WO2016206513A1 - 加速处理数据的方法、分配装置和交换机 - Google Patents

加速处理数据的方法、分配装置和交换机 Download PDF

Info

Publication number
WO2016206513A1
WO2016206513A1 PCT/CN2016/083045 CN2016083045W WO2016206513A1 WO 2016206513 A1 WO2016206513 A1 WO 2016206513A1 CN 2016083045 W CN2016083045 W CN 2016083045W WO 2016206513 A1 WO2016206513 A1 WO 2016206513A1
Authority
WO
WIPO (PCT)
Prior art keywords
packet
message
acceleration
accelerated
type
Prior art date
Application number
PCT/CN2016/083045
Other languages
English (en)
French (fr)
Inventor
许珊珊
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2016206513A1 publication Critical patent/WO2016206513A1/zh

Links

Images

Definitions

  • the present application relates to, but is not limited to, the field of network technologies, and in particular, a method, a distribution device, and a switch for accelerating processing of data.
  • OAM Operaation Administration and Maintenance
  • CFM Connectivity Fault Management
  • BFD Bidirectional Forwarding Detection
  • the main purpose of the present application is to provide a method, a distribution device, and a switch for speeding up processing of data, thereby effectively improving data processing capabilities.
  • the present application proposes a method for speeding up processing of data, including:
  • the packet is added to the acceleration packet queue to process the acceleration packet in the acceleration packet queue by the acceleration processing device.
  • the packet is added to the non-accelerated packet queue.
  • the non-accelerated message in the non-accelerated message queue is processed by a conventional processing device.
  • the packet is added to the acceleration packet queue, so that the acceleration processing device processes the acceleration packet in the acceleration packet queue.
  • the steps include:
  • the type of the packet is an acceleration packet, determining a service type of the packet;
  • the step of determining the type of the received message includes:
  • the type of the packet is an acceleration packet
  • the type of the packet is a non-accelerated packet.
  • the application further provides a computer readable storage medium storing computer executable instructions that are implemented when the computer executable instructions are executed.
  • the application also provides a device for accelerating the processing of data, comprising:
  • a type judging module configured to determine a type of the message received through the network interface
  • the acceleration message adding module is configured to add the message to the acceleration message queue when the type of the message is an acceleration message, so that the acceleration processing device processes the acceleration report in the acceleration message queue.
  • the non-accelerated message adding module is configured to: when the type of the message is a non-accelerated message, send the message to the non-accelerated message queue to process the non-accelerated message by the conventional processing device. Non-accelerated messages in the queue.
  • the acceleration packet adding module is configured to determine a service type of the packet when the type of the packet is an acceleration packet, and set the packet according to a service type of the packet. Add to And in the acceleration packet queue corresponding to the service type, the acceleration processing device sequentially accelerates and processes the acceleration packets in each of the acceleration message queues.
  • the type determining module is configured to determine, according to a protocol type of the packet, a type of the packet: when the protocol type of the packet is an operation management and maintenance OAM protocol, where the packet is The type is an acceleration packet.
  • the protocol type of the packet is not the OAM protocol, the type of the packet is a non-accelerated packet.
  • the present application also provides a switch for speeding up processing of data, including a distribution device, an acceleration processing device, and a conventional processing device;
  • the dispensing device comprises:
  • a type judging module configured to determine a type of the received message
  • the acceleration message adding module is configured to add the message to the acceleration message queue when the type of the message is an acceleration message, to process, by the acceleration processing device, the cached message queue Accelerate the message;
  • the non-accelerated message adding module is configured to add the message to the non-accelerated message queue when the type of the message is a non-accelerated message, to process the non-accelerated message by the conventional processing device.
  • the acceleration processing device is configured to process an accelerated message in the accelerated message queue
  • the conventional processing device is configured to process non-accelerated messages in the non-accelerated message queue.
  • the acceleration processing device is configured to batch process the accelerated packets in the plurality of the accelerated message queues corresponding to the respective accelerated services according to the accelerated message queue in which the acceleration message is located.
  • the conventional processing device is configured to limit the rate at which the non-accelerated message in the non-accelerated message queue is sent and received.
  • the switch in the present application includes a plurality of processing devices, and the acceleration message and the non-acceleration message are allocated to different processing devices for processing, and the acceleration processing device for processing the accelerated message does not need to determine the need.
  • the priority of the service corresponding to the packet queue in which the processed packet is located can process the accelerated packet in batches, which is beneficial to improving the efficiency of processing the accelerated message, and avoids processing the non-accelerated message by the accelerated processing device.
  • the efficiency of processing the accelerated message is reduced.
  • the conventional processing device has the ability to process the non-accelerated message and the control program without processing the accelerated message requiring more computing resources, thereby reducing the load of the conventional processing device and improving the load. The efficiency of the switch processing messages.
  • FIG. 1 is a flowchart of a first embodiment of a method for speeding up processing of data provided by the present application
  • step S30 of the second embodiment of the method for speeding up processing of data provided by the present application is a flowchart of step S30 of the second embodiment of the method for speeding up processing of data provided by the present application
  • step S20 of the third embodiment of the method for speeding up processing of data provided by the present application is a flowchart of step S20 of the third embodiment of the method for speeding up processing of data provided by the present application
  • FIG. 4 is a schematic diagram of an apparatus for distributing accelerated processing data provided by the present application.
  • FIG. 5 is a schematic structural diagram of a switch for speeding up processing data provided by the present application.
  • FIG. 1 is a flowchart of a first embodiment of a method for speeding up processing of data provided by the present application.
  • the method for speeding up processing data mentioned in this embodiment includes:
  • Step S20 determining the type of the received message
  • the switch in this embodiment includes a plurality of processing devices, and the plurality of processing devices can be divided into two groups, one set is configured to process control messages and other messages requiring less computing resources, and the other group is set to be Accelerated messages requiring more computing resources.
  • This embodiment takes two processing devices as an example.
  • the switch receives the packet through the network interface, and the packet is classified by the distribution device in the switch.
