请求进行处理。
步骤204、接收I层存储设备返回的存储访问响应，向主机和备份机发送所述存储访问响应。
本发明实施例在接收到I层存储设备返回的存储访问响应后，根据存储访问响应中的消息特征或序列号来匹配对应的存储访问请求，然后向主机和备份机复制发送该存储访问响应，以便于主机和备份机获取存储访问结果。
在本发明实施例提供的云计算平台上的消息处理方法中，在接收到主机和备份机向外发出的输出请求消息后，发送其中一个输出请求消息，在接收 到外部设备返回的输出响应消息后，向主机和备份机复制发送该输出响应消息。与现有技术相比，本发明实施例将输出响应消息发送到所有的主机和备份机，主机和备份机执行同样的输出响应消息，因而主机和备份机之间不需要进行数据同步和状态同步，从而提高了双机备份的稳定性，进一步提高了双机切换速度。
进一步的，本发明实施例还在接收到存储访问请求后发送其中一个到共享的I层存储设备，并将接收到的存储访问响应向所有的主机和备份机发送，从而实现共享的存储设备的访问。与现有技术相比，本发明实施例中的主机和备份机可以共享I层存储设备，降低了双机备份场景下的存储资源的冗余。
参见图4，图4是本发明实施例三提供的云计算平台上的消息处理方法的流程图。
如图所示，应用程序(例如计费应用)运行在虚拟化的容器内，虚拟化容器接管计费应用的所有资源访问请求，包括网络资源，磁盘资源、CPU资源。本实施例以计费应用中的融合记账部分(convergent billing part，CBP)为例来说明，计费应用所在的虚拟化容器共享存储资源，虚拟化容器的网络资源和存储资源独立部署，主机和备份机1、2上均部署了融合记账部分。
在主机和备份机部署完成后，消息处理装置本地存储了路由表，以便于后续的消息发送。
步骤301、接收外部计费请求消息。
其中，消息处理装置接收到的计费请求中携带消息特征，消息特征在本实施例中具体可以为会话标识(session-id)和计费请求号码(request number)。对于不同的外部设备先后发送的计费请求消息，则可以通过外部设备的标识(例如用户设备标识)来进行区分。外部计费请求消息的目的地址可以为主机和/或备份机，还可以为本实施例中的消息处理装置，消息处理装置均可接收到该外部计费请求消息。
步骤302、发送计费请求消息到主机和备份机。
其中，消息处理装置根据本地的路由表发送计费请求消息到主机和备份机1、2。
步骤303、主机和备份机处理计费请求消息，返回计费响应消息。
其中，主机和备份机1、2根据该计费请求独立执行计费处理，返回计费响应消息，消息处理装置执行步骤304。
此外，本发明实施例中主机和备份机1、2在处理计费请求消息后，还需要向I层存储设备写入数据，即执行步骤306。
步骤304、接收主机和备份机发送的计费响应消息，向外发送一个第一类计费响应消息。
其中，消息处理装置可以通过会话标识(session-id)、请求号码(request number)来识别不同的虚拟机针对同一计费请求消息返回的计费响应消息。
在本实施例中，主机或备份机在执行计费请求消息后，返回的计费响应消息中的计费结果可能不同，此时消息处理装置获取重复次数最多的计费结果对应的第一类计费响应消息，然后从第一类计费响应消息中选择一个向外发送。
步骤305、根据第一类计费响应消息对应的虚拟机重建第一类计费响应消息对应的虚拟机以外的主机或备份机。
消息处理装置根据第一类计费响应消息对应的虚拟机重建第一类计费响应消息对应的虚拟机以外的虚拟机(可以为主机或备份机)，以便于第一类计费响应消息对应的虚拟机以外的主机或备份机恢复正常，进而提高双机备份的可靠性。
特殊情况下，当所有的虚拟机的输出结果各不相同时，则根据预定义虚拟机创建规则，对所有的虚拟机进行重建。此外，本发明实施例还在虚拟机出现明显运行异常(例如无法返回响应消息)时，对该虚拟机进行重建，例如根据第一类计费响应消息对应的虚拟机重建该无法返回响应消息的虚拟机。
