WO2016202224A1 - 一种传输层参数调整方法和装置 - Google Patents

一种传输层参数调整方法和装置 Download PDF

Info

Publication number
WO2016202224A1
WO2016202224A1 PCT/CN2016/085542 CN2016085542W WO2016202224A1 WO 2016202224 A1 WO2016202224 A1 WO 2016202224A1 CN 2016085542 W CN2016085542 W CN 2016085542W WO 2016202224 A1 WO2016202224 A1 WO 2016202224A1
Authority
WO
WIPO (PCT)
Prior art keywords
network device
forwarding path
identifier
authentication
indicator
Prior art date
Application number
PCT/CN2016/085542
Other languages
English (en)
French (fr)
Inventor
程剑
尤建洁
李峰
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
Publication of WO2016202224A1 publication Critical patent/WO2016202224A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/12—Shortest path evaluation
    • H04L45/124—Shortest path evaluation using a combination of metrics

Definitions

  • the present invention relates to the field of communications technologies, and in particular, to a transmission layer parameter adjustment method and apparatus.
  • transport layer protocols are used to provide transport services for end-to-end connections. This service is divided into two types: reliable and unreliable.
  • the Transmission Control Protocol (English: Transmission Control Protocol, TCP for short) is a typical reliable transmission service, and the User Datagram Protocol (English: User Datagram Protocol, UDP for short). It is a typical unreliable transmission service.
  • the transport layer protocol also has a datagram congestion control protocol (English: Datagram Congestion Control Protocol, DCCP) and a flow control transmission protocol (English: Stream Control Transmission Protocol, SCTP), but on the network. Most of the traffic is transmitted over TCP and UDP.
  • TCP throughput is related to a variety of factors and can be formulated as follows:
  • the BW is the link bandwidth
  • the WindowSize is the congestion window
  • the RTT is the link delay
  • the MSS is the maximum fragmented packet
  • c is a constant
  • ⁇ is the system packet loss rate.
  • the above formula indicates that the throughput of TCP is less than or equal to the minimum value of three expressions.
  • the physical meaning is that when the link delay and packet loss rate are not particularly large, the maximum throughput of TCP is limited by the link bandwidth; When the link delay is large and the packet loss rate is not particularly large, the maximum throughput of TCP is limited by the congestion window of the sender. When the packet loss rate is large, the maximum throughput of TCP is limited by the packet loss rate.
  • link bandwidth, link delay and packet loss rate are important external factors.
  • the size of the congestion window is an important operational indicator.
  • the current operating system (such as Linux, FreeBSD, etc.) has realized the function of recording indicators (English: metrics), that is, whenever a stream ends, the indicator of the stream is recorded, and the Internet of the receiving end of the stream is used.
  • the protocol (English: Internet Protocol, IP address for short) is used as an index to save the indicator of the stream. For example, in FIG. 1, when the host H2 sends a TCP stream to the host H1, the TCP Metrics recorded locally in the H2 is indexed by the IP address 10.1.1.10 of the H1, and the link delay, the packet loss rate, the congestion window, and the chain are recorded.
  • H2 and H1 can be clients, such as computers, mobile phones and other terminal devices, or can be cloud side, data center or network side servers.
  • the previously saved indicator can be used as a reference to set the TCP parameters, so that the newly established TCP stream can avoid some blind heuristics and speed up congestion.
  • the convergence speed of the window improves the transmission efficiency of the message.
  • Equal-Cost Multipath Routing if there are multiple different forwarding paths to the same destination address, only the IP address of the peer is used as the Metrics record.
  • the index is not appropriate because the forwarding path of each stream may be different, and the link delay, packet loss rate, and link bandwidth may be different for different forwarding paths.
  • FIG 2 there are two forwarding paths for the flow from host H2 to host H1, that is, R4>R2>R1 and R4>R3>R1, which forwarding path is established by the flow established by H2, and the ECMP-capable router R4 decides that H2 does not know or care, which will result in the matching record of the queried transport layer metrics may have no reference value for a certain stream, and thus can not adjust the transport layer parameters of the stream.
  • the embodiment of the present invention provides a method and a device for adjusting a parameter of a transport layer.
  • the IP address of the peer is used as the index of the Metrics record
  • the matching record of the transport layer indicator queried by the sender of the stream may not be adjusted.
  • the present invention provides a method for adjusting a transmission layer parameter, including:
  • the first device obtains a matching record in the pre-stored transport layer indicator, and adjusts a transport layer parameter of the stream according to the content of the matching record, where the index of the matching record includes an IP address of the second device. And an identifier of the network device included in the first forwarding path.
  • the determining, by the first device, the identifier of the network device that is included in the first forwarding path of the flow includes:
  • the method further includes:
  • the first device creates a new record in the pre-stored transport layer indicator, where the index of the new record includes the second device The IP address and the identifier of the network device included in the first forwarding path.
  • the method further includes:
  • Adjusting, by the first device, the transport layer parameters of the stream according to the content of the matching record including:
  • the first device adjusts the transport layer parameters of the stream according to the content of the matching record and/or the obtained index of the network device.