  • the distribution device distinguishes the type of the packet by using an ACL (Access Control List) or a unicast route.
  • ACL Access Control List
  • the packets that require more computing resources are accelerated packets and are added to the accelerated packet queue.
  • the packets of the resource are ordinary packets, that is, non-accelerated packets, which are added to the non-accelerated message queue.
  • Step S30 When the type of the packet is an acceleration packet, the packet is added to the acceleration packet queue, so that the acceleration processing device processes the acceleration packet in the accelerated message queue.
  • the distribution device sequentially adds the acceleration message to the acceleration message queue.
  • the acceleration processing device can process the acceleration packets in the multiple acceleration packet queues corresponding to the multiple acceleration services, and the different types of acceleration packets can be added to different acceleration packet queues, and can be grouped according to the service type of the packet. Or, in the order of the message, put them into each accelerated message queue. After the packets are grouped, the accelerated packets in each accelerated packet queue are accelerated. Because the packets in the accelerated packet queue require more computing resources to speed up the processing, the packets in the same accelerated packet queue can be processed in batches at the same time, which is beneficial to improve the efficiency of processing the packets.
  • Step S40 When the type of the packet is a non-accelerated packet, the packet is added to the non-accelerated packet queue to process the non-accelerated packet in the non-accelerated packet queue by the conventional processing device.
  • the non-accelerated message requires less computing resources and does not require a shorter response time, it does not need to be accelerated, and can be processed by a conventional processing device, thereby avoiding an increase in the load of the acceleration processing device set to handle the accelerated service, while avoiding the non-accelerated report.
  • the longer response time required by the text reduces the efficiency of the acceleration processing device in processing the accelerated message, thereby avoiding the impact on the efficiency of processing the accelerated message.
  • the switch in this embodiment includes an acceleration processing device and a conventional processing device, and the accelerated message and the non-accelerated message are respectively processed by the corresponding acceleration processing device or the conventional processing device, which is beneficial to improving the efficiency of processing the accelerated message and avoiding The efficiency of processing the accelerated message is reduced because the acceleration processing device processes the non-accelerated message.
  • the conventional processing device has the ability to process the non-accelerated message and the control program without processing the acceleration report requiring more computing resources. Therefore, the load of the conventional processing device is reduced, and the efficiency of processing the message by the switch is improved.
  • FIG. 2 is a flowchart of step S30 of the second embodiment of the method for speeding up processing of data provided by the present application.
  • This embodiment includes the steps in the embodiment shown in FIG. 1, wherein step S30 includes:
  • Step S31 When the type of the packet is an acceleration packet, determine the service type of the packet.
  • Step S32 Add the packet to the acceleration packet queue corresponding to the service type according to the service type of the packet, so that the acceleration processing device separately accelerates and processes the acceleration packet in each of the accelerated packet queues.
  • the acceleration processing device of the embodiment processes the acceleration packets in the plurality of accelerated message queues, and the different service types correspond to different acceleration message queues.
  • the distribution device will be different.
  • the service type packet is added to the corresponding acceleration packet queue.
  • FIG. 3 is a flowchart of step S20 of the third embodiment of the method for speeding up processing of data provided by the present application.
  • This embodiment includes the steps in the embodiment shown in FIG. 1, wherein step S20 includes:
  • Step S21 Determine the type of the packet according to the protocol type of the packet:
  • Step S22 When the protocol type of the packet is the OAM protocol, the type of the packet is an acceleration packet.
  • step S23 when the protocol type of the packet is not the OAM protocol, the type of the packet is a non-accelerated packet.
  • the packet when determining the packet type, the packet is classified according to the protocol type of the packet.
  • the OAM protocol such as CFM and BFD, has high requirements on time performance. Therefore, packets with the protocol type of OAM can be accelerated as acceleration packets.
  • the priority of the processing packet can be defined in advance, and the packet with the priority higher than the threshold is used as the acceleration packet to implement packet classification.
  • determining whether the packet is an accelerated packet according to the type of the packet protocol is simple, and the method for determining the efficiency of the packet is improved, thereby improving the processing efficiency of the switch.
  • Embodiments of the present invention further provide a computer readable storage medium storing computer executable instructions that are implemented when the computer executable instructions are executed.
  • FIG. 4 is a schematic diagram of an apparatus for distributing accelerated processing data provided by the present application.
  • the device for accelerating processing data mentioned in this embodiment includes:
  • the type determining module 120 is configured to determine the type of the received message
  • the acceleration message adding module 130 is configured to add the message to the acceleration message queue when the type of the message is an acceleration message, so that the acceleration processing device processes the acceleration message in the accelerated message queue;
  • the non-acceleration packet adding module 110 is configured to add a packet to the non-accelerated packet queue when the type of the packet is a non-accelerated packet, so that the non-acceleration in the non-accelerated packet queue is processed by the conventional processing device. Message.
  • the switch in this embodiment includes a plurality of processing devices, and the plurality of processing devices can be divided into two groups, one set to process control messages and other messages requiring less computing resources, and the other set to process requirements. Accelerated messages with more computing resources.
  • This embodiment takes two processing devices as an example.
  • the switch receives the packet through the network interface, and the packet is classified by the distribution device in the switch.
  • the device allocates packets of different computing resources to the accelerating packet queue by using an ACL or a unicast route.
  • the packets that require less computing resources are ordinary packets. That is, non-accelerated messages are added to the non-accelerated message queue.
  • the distribution device sequentially adds the acceleration message to the acceleration message queue.
  • the acceleration processing device can process the acceleration packets in the multiple acceleration packet queues corresponding to the multiple acceleration services, and the different types of acceleration packets can be added to different acceleration packet queues, and can be grouped according to the service type of the packet. Or, in the order of the message, put them into each accelerated message queue. After the packets are grouped, the accelerated packets in each accelerated packet queue are accelerated. Due to acceleration in the message queue Packets require more computing resources to speed up processing. Therefore, when processing, batches of packets in the same accelerated packet queue can be processed in batches, which is beneficial to improve the efficiency of processing packets.