步骤306、主机和备份机向I层存储设备发送数据写入请求。
在本实施例中，主机和备份机1、2还需要向I层存储设备进行计费信息存储，向I层存储设备发送存储访问请求。在本实施例中，消息处理装置可以拦截所有的存储访问请求，存储访问请求具体为数据写入请求。在具体实现时，主机和备份机1、2发送的数据写入请求中的数据可能不一致。
步骤307、接收主机和备份机发送的多个数据写入请求，发送一个第一类数据写入请求到I层存储设备。
在本实施例中，数据写入请求中的消息消息特征为数据写入请求中的关键字(对应不同的待写入数据)。消息处理装置接收到数据写入请求后，通过数据写入请求中的关键字来识别主机和备份机1、2向I层存储设备发送的数据写入请求。
此外，数据写入请求中携带待写入数据，本发明实施例从具有相同关键字的多个数据写入请求中，确定重复次数最多的待写入数据对应的第一类数据写入请求，并向I层存储设备发送所述第一类数据写入请求中的一个数据写入请求，进而提高数据写入的正确性。
I层存储设备根据接收到的数据写入请求写入计费数据，并返回数据写入响应。
步骤308、接收I层存储设备返回的数据写入响应。
消息处理装置接收I层存储设备返回的数据写入响应。
步骤309、向主机和备份机复制发送该数据写入响应。
消息处理装置在接收到与数据写入请求对应的数据写入响应后，根据关键字来匹配对应的数据写入请求，然后根据内部存储的路由表向主机和备份机1、2发送接收到的数据写入响应，完成数据写入过程。
参见图5，图5是本发明实施例四提供的云计算平台上的消息处理方法的流程图。
如图所示，应用程序(例如呼叫处理应用)运行在虚拟化的容器内，虚拟化容器接管应用的所有资源访问请求，包括网络资源，磁盘资源、CPU资源。本实施例以呼叫处理应用中的服务控制部分(service control part，SCP)为例来说明，主机和备份机1、2上均部署了服务控制部分。
本发明实施例提供的双机备份场景下的智能网呼叫消息处理方法主要包括以下步骤：
步骤501、接收外部上报的呼叫控制请求。
具体的，消息处理装置接收外部的业务交换点(Service Switching Point，SSP)上报的呼叫控制请求。
步骤502、向主机和备份机发送呼叫控制请求。
其中，消息处理装置根据内部储存的路由表发送呼叫控制请求给所有的 主机和备份机。
主机和备份机各自独立执行呼叫控制，返回呼叫控制响应，呼叫控制响应中的(CAMEL Application Part，CAP)协议的事务处理能力应用部分(Transaction Capabilities Application Part，TCAP)层的会话标识(dialog id)可以唯一标识本次呼叫请求。
步骤503、接收主机和备份机发送的呼叫控制响应。
消息处理装置可以接收所有的主机和备份机1、2返回的呼叫控制响应。
步骤504、发送健康度最高的虚拟机返回的呼叫控制响应。
具体的，消息处理装置中存储有每个主机和备份机的健康度数据，健康度数据是根据历史的呼叫控制响应来计算得到的，例如根据返回的呼叫控制响应正确的次数占返回的呼叫控制响应的总次数的比例确定。
消息处理装置发送健康度最高的虚拟机返回的呼叫控制响应，根据呼叫控制响应中的CAP协议的TCAP层的会话标识来区分重复的呼叫控制响应并丢弃，从而完成呼叫控制过程。
步骤505、根据接收到的呼叫控制响应更新对应的主机和备份机的健康度。
本发明实施例在向外发送呼叫控制响应后，对所有主机和备份机本次返回的呼叫控制响应进行正确性判断，即统计所有的呼叫控制响应中不同的呼叫控制结果重复出现的次数，将重复出现的次数最大的响应结果对应的呼叫控制响应确定为正确的呼叫控制响应(若所有的呼叫控制响应中的呼叫控制结果均不相同，则可以将所有的呼叫控制响应均确定为错误的呼叫控制响应)，而将其他的响应结果对应的呼叫控制响应确定为错误的呼叫控制响应，进而更新呼叫控制响应对应的主机和备份机的健康度，以便于在发送呼叫控制响应时确定健康度最高的虚拟机。
此外，若主机或备份机未返回呼叫控制响应，则可以确定该虚拟机返回错误的呼叫控制响应，更新该虚拟机的健康度。
步骤506、重建健康度低于阈值的主机或备份机。
在本实施例中，若主机或备份机的健康度低于阈值(例如95％)，则确定该虚拟机出现异常，重建该虚拟机。例如，根据预先定义的规则重建该虚拟机或根据健康度最高的虚拟机重建健康度低于阈值的主机或备份机。
在本发明实施例中，通过部署多个主机和备份机，主机和备份机同时执行呼叫请求，双机备份来提高了呼叫控制过程的稳定性。此外，由于云环境下的单个应用程序配置可以很低，本发明实施例可以通过低成本虚拟机的规模化降低电信硬件的成本。
参见图6，图6是本发明实施例五提供的消息处理装置的结构示意图。
如图所示，本发明实施例提供的消息处理装置主要包括：
输入请求处理单元610，用于接收输入请求消息，发送所述请求消息到部署在所述云计算平台上的主机和备份机，所述主机和备份机上运行相同的应用程序，所述主机和备份机用于实现双机备份；
输入响应处理单元620，用于接收所述主机和备份机返回的多个输入响应消息，向外发送其中一个输入响应消息。
在本发明实施例提供的消息处理装置中，输入响应处理单元620接收主机和备份机返回的多个输入响应消息，向外发送其中一个输入响应消息具体可以包括：
输入响应处理单元620接收所述主机和备份机返回的多个输入响应消息，向外发送最先到达的响应消息；或者
输入响应处理单元620接收所述主机和备份机返回的多个输入响应消息，所述输入响应消息中携带响应结果，确定重复次数最多的响应结果对应的第一类输入响应消息，向外发送所述第一类输入响应消息中的一个输入响应消息；或者
输入响应处理单元620接收所述主机和备份机返回的多个输入响应消息，所述输入响应消息中携带响应结果，在所述多个输入响应消息的响应结果均不相同时，选择预先定义的主机或备份机的输入响应消息向外发送；或者
输入响应处理单元620接收所述主机和备份机返回的多个输入响应消息，发送主机和备份机中健康度最高的虚拟机返回的输入响应消息。
其中，输入响应处理单元620的具体处理过程参见上述的实施例一中的步骤102的相关描述，在此不再赘述。
进一步参见图6，本发明实施例提供的消息处理装置还包括：
虚拟机重建单元630，用于在所述多个输入响应消息的响应结果出现异常时，重建所述主机和备份机，或者在所述主机或备份机的健康度低于阈值时，重建所述健康度低于阈值的主机或备份机。其中，出现异常的情况包括：多个输入响应消息的响应结果均不相同或多个输入响应消息中只存在部分响应结果不同的响应消息，即另一部分输入响应消息的响应结果相同，虚拟机重建单元630重建虚拟机的具体过程参见上述实施例一中的步骤103的描述，在此不再赘述。
健康度更新单元640，用于根据接收到的输入响应消息更新所述主机和备份机的健康度，所述主机和备份机的健康度根据各自发送的携带正确的响应结果的输入响应消息的数量计算得到。其中，健康度更新单元640具体可以根据虚拟机发送的输入响应消息的总量和携带正确的响应结果的输入响应消息来计算得到。
输出请求处理单元650，用于接收主机和备份机发送的多个输出请求消息，向外发送其中一个输出请求消息；输出响应处理单元660，用于接收输出响应消息，向所述主机和备份机发送所述输出响应消息。其中，输出请求处理单元650和输出响应处理单元660的对输出请求消息和输出响应消息的具体处理过程参见上述实施例二中步骤201-202的相关描述，在此不再赘述。
存储请求处理单元670，用于接收主机和备份机发送的多个存储访问请求，向所述主机和备份机共享的I层存储设备发送其中一个存储访问请求；存储响应处理单元680，用于接收所述I层存储设备返回的存储访问响应，向所述主机和备份机发送所述存储访问响应。其中，存储响应处理单元680向主机和备份机共享的I层存储设备发送其中一个存储访问请求包括：存储响应处理单元680向所述主机和备份机共享的I层存储设备发送最先到达的存储访问请求。