  • the non-accelerated message requires less computing resources and does not require a shorter response time, it does not need to be accelerated, and can be processed by a conventional processing device, thereby avoiding an increase in the load of the acceleration processing device set to handle the accelerated service, while avoiding the non-accelerated report.
  • the longer response time required by the text reduces the efficiency of the acceleration processing device in processing the accelerated message, thereby avoiding the impact on the efficiency of processing the accelerated message.
  • the switch in this embodiment includes an acceleration processing device and a conventional processing device, and the accelerated message and the non-accelerated message are respectively processed by the corresponding acceleration processing device or the conventional processing device, which is beneficial to improving the efficiency of processing the accelerated message and avoiding The efficiency of processing the accelerated message is reduced because the acceleration processing device processes the non-accelerated message.
  • the conventional processing device has the ability to process the non-accelerated message and the control program without processing the accelerated message requiring more computing resources. Thereby reducing the load of the conventional processing device and improving the efficiency of the switch processing the message.
  • the acceleration packet adding module 130 is configured to determine a service type of the packet when the type of the packet is an acceleration packet.
  • the packet is added to the acceleration packet queue corresponding to the service type, so that the acceleration processing device accelerates the acceleration of the acceleration packet in each of the accelerated packet queues.
  • the acceleration processing device of the embodiment processes the acceleration packets in the plurality of accelerated message queues, and the different service types correspond to different acceleration message queues.
  • the distribution device will be different.
  • the service type packet is added to the corresponding acceleration packet queue.
  • the allocation device in this embodiment adds the packet to the non-additional packet.
  • the fast message queue is sent to the conventional processing device for processing. Since the conventional processing device also needs to process the control message, in order to avoid the burden of the conventional processing device, the conventional processing device needs to limit the rate of sending and receiving non-accelerated messages, and avoids a large number of non-accelerated messages being sent to the conventional processing device, causing the conventional processing device to be overloaded. Big. At the same time, since the accelerated message and the non-accelerated message are classified into their respective queues, the longer response time required to process the control message and the non-accelerated message does not affect the efficiency of processing the accelerated message, which is beneficial to satisfy the acceleration. The processing efficiency of packets improves the efficiency of packets processed by the switch.
  • the type judging module 120 is configured to determine the type of the packet according to the protocol type of the packet: when the protocol type of the packet is OAM, the type of the packet is an accelerated packet; and the protocol type of the packet. When the OAM protocol is not used, the type of the packet is non-accelerated.
  • the packet when determining the packet type, the packet is classified according to the protocol type of the packet.
  • the OAM protocol such as CFM and BFD, has high requirements on time performance. Therefore, packets with the protocol type of OAM can be accelerated as acceleration packets.
  • the priority of the processing packet can be defined in advance, and the packet with the priority higher than the threshold is used as the acceleration packet to implement packet classification.
  • determining whether the packet is an accelerated packet according to the type of the packet protocol is simple, and the method for determining the efficiency of the packet is improved, thereby improving the processing efficiency of the switch.
  • FIG. 5 is a schematic structural diagram of a switch for speeding up processing data provided by the present application.
  • the switch for speeding up processing data proposed in this embodiment includes a distribution device 100, an acceleration processing device 400, and a conventional processing device 500;
  • the dispensing device 100 includes:
  • the type determining module 120 is configured to determine the type of the received message
  • the acceleration message adding module 130 is configured to add the message to the acceleration message queue when the type of the message is an acceleration message, so that the acceleration processing device 400 processes the acceleration message in the accelerated message queue;
  • the non-acceleration packet adding module 110 is configured to report when the type of the packet is a non-accelerated packet.
  • the text is added to the non-accelerated message queue to process the non-accelerated message in the non-accelerated message queue by the conventional processing device 500;
  • the acceleration processing device 400 is configured to process the accelerated message in the accelerated message queue
  • the conventional processing device 500 is configured to process non-accelerated messages in the non-accelerated message queue.
  • the acceleration processing device 400 is configured to batch process the accelerated packets in the plurality of accelerated message queues corresponding to the respective accelerated services according to the accelerated message queue in which the accelerated message is located.
  • the conventional processing device 500 is configured to limit the rate at which the non-accelerated messages in the non-accelerated message queue are sent and received.
  • the switch in this embodiment includes the acceleration processing device 400 and the conventional processing device 500.
  • the acceleration processing device 400 and the conventional processing device 500 can be as shown in FIG. 5, and the structure and principle thereof can be referred to the above embodiments, and details are not described herein. Since the switch in the embodiment includes the acceleration processing device 400 and the normal processing device 500, the acceleration message and the non-acceleration message are allocated to the corresponding acceleration processing device 400 and the conventional processing device 500 for processing, which is advantageous for improving the processing acceleration report.
  • the accelerated message of the resource is calculated, thereby reducing the load of the conventional processing device 500 and improving the efficiency of the switch processing the message.
  • the switch in the present application includes a plurality of processing devices, and the acceleration message and the non-acceleration message are respectively processed to different processing devices, and the acceleration processing device for processing the acceleration message does not need to determine the message in which the message to be processed is located.
  • the priority of the service corresponding to the queue can process the accelerated message in batches, which is beneficial to improving the efficiency of processing the accelerated message, and avoids reducing the efficiency of processing the accelerated message by the accelerated processing device processing the non-accelerated message.
  • the conventional processing device has the ability to process the non-accelerated message and the control program without processing the accelerated message requiring more computing resources, thereby reducing the load of the conventional processing device and improving the efficiency of the switch processing the message.