在本发明实施例中，存储请求处理单元670和存储响应处理单元680对存储访问请求和存储访问响应的具体处理过程参见上述实施例二中步骤203-204的相关描述，在此不再赘述。与现有技术相比，本发明实施例中的消息处理装置可以使得主机和备份机可以共享I层存储设备，降低了双机备份场景下的存储资源的冗余。
参见图7，图7是本发明实施例六提供的消息处理装置的结构示意图。
如图所示，本发明实施例提供的消息处理装置主要包括：
处理器710、存储器720和总线730，所述存储器720用于存储计算机执行指令，所述处理器710与所述存储器720通过所述总线730连接，当所述装置运行时，所述处理器710执行所述计算机执行指令，以使所述装置执行如下方法：
接收输入请求消息，发送所述请求消息到部署在所述云计算平台上的主机和备份机，所述主机和备份机上运行相同的应用程序；
接收所述主机和备份机返回的多个输入响应消息，向外发送其中一个输入响应消息。
本发明实施例六提供的消息处理装置执行上述方法的具体过程和前述对应的消息处理方法实施例一、实施例二、实施例三和实施例四中的执行过程相同，具有相同的有益效果，详情参见上述实施例，在此不再赘述。
参见图8，图8是本发明实施例七提供的消息处理系统的结构示意图。
如图所示，本发明实施例提供的消息处理系统主要包括：主机810、备份机820，以及消息处理装置830。其中，消息处理装置830的具体结构和功能与上述实施例五、六提供的消息处理装置相同，详情参见上述实施例，在此不再赘述。本发明实施例提供的消息处理系统将输入请求消息发送到所有的主机和备份机执行，主机和备份机执行同样的输入请求消息，因而主机和备份机之间不需要进行数据同步和状态同步，从而提高了双机备份的稳定性，进一步提高了双机切换速度。
本领域普通技术人员将会理解，本发明的各个方面、或各个方面的可能实现方式可以被具体实施为系统、方法或者计算机程序产品。因此，本发明的各方面、或各个方面的可能实现方式可以采用完全硬件实施例、完全软件实施例(包括固件、驻留软件等等)，或者组合软件和硬件方面的实施例的形式，在这里都统称为“电路”、“模块”或者“系统”。此外，本发明的各方面、或各个方面的可能实现方式可以采用计算机程序产品的形式，计算机程序产品是指存储在计算机可读介质中的计算机可读程序代码。
计算机可读介质可以是计算机可读信号介质或者计算机可读存储介质。 计算机可读存储介质包含但不限于电子、磁性、光学、电磁、红外或半导体系统、设备或者装置，或者前述的任意适当组合，如随机存取存储器(RAM)、只读存储器(ROM)、可擦除可编程只读存储器(EPROM或者快闪存储器)、光纤、便携式只读存储器(CD-ROM)。
计算机中的处理器读取存储在计算机可读介质中的计算机可读程序代码，使得处理器能够执行在流程图中每个步骤、或各步骤的组合中规定的功能动作；生成实施在框图的每一块、或各块的组合中规定的功能动作的装置。
计算机可读程序代码可以完全在用户的计算机上执行、部分在用户的计算机上执行、作为单独的软件包、部分在用户的计算机上并且部分在远程计算机上，或者完全在远程计算机或者服务器上执行。也应该注意，在某些替代实施方案中，在流程图中各步骤、或框图中各块所注明的功能可能不按图中注明的顺序发生。例如，依赖于所涉及的功能，接连示出的两个步骤、或两个块实际上可能被大致同时执行，或者这些块有时候可能被以相反顺序执行。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本发明的范围。
以上所述，仅为本发明的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本发明揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本发明的保护范围之内。因此，本发明的保护范围应所述以权利要求的保护范围为准。