Landscapes

  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

一种加速处理数据的方法，包括：判断接收的报文的类型；当所述报文的类型为加速报文时，将所述报文添加到加速报文队列中，以由加速处理装置处理所述加速报文；当所述报文的类型为非加速报文时，将所述报文添加到非加速报文队列中，以由常规处理装置处理非加速报文。还公开了一种加速处理数据的CPU核和交换机。本发明中的交换机包括多个处理装置，将加速报文与非加速报文分配到不同的处理装置进行处理，有利于提高处理加速报文的效率，且避免了因加速处理装置处理非加速报文而降低处理加速报文的效率，减少了常规处理装置的负荷，提高了交换机处理报文的效率。

Description

加速处理数据的方法、分配装置和交换机 技术领域
本申请涉及但不限于网络技术领域，特别是加速处理数据的方法、分配装置和交换机。
背景技术
随着互联网络技术的快速发展以及网络基础设施的不断完善，网络规模不断扩大，人们对网络的质量要求也越来越高。同时，为了保障网络的正常运行，还需要快速检测并解决网络连接中的问题。因此，为确保网络质量，CFM(Connectivity Fault Management，连接性故障管理)、BFD(Bidirectional Forwarding Detection，双向转发检测)等OAM(Operation Administration and Maintenance，操作管理维护)协议应运而生，旨在提高网络的可靠性和可维护性。然而，由于这些协议对时间性能的要求很高，而相关技术的交换机中一个或多个处理装置上随机处理收发的报文和控制程序，容易让处理装置超负荷，无法保证程序的有效运行，数据处理能力也不能满足快速收发报文的性能需求。
发明内容
以下是对本文详细描述的主题的概述。本概述并非是为了限制权利要求的保护范围。
本申请的主要目的为提供一种加速处理数据的方法、分配装置和交换机，有效提高数据处理能力。
本申请提出一种加速处理数据的方法，包括：
判断接收的报文的类型；
当所述报文的类型为加速报文时，将所述报文添加到加速报文队列中，以由加速处理装置处理所述加速报文队列中的所述加速报文；
当所述报文的类型为非加速报文时，将所述报文添加到非加速报文队列 中，以由常规处理装置处理所述非加速报文队列中的所述非加速报文。
可选地，所述当报文的类型为加速报文时，将所述报文添加到加速报文队列中，以由加速处理装置处理所述加速报文队列中的所述加速报文的步骤包括：
当所述报文的类型为加速报文时，判断所述报文的业务类型；
按照所述报文的业务类型，将所述报文添加到所述业务类型对应的加速报文队列中，以由所述加速处理装置分别批量加速处理各所述加速报文队列中的加速报文。
可选地，所述判断接收的报文的类型的步骤包括：
根据所述报文的协议类型，判断所述报文的类型：
当所述报文的协议类型为操作管理维护OAM协议时，所述报文的类型为加速报文；
当所述报文的协议类型不是OAM协议时，所述报文的类型为非加速报文。
本申请另外提供一种计算机可读存储介质，存储有计算机可执行指令，所述计算机可执行指令被执行时实现上述方法。
本申请还提出一种加速处理数据的分配装置，包括：
类型判断模块，设置成判断通过网络接口接收的报文的类型；
加速报文添加模块，设置成当所述报文的类型为加速报文时，将所述报文添加到加速报文队列中，以由加速处理装置处理所述加速报文队列中的加速报文；
非加速报文添加模块，设置成当所述报文的类型为非加速报文时，将所述报文发送给添加到非加速报文队列，以由常规处理装置处理所述非加速报文队列中的非加速报文。
可选地，所述加速报文添加模块是设置成当所述报文的类型为加速报文时，判断所述报文的业务类型；按照所述报文的业务类型，将所述报文添加 到所述业务类型对应的加速报文队列中，以由所述加速处理装置分别批量加速处理各所述加速报文队列中的加速报文。
可选地，所述类型判断模块是设置成根据所述报文的协议类型，判断所述报文的类型：当所述报文的协议类型为操作管理维护OAM协议时，所述报文的类型为加速报文；当所述报文的协议类型不是OAM协议时，所述报文的类型为非加速报文。
本申请还提出一种加速处理数据的交换机，包括分配装置、加速处理装置和常规处理装置；其中，
所述分配装置包括：
类型判断模块，设置成判断接收的报文的类型；
加速报文添加模块，设置成当所述报文的类型为加速报文时，将所述报文添加到加速报文队列中，以由加速处理装置处理所述加速报文队列中的所述加速报文；
非加速报文添加模块，设置成当所述报文的类型为非加速报文时，将所述报文添加到非加速报文队列，以由所述常规处理装置处理所述非加速报文队列中的非加速报文；
所述加速处理装置，设置成处理所述加速报文队列中的加速报文；
所述常规处理装置，设置成处理所述非加速报文队列中的非加速报文。
可选地，所述加速处理装置是设置成根据所述加速报文所在的所述加速报文队列，批量处理分别对应于各个加速业务的多个所述加速报文队列中的加速报文。
可选地，所述常规处理装置是设置成限制收发所述非加速报文队列中的非加速报文的速率。
本申请中的交换机包括多个处理装置，将加速报文与非加速报文分配给不同的处理装置进行处理，用于处理加速报文的加速处理装置无需判断需要 处理的报文所在的报文队列对应的业务的优先级，就能批量地处理加速报文，有利于提高处理加速报文的效率，且避免了因所述加速处理装置处理非加速报文而降低处理加速报文的效率，同时，所述常规处理装置有能力处理非加速报文以及控制程序，而不需要处理要求较多计算资源的加速报文，从而减少了常规处理装置的负荷，提高了交换机处理报文的效率。
在阅读并理解了附图和详细描述后，可以明白其他方面。
附图概述
图1为本申请提供的加速处理数据的方法的第一实施例的流程图；
图2为本申请提供的加速处理数据的方法的第二实施例的步骤S30的流程图；
图3为本申请提供的加速处理数据的方法的第三实施例的步骤S20的流程图；
图4为本申请提供的加速处理数据的分配装置的示意图；
图5为本申请提供的加速处理数据的交换机的结构示意图。
本申请目的的实现、功能特点及优点将结合实施例，参照附图做进一步说明。
本发明的较佳实施方式
应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
如图1所示，图1为本申请提供的加速处理数据的方法的第一实施例的流程图。本实施例提到的加速处理数据的方法，包括：
步骤S20，判断接收的报文的类型；
在本实施例中的交换机包括多个处理装置，可将多个处理装置分为两组，一组设置成处理控制报文和其他要求较少计算资源的报文，另一组设置成处 理要求较多计算资源的加速报文。本实施例以两个处理装置为例。交换机通过网络接口接收报文，并由交换机中的分配装置对报文进行分类。
分配装置通过ACL(Access Control Lists，访问控制列表)或单播路由对报文的类型进行区分，要求较多计算资源的报文为加速报文，添加到加速报文队列中；要求较少计算资源的报文为普通报文，即非加速报文，添加到非加速报文列队中。
步骤S30，当报文的类型为加速报文时，将报文添加到加速报文队列中，以由加速处理装置处理加速报文队列中的所述加速报文；
对于加速报文，分配装置依次将加速报文添加到加速报文队列中。加速处理装置可处理多个加速业务分别对应的多个加速报文队列中的加速报文，不同类型的加速报文可添加到不同的加速报文队列中，可根据报文的业务类型进行分组，或按照报文的顺序依次放入各加速报文队列中。将报文分组后，分别对各加速报文队列中的加速报文进行加速处理。由于加速报文队列中的报文都要求较多计算资源以加速处理，因此在处理时，可同时批量处理同一加速报文队列中的各个报文，有利于提高处理报文的效率。
步骤S40，当报文的类型为非加速报文时，将报文添加到非加速报文队列中，以由常规处理装置处理所述非加速报文队列中的所述非加速报文。
由于非加速报文要求较少的计算资源，无须较短的响应时间，因此无须加速处理，可由常规处理装置处理，避免增加设置成处理加速业务的加速处理装置的负荷，同时避免因非加速报文所要求的较长的响应时间而降低加速处理装置处理加速报文的效率，进而避免影响对加速报文进行处理的效率。
本实施例中的交换机包括加速处理装置和常规处理装置，加速报文与非加速报文分别由对应的加速处理装置或常规处理装置进行处理，有利于提高处理加速报文的效率，且避免了因所述加速处理装置处理非加速报文而降低处理加速报文的效率，同时，所述常规处理装置有能力处理非加速报文以及控制程序，而不需要处理要求较多计算资源的加速报文，从而减少了常规处理装置的负荷，提高了交换机处理报文的效率。
如图2所示，图2为本申请提供的加速处理数据的方法的第二实施例的步骤S30的流程图。本实施例包括图1所示实施例中的步骤，其中的步骤S30包括：
步骤S31，当报文的类型为加速报文时，判断报文的业务类型；
步骤S32，按照报文的业务类型，将报文添加到业务类型对应的加速报文队列中，以由所述加速处理装置分别批量加速处理各所述加速报文队列中的加速报文。
本实施例的加速处理装置处理多个加速报文队列中的加速报文，不同的业务类型对应不同的加速报文队列，分配装置在为报文分组时，根据报文的业务类型，将不同业务类型的报文添加到对应的加速报文队列中。在处理各加速报文时，为节省处理时间，无须区分加速报文队列中各加速报文的优先级，将加速报文队列中的多个加速报文批量处理，有利于提高处理报文的效率。同时，由于同时批量处理多个加速报文，只需要调用一次***资源提供的API(Application Programming Interface，应用程序编程接口)，无须为每一个报文都调用一次API，有利于节省***资源，进一步提高处理报文的效率。
如图3所示，图3为本申请提供的加速处理数据的方法的第三实施例的步骤S20的流程图。本实施例包括图1所示实施例中的步骤，其中的步骤S20包括：
步骤S21，根据报文的协议类型，判断报文的类型：
步骤S22，当报文的协议类型为OAM协议时，报文的类型为加速报文；
步骤S23，当报文的协议类型不是OAM协议时，报文的类型为非加速报文。
本实施例在判断报文类型时，根据报文的协议类型对报文进行分类。由于CFM、BFD等OAM协议对时间性能的要求很高，因此可将协议类型为OAM协议的报文作为加速报文进行加速处理。此外，还可以预先在报文中设置报文类型，预先定义报文是否需要加速处理，根据预先定义的类型进行 分类。或者，还可以预先定义处理报文的优先级别，将优先级别高于阈值的报文作为加速报文，实现报文分类。本实施例中，根据报文协议类型判断报文是否为加速报文，判断方式简单，有利于提高对报文进行分类的效率，进而提高交换机的处理效率。
本发明实施例另外提供一种计算机可读存储介质，存储有计算机可执行指令，所述计算机可执行指令被执行时实现上述方法。
图4为本申请提供的加速处理数据的分配装置的示意图。如图4所示，本实施例中提到的加速处理数据的分配装置，包括：
类型判断模块120，设置成判断接收的报文的类型；
加速报文添加模块130，设置成当报文的类型为加速报文时，将报文添加到加速报文队列中，以由加速处理装置处理加速报文队列中的所述加速报文；
非加速报文添加模块110，设置成当报文的类型为非加速报文时，将报文添加到非加速报文队列，以由常规处理装置处理所述非加速报文队列中的非加速报文。
在本实施例中的交换机包括多个处理装置，可将多个处理装置分为两组，一组设置成处理控制报文和其他要求较少计算资源的报文，另一组设置成处理要求较多计算资源的加速报文。本实施例以两个处理装置为例。交换机通过网络接口接收报文，并由交换机中的分配装置对报文进行分类。
分配装置通过ACL或单播路由对报文的类型进行区分，要求较多计算资源的报文为加速报文，添加到加速报文队列中；要求较少计算资源的报文为普通报文，即非加速报文，添加到非加速报文列队中。
对于加速报文，分配装置依次将加速报文添加到加速报文队列中。加速处理装置可处理多个加速业务分别对应的多个加速报文队列中的加速报文，不同类型的加速报文可添加到不同的加速报文队列中，可根据报文的业务类型进行分组，或按照报文的顺序依次放入各加速报文队列中。将报文分组后，分别对各加速报文队列中的加速报文进行加速处理。由于加速报文队列中的 报文都是要求较多计算资源以加速处理，因此在处理时，可同时批量处理同一加速报文队列中的各个报文，有利于提高处理报文的效率。