Claims (16)

  1. 一种云计算平台上的消息处理方法，其特征在于，包括：
    接收输入请求消息，发送所述请求消息到部署在所述云计算平台上的主机和备份机，所述主机和备份机上运行相同的应用程序；
    接收所述主机和备份机返回的多个输入响应消息，向外发送其中一个输入响应消息。
  2. 如权利要求1所述的方法，其特征在于，所述接收主机和备份机返回的多个输入响应消息，向外发送其中一个输入响应消息包括：
    接收所述主机和备份机返回的多个输入响应消息，向外发送最先到达的响应消息；或者
    接收所述主机和备份机返回的多个输入响应消息，所述输入响应消息中携带响应结果，确定重复次数最多的响应结果对应的第一类输入响应消息，向外发送所述第一类输入响应消息中的一个输入响应消息；或者
    接收所述主机和备份机返回的多个输入响应消息，所述输入响应消息中携带响应结果，在所述多个输入响应消息的响应结果均不相同时，选择预先定义的主机或备份机的输入响应消息向外发送；或者
    接收所述主机和备份机返回的多个输入响应消息，发送所述主机和备份机中健康度最高的虚拟机返回的输入响应消息。
  3. 如权利要求2所述的方法，其特征在于，还包括：
    根据接收到的输入响应消息更新所述主机和备份机的健康度，所述主机和备份机的健康度根据各自发送的携带正确的响应结果的输入响应消息的数量计算得到。
  4. 如权利要求2所述的方法，其特征在于，所述方法还包括：
    在所述多个输入响应消息的响应结果出现异常时，重建所述主机和备份机；或者
    在所述主机或备份机的健康度低于阈值时，重建所述健康度低于阈值的主机或备份机。
  5. 如权利要求1所述的方法，其特征在于，所述接收主机和备份机返回 的多个输入响应消息，向外发送其中一个输入响应消息具体包括：
    接收所述主机和备份机返回的多个输入响应消息，向外发送最先返回的输入响应消息；
    根据所述输入响应消息的消息特征识别与所述最先返回的输入响应消息重复的消息并丢弃。
  6. 如权利要求1-5任一项所述的方法，其特征在于，还包括：
    接收主机和备份机发送的多个输出请求消息，向外发送其中一个输出请求消息；
    接收输出响应消息，向所述主机和备份机发送所述输出响应消息。
  7. 如权利要求1-5任一项所述的方法，其特征在于，所述方法还包括：
    接收主机和备份机发送的多个存储访问请求，向所述主机和备份机共享的I层存储设备发送其中一个存储访问请求；
    接收所述I层存储设备返回的存储访问响应，向所述主机和备份机发送所述存储访问响应。
  8. 如权利要求7所述的方法，其特征在于，所述向主机和备份机共享的I层存储设备发送其中一个存储访问请求包括：
    向所述主机和备份机共享的I层存储设备发送最先到达的存储访问请求。
  9. 一种云计算平台上的消息处理装置，其特征在于，包括：
    输入消息处理单元，用于接收输入请求消息，发送所述请求消息到部署在所述云计算平台上的主机和备份机，所述主机和备份机用于实现双机备份，所述主机和备份机上运行相同的应用程序；
    输入响应处理单元，用于接收所述主机和备份机返回的多个输入响应消息，向外发送其中一个输入响应消息。
  10. 如权利要求9所述的消息处理装置，其特征在于，所述输入响应处理单元接收主机和备份机返回的多个输入响应消息，向外发送其中一个输入响应消息包括：
    输入响应处理单元接收所述主机和备份机返回的多个输入响应消息，向外发送最先到达的响应消息；或者
    输入响应处理单元接收所述主机和备份机返回的多个输入响应消息，所 述输入响应消息中携带响应结果，确定重复次数最多的响应结果对应的第一类输入响应消息，向外发送所述第一类输入响应消息中的一个输入响应消息；或者
    输入响应处理单元接收所述主机和备份机返回的多个输入响应消息，所述输入响应消息中携带响应结果，在所述多个输入响应消息的响应结果均不相同时，选择预先定义的主机或备份机的输入响应消息向外发送；或者
    输入响应处理单元接收所述主机和备份机返回的多个输入响应消息，发送主机和备份机中健康度最高的虚拟机返回的输入响应消息。
  11. 如权利要求10所述的消息处理装置，其特征在于，还包括：
    虚拟机重建单元，用于在所述多个输入响应消息的响应结果出现异常时，重建所述主机和备份机，或者在所述主机或备份机的健康度低于阈值时，重建所述健康度低于阈值的主机或备份机。
  12. 如权利要求10所述的消息处理装置，其特征在于，还包括：
    健康度更新单元，用于根据接收到的输入响应消息更新所述主机和备份机的健康度，所述主机和备份机的健康度根据各自发送的携带正确的响应结果的输入响应消息的数量计算得到。
  13. 如权利要求9-12任一项所述的消息处理装置，其特征在于，还包括：
    输出消息处理单元，用于接收主机和备份机发送的多个输出请求消息，向外发送其中一个输出请求消息；
    输出响应处理单元，用于接收输出响应消息，向所述主机和备份机发送所述输出响应消息。
  14. 如权利要求9-12任一项所述的消息处理装置，其特征在于，还包括：
    存储消息处理单元，用于接收主机和备份机发送的多个存储访问请求，向所述主机和备份机共享的I层存储设备发送其中一个存储访问请求；
    存储响应处理单元，用于接收所述I层存储设备返回的存储访问响应，向所述主机和备份机发送所述存储访问响应。
  15. 如权利要求14所述的消息处理装置，其特征在于，所述存储响应处理单元向主机和备份机共享的I层存储设备发送其中一个存储访问请求包括：
    存储响应处理单元向所述主机和备份机共享的I层存储设备发送最先到 达的存储访问请求。
  16. 一种云计算平台上的消息处理系统，包括部署在所述云计算平台上的主机和备份机，其特征在于，还包括如权利要求9-15任一项所述的消息处理装置。
PCT/CN2015/076985 2014-10-08 2015-04-20 计费方法和设备、接入设备、服务质量控制方法和设备 WO2016054906A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP15848459.2A EP3193482B1 (en) 2014-10-08 2015-04-20 Message processing method and apparatus
US15/481,962 US10567507B2 (en) 2014-10-08 2017-04-07 Message processing method and apparatus, and message processing system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410528631.0A CN105577711B (zh) 2014-10-08 2014-10-08 消息处理方法、装置及消息处理系统
CN201410528631.0 2014-10-08

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US15/481,962 Continuation US10567507B2 (en) 2014-10-08 2017-04-07 Message processing method and apparatus, and message processing system

Publications (1)

Publication Number Publication Date
WO2016054906A1 true WO2016054906A1 (zh) 2016-04-14

Family

ID=55652540

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/076985 WO2016054906A1 (zh) 2014-10-08 2015-04-20 计费方法和设备、接入设备、服务质量控制方法和设备

Country Status (4)

Country Link
US (1) US10567507B2 (zh)
EP (1) EP3193482B1 (zh)
CN (1) CN105577711B (zh)
WO (1) WO2016054906A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7028503B2 (ja) * 2018-03-29 2022-03-02 Ｎｅｃプラットフォームズ株式会社 コンピュータシステム、通信システム、コンピュータシステムによる制御方法及びプログラム

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103368993A (zh) * 2012-03-28 2013-10-23 上海地面通信息网络有限公司 一种虚拟云主机备份控制装置
US8615584B2 (en) * 2009-12-03 2013-12-24 International Business Machines Corporation Reserving services within a cloud computing environment
CN103532764A (zh) * 2013-10-24 2014-01-22 中国科学院信息工程研究所 多机热备份方法及系统
CN103761166A (zh) * 2014-01-22 2014-04-30 上海交通大学 一种虚拟化环境下针对网络服务的双机热备份容灾系统及其方法
CN103885857A (zh) * 2014-03-10 2014-06-25 中标软件有限公司 虚拟机容错方法以及设备