由于非加速报文要求较少的计算资源，无须较短的响应时间，因此无须加速处理，可由常规处理装置处理，避免增加设置成处理加速业务的加速处理装置的负荷，同时避免因非加速报文所要求的较长的响应时间而降低加速处理装置处理加速报文的效率，进而避免影响对加速报文进行处理的效率。
本实施例中的交换机包括加速处理装置和常规处理装置，加速报文与非加速报文分别由对应的加速处理装置或常规处理装置进行处理，有利于提高处理加速报文的效率，且避免了因加速处理装置处理非加速报文而降低处理加速报文的效率，同时，所述常规处理装置有能力处理非加速报文以及控制程序，而不需要处理要求较多计算资源的加速报文，从而减少了常规处理装置的负荷，提高了交换机处理报文的效率。
可选地，加速报文添加模块130是设置成当报文的类型为加速报文时，判断报文的业务类型；
按照报文的业务类型，将报文添加到业务类型对应的加速报文队列中，以由所述加速处理装置分别批量加速处理各所述加速报文队列中的加速报文。
本实施例的加速处理装置处理多个加速报文队列中的加速报文，不同的业务类型对应不同的加速报文队列，分配装置在为报文分组时，根据报文的业务类型，将不同业务类型的报文添加到对应的加速报文队列中。在处理各加速报文时，为节省处理时间，无须区分加速报文队列中各加速报文的优先级，将加速报文队列中的多个加速报文批量处理，有利于提高处理报文的效率。同时，由于同时批量处理多个加速报文，只需要调用一次***资源提供的API(Application Programming Interface，应用程序编程接口)，无须为每一个报文都调用一次API，有利于节省***资源，进一步提高处理报文的效率。
本实施例的分配装置在当前报文为非加速报文时，将该报文添加到非加 速报文队列中，并发送给常规处理装置进行处理。由于常规处理装置还需要处理控制报文，为了避免常规处理装置负担过大，常规处理装置需要限制收发非加速报文的速率，避免大量非加速报文发送给常规处理装置造成常规处理装置负荷过大。同时，由于将加速报文和非加速报文分类到各自的队列，处理控制报文和非加速报文所需的较长的响应时间不会影响处理加速报文的效率，有利于满足对加速报文的处理效率，提高交换机处理报文的效率。
可选地，类型判断模块120是设置成根据报文的协议类型，判断报文的类型：当报文的协议类型为OAM协议时，报文的类型为加速报文；当报文的协议类型不是OAM协议时，报文的类型为非加速报文。
本实施例在判断报文类型时，根据报文的协议类型对报文进行分类。由于CFM、BFD等OAM协议对时间性能的要求很高，因此可将协议类型为OAM协议的报文作为加速报文进行加速处理。此外，还可以预先在报文中设置报文类型，预先定义报文是否需要加速处理，根据预先定义的类型进行分类。或者，还可以预先定义处理报文的优先级别，将优先级别高于阈值的报文作为加速报文，实现报文分类。本实施例中，根据报文协议类型判断报文是否为加速报文，判断方式简单，有利于提高对报文进行分类的效率，进而提高交换机的处理效率。
图5为本申请提供的加速处理数据的交换机的结构示意图。如图5所示，本实施例提出的加速处理数据的交换机，包括分配装置100、加速处理装置400和常规处理装置500；
分配装置100包括：
类型判断模块120，设置成判断接收的报文的类型；
加速报文添加模块130，设置成当报文的类型为加速报文时，将报文添加到加速报文队列中，以由加速处理装置400处理加速报文队列中的加速报文；
非加速报文添加模块110，设置成当报文的类型为非加速报文时，将报 文添加到非加速报文队列，以由常规处理装置500处理所述非加速报文队列中的非加速报文；
加速处理装置400，设置成处理所述加速报文队列中的加速报文；
常规处理装置500，设置成处理非加速报文队列中的非加速报文。
可选地，加速处理装置400是设置成根据所述加速报文所在的所述加速报文队列，批量处理分别对应于各个加速业务的多个加速报文队列中的加速报文。
可选地，常规处理装置500是设置成限制收发所述非加速报文队列中的非加速报文的速率。
本实施例中的交换机包括了加速处理装置400和常规处理装置500，加速处理装置400和常规处理装置500可以如图5所示，其结构和原理可参照上述实施例，在此不做赘述。由于本实施例中的交换机包括加速处理装置400和常规处理装置500，将加速报文与非加速报文分配在各自对应的加速处理装置400和常规处理装置500进行处理，有利于提高处理加速报文的效率，且避免了因加速处理装置400处理非加速报文而降低处理加速报文的效率，同时，常规处理装置500有能力处理非加速报文以及控制程序，而不需要处理要求较多计算资源的加速报文，从而减少了常规处理装置500的负荷，提高了交换机处理报文的效率。
以上仅为本申请的可选实施例，并非因此限制本申请的专利范围，凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本申请的专利保护范围内。
本领域普通技术人员可以理解上述方法中的全部或部分步骤可通过程序来指令相关硬件(例如处理器)完成，所述程序可以存储于计算机可读存储介质中，如只读存储器、磁盘或光盘等。可选地，上述实施例的全部或部分步骤也可以使用一个或多个集成电路来实现。相应地，上述实施例中的各模块/单元可以采用硬件的形式实现，例如通过集成电路来实现其相应功能，也可以采用软件功能模块的形式实现，例如通过处理器执行存储于存储器中的 程序/指令来实现其相应功能。本发明实施例不限制于任何特定形式的硬件和软件的结合。
工业实用性
本申请中的交换机包括多个处理装置，将加速报文与非加速报文分别给不同的处理装置进行处理，用于处理加速报文的加速处理装置无需判断需要处理的报文所在的报文队列对应的业务的优先级，就能批量地处理加速报文，有利于提高处理加速报文的效率，且避免了因所述加速处理装置处理非加速报文而降低处理加速报文的效率，同时，所述常规处理装置有能力处理非加速报文以及控制程序，而不需要处理要求较多计算资源的加速报文，从而减少了常规处理装置的负荷，提高了交换机处理报文的效率。