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6889338B2 (en) * 2001-08-15 2005-05-03 Nortel Networks Limited Electing a master server using election periodic timer in fault-tolerant distributed dynamic network systems
US7434096B2 (en) * 2006-08-11 2008-10-07 Chicago Mercantile Exchange Match server for a financial exchange having fault tolerant operation
US8364481B2 (en) * 2008-07-02 2013-01-29 Google Inc. Speech recognition with parallel recognition tasks
JP5822452B2 (ja) * 2010-10-22 2015-11-24 株式会社インテック ストレージサービス提供装置、システム、サービス提供方法、及びサービス提供プログラム
US8788871B2 (en) * 2011-06-28 2014-07-22 International Business Machines Corporation Unified, workload-optimized, adaptive RAS for hybrid systems
US20130030928A1 (en) * 2011-07-25 2013-01-31 Lopez Eric R Methods and systems for electronically marketing a product through affiliate sponsorships via a computer network
JP5957269B2 (ja) * 2012-04-09 2016-07-27 クラリオン株式会社 音声認識サーバ統合装置および音声認識サーバ統合方法
US9313252B2 (en) * 2012-04-20 2016-04-12 Microsoft Technology Licensing, Llc Binding crud-type protocols in distributed agreement protocols
CN103501290B (zh) * 2013-09-18 2017-10-24 万达信息股份有限公司 一种基于动态备份虚拟机的高可靠服务系统构建方法
US9532103B2 (en) * 2013-12-16 2016-12-27 Vmware, Inc. Multi-user support for set top boxes and game consoles
US9692853B2 (en) * 2014-04-22 2017-06-27 Pismo Labs Technology Limited Methods and systems for processing a DNS request

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8615584B2 (en) * 2009-12-03 2013-12-24 International Business Machines Corporation Reserving services within a cloud computing environment
CN103368993A (zh) * 2012-03-28 2013-10-23 上海地面通信息网络有限公司 一种虚拟云主机备份控制装置
CN103532764A (zh) * 2013-10-24 2014-01-22 中国科学院信息工程研究所 多机热备份方法及系统
CN103761166A (zh) * 2014-01-22 2014-04-30 上海交通大学 一种虚拟化环境下针对网络服务的双机热备份容灾系统及其方法
CN103885857A (zh) * 2014-03-10 2014-06-25 中标软件有限公司 虚拟机容错方法以及设备

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3193482A4 *

Also Published As

Publication number Publication date
EP3193482A1 (en) 2017-07-19
EP3193482A4 (en) 2017-11-08
US10567507B2 (en) 2020-02-18
US20170214742A1 (en) 2017-07-27
CN105577711B (zh) 2019-05-03
EP3193482B1 (en) 2020-02-05
CN105577711A (zh) 2016-05-11

Similar Documents

Publication Publication Date Title
US11079966B2 (en) Enhanced soft fence of devices
US10347542B2 (en) Client-initiated leader election in distributed client-server systems
JP6296632B2 (ja) マルチスレッドアプリケーションのためのセッションをリマッピングするための技術
US9423956B2 (en) Emulating a stretched storage device using a shared storage device
US10558568B2 (en) Restriction of validation to active tracks in non-volatile storage during a warmstart recovery
JP2017531250A (ja) 粒状同期／半同期アーキテクチャ
US11379329B2 (en) Validation of data written via two different bus interfaces to a dual server based storage controller
JP2014522052A (ja) ハードウェア故障の軽減
CN112395047A (zh) 虚拟机故障疏散方法、系统及计算机可读介质
US10606489B2 (en) Sidefiles for management of data written via a bus interface to a storage controller during consistent copying of data
US10691353B1 (en) Checking of data difference for writes performed via a bus interface to a dual-server storage controller
CN111381770B (zh) 一种数据存储切换方法、装置、设备及存储介质
WO2016054906A1 (zh) 计费方法和设备、接入设备、服务质量控制方法和设备
CN114296875A (zh) 基于容错系统的数据同步方法、系统及计算机可读介质
CN111104199A (zh) 虚拟机高可用的方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15848459

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

REEP Request for entry into the european phase

Ref document number: 2015848459

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2015848459

Country of ref document: EP