Claims (9)

  1. 一种加速处理数据的方法，包括：
    判断接收的报文的类型；
    当所述报文的类型为加速报文时，将所述报文添加到加速报文队列中，以由加速处理装置处理所述加速报文队列中的所述加速报文；
    当所述报文的类型为非加速报文时，将所述报文添加到非加速报文队列中，以由常规处理装置处理所述非加速报文队列中的所述非加速报文。
  2. 如权利要求1所述的加速处理数据的方法，其中，所述当报文的类型为加速报文时，将所述报文添加到加速报文队列中，以由加速处理装置处理所述加速报文队列中的所述加速报文的步骤包括：
    当所述报文的类型为加速报文时，判断所述报文的业务类型；
    按照所述报文的业务类型，将所述报文添加到所述业务类型对应的加速报文队列中，以由所述加速处理装置分别批量加速处理各所述加速报文队列中的加速报文。
  3. 如权利要求1或2所述的加速处理数据的方法，其中，所述判断接收的报文的类型的步骤包括：
    根据所述报文的协议类型，判断所述报文的类型：
    当所述报文的协议类型为操作管理维护OAM协议时，所述报文的类型为加速报文；
    当所述报文的协议类型不是OAM协议时，所述报文的类型为非加速报文。
  4. 一种加速处理数据的分配装置，包括：
    类型判断模块，设置成判断接收的报文的类型；
    加速报文添加模块，设置成当所述报文的类型为加速报文时，将所述报文添加到加速报文队列中，以由加速处理装置处理所述加速报文队列中的所述加速报文；
    非加速报文添加模块，设置成当所述报文的类型为非加速报文时，将所 述报文添加到非加速报文队列，以由常规处理装置处理所述非加速报文队列中的非加速报文。
  5. 如权利要求4所述的加速处理数据的分配装置，其中，所述加速报文添加模块是设置成当所述报文的类型为加速报文时，判断所述报文的业务类型；
    按照所述报文的业务类型，将所述报文添加到所述业务类型对应的加速报文队列中，以由所述加速处理装置分别批量加速处理各所述加速报文队列中的加速报文。
  6. 如权利要求4或5所述的加速处理数据的分配装置，其中，所述类型判断模块是设置成根据所述报文的协议类型，判断所述报文的类型：当所述报文的协议类型为操作管理维护OAM协议时，则所述报文的类型为加速报文；当所述报文的协议类型不是OAM协议时，则所述报文的类型为非加速报文。
  7. 一种加速处理数据的交换机，包括分配装置、加速处理装置和常规处理装置；其中，
    所述分配装置为权利要求4至6中任一项所述的分配装置；
    所述加速处理装置，设置成处理所述加速报文队列中的加速报文；
    所述常规处理装置，设置成处理所述非加速报文队列中的非加速报文。
  8. 如权利要求7所述的加速处理数据的交换机，其中，所述加速处理装置是设置成根据所述加速报文所在的所述加速报文队列，批量处理分别对应于各个加速业务的多个所述加速报文队列中的加速报文。
  9. 根据权利要求7或8所述的加速处理数据的交换机，其中，所述常规处理装置是设置成限制收发所述非加速报文队列中的非加速报文的速率。
PCT/CN2016/083045 2015-06-26 2016-05-23 加速处理数据的方法、分配装置和交换机 WO2016206513A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510366655.5 2015-06-26
CN201510366655.5A CN106330762A (zh) 2015-06-26 2015-06-26 交换机加速处理数据的方法、cpu核和交换机

Publications (1)

Publication Number Publication Date
WO2016206513A1 true WO2016206513A1 (zh) 2016-12-29

Family

ID=57584603

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/083045 WO2016206513A1 (zh) 2015-06-26 2016-05-23 加速处理数据的方法、分配装置和交换机

Country Status (2)

Country Link
CN (1) CN106330762A (zh)
WO (1) WO2016206513A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111327526A (zh) * 2018-12-14 2020-06-23 就肆电竞股份有限公司 资料传输加速***

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108667765A (zh) * 2017-03-28 2018-10-16 深圳市中兴微电子技术有限公司 一种数据处理方法及装置
CN107566289A (zh) * 2017-08-21 2018-01-09 杭州迪普科技股份有限公司 一种基于流分类的控制核流量限速方法及装置
CN107426113B (zh) * 2017-09-13 2020-03-17 迈普通信技术股份有限公司 报文接收方法及网络设备
CN111294291B (zh) * 2020-01-16 2022-07-12 新华三信息安全技术有限公司 一种协议报文的处理方法和装置
CN113114584B (zh) * 2021-03-01 2023-02-28 杭州迪普科技股份有限公司 一种网络设备的保护方法及装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1913486A (zh) * 2005-08-10 2007-02-14 中兴通讯股份有限公司 一种增强协议报文安全的方法和装置
CN101056222A (zh) * 2007-05-17 2007-10-17 华为技术有限公司 一种深度报文检测方法、网络设备及***
CN101471854A (zh) * 2007-12-29 2009-07-01 华为技术有限公司 一种转发报文的方法及装置
CN101541038A (zh) * 2009-04-27 2009-09-23 杭州华三通信技术有限公司 增强无线局域网承载的上层应用稳定性的方法和设备

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1913486A (zh) * 2005-08-10 2007-02-14 中兴通讯股份有限公司 一种增强协议报文安全的方法和装置
CN101056222A (zh) * 2007-05-17 2007-10-17 华为技术有限公司 一种深度报文检测方法、网络设备及***
CN101471854A (zh) * 2007-12-29 2009-07-01 华为技术有限公司 一种转发报文的方法及装置
CN101541038A (zh) * 2009-04-27 2009-09-23 杭州华三通信技术有限公司 增强无线局域网承载的上层应用稳定性的方法和设备

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111327526A (zh) * 2018-12-14 2020-06-23 就肆电竞股份有限公司 资料传输加速***

Also Published As

Publication number Publication date
CN106330762A (zh) 2017-01-11

Similar Documents

Publication Publication Date Title
WO2016206513A1 (zh) 加速处理数据的方法、分配装置和交换机
US11422839B2 (en) Network policy implementation with multiple interfaces
KR101670642B1 (ko) 클라이언트 디바이스 상에서의 패킷 송신을 스케줄링하기 위한 시스템 및 방법
US10423790B2 (en) Intelligent identification of stressed machines for data security management
US10942786B2 (en) Network management
CN108432194B (zh) 一种拥塞处理的方法、主机及***
US10764132B2 (en) Scale-out association method and apparatus, and system
US10545896B2 (en) Service acceleration method and apparatus
WO2016107152A1 (zh) 虚拟机实例的部署方法和装置及设备
CN104301066A (zh) 数据包传输方法及装置
CN102891809B (zh) 多核网络设备报文按接口保序方法及***
US10263809B2 (en) Selecting an optimal network device for reporting flow table misses upon expiry of a flow in a software defined network
WO2015101091A1 (zh) 一种分布式资源调度方法及装置
US10476746B2 (en) Network management method, device, and system
WO2017185615A1 (zh) 一种业务处理设备的业务状态确定方法及调度设备
CN102611630B (zh) 一种报文接收控制方法及***
WO2016149945A1 (zh) 一种生命周期事件的处理方法及vnfm
JP2011203810A (ja) サーバ、計算機システム及び仮想計算機管理方法
WO2017173728A1 (zh) 一种广播消息发送方法及装置、移动终端
CN106470421A (zh) 一种防止恶意终端非法占用核心网资源的方法和设备
US11012378B2 (en) Methods and apparatus for shared buffer allocation in a transport node
JP2016046669A (ja) パケット処理装置、プログラム及び方法
CN111427673B (zh) 一种负载均衡方法、装置及设备
CN114124856A (zh) 一种流量控制的方法、装置及存储介质
US20220214926A1 (en) Virtual machine monitoring device, virtual machine monitoring method, and program

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: 16813629

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16813629

Country of ref document: EP

Kind code of ref document: A1