  • the first device acquires an indicator of the network device from a network device corresponding to the determined network device identifier Thereafter, the method further includes:
  • the obtained metric of the network device is added to the new record.
  • the first device sends the acquired five-tuple to the branch point, and receives the branch
  • the decision information returned by the point including:
  • the first device sends the acquired quintuple to the branch point through the SPUD message, and receives the decision information returned by the branch point through the SPUD message.
  • an indicator of the network device including:
  • the first device acquires an indicator of the network device from the network device by using an SPUD message.
  • the indicator of the network device includes at least one of the following:
  • the bandwidth includes at least two of currently used bandwidth, available bandwidth, and total bandwidth.
  • the first device determines the first After forwarding the identifier of the network device included in the path, the method further includes:
  • the first device requests, by using an SPUD message, at least one network device included in the first forwarding path to reserve resources for the stream; and/or
  • the first device requests, by using the SPUD message, at least one network device included in the first forwarding path to set a forwarding priority of the flow to a high priority.
  • the first device determines the first Before forwarding the identifier of the network device included in the path, the method further includes:
  • the first device sends an authentication and authentication request to the authentication and authentication controller
  • the first device receives a key that is returned by the authentication and authentication controller after the authentication and authentication request, and the key is used to identify the identity and authority of the first device;
  • the method further includes:
  • the first device When the first device sends a request to the network device included in the first forwarding path by using the SPUD message, the first device carries the key in the SPUD message, so as to receive the requested network device according to the maintained endpoint.
  • the permission table and the key determine whether to respond to the request.
  • the present invention provides a transmission layer parameter adjustment apparatus, including:
  • a processing unit configured to determine an identifier of the network device included in the first forwarding path of the flow, where the flow flows from the first device to the second device by using the first forwarding path;
  • an adjusting unit configured to obtain a matching record in the pre-stored transport layer indicator, and adjust a transport layer parameter of the stream according to the content of the matching record, where the index of the matching record includes an IP address of the second device And an identifier of the network device included in the first forwarding path.
  • the processing unit is specifically configured to:
  • branch point Determining a branch point of the plurality of forwarding paths from the first device to the second device, where the branch point is a network device that intersects the plurality of forwarding paths starting from the first device and has a branch;
  • the apparatus further includes:
  • An update unit configured to create a new record in the pre-stored transport layer indicator when the matching record is not obtained in the pre-stored transport layer indicator, where the index of the new record includes the second device The IP address and the identifier of the network device included in the first forwarding path.
  • the device further includes:
  • An obtaining unit configured to acquire an indicator of the network device from a network device corresponding to the determined network device identifier
  • the adjusting unit is specifically configured to:
  • the updating unit is further configured to:
  • the adjusting unit acquires the matching record, update the content in the matching record by using the obtained indicator of the network device;
  • the obtained metric of the network device is added to the new record.
  • the processing unit is specifically configured to:
  • the acquired five-tuple is sent to the branch point through the SPUD message, and the decision information returned by the branch point through the SPUD message is received.
  • the obtaining unit is specifically configured to:
  • the indicator of the network device includes at least one of the following:
  • the bandwidth includes at least two of currently used bandwidth, available bandwidth, and total bandwidth.
  • the device further includes:
  • a requesting unit configured to request, by using an SPUD message, at least one network device included in the first forwarding path to reserve resources for the flow; and/or requesting, by using an SPUD message, at least one network included in the first forwarding path
  • the device sets the forwarding priority of the stream to a high priority.
  • the device further includes:
  • An authentication and authentication unit configured to send an authentication and authentication request to the authentication and authentication controller; and receive a key returned by the authentication and authentication controller after passing the authentication and authentication request, the key
  • the identifier and the authority for identifying the first device when sending a request to the network device included in the first forwarding path by using an SPUD message, carrying the key in the SPUD message, so as to receive the The requesting network device determines whether to respond to the request based on the maintained endpoint permission table and the key.
  • the identifier of the network device included in the forwarding path of the flow is used as the index of the matching record, so that the sending end of the stream is in accordance with the obtained transport layer before the flow is created or during the process of sending the flow.
  • the matching record of the indicator adjusts the transport layer parameters of the stream.
  • FIG. 2 is a schematic diagram of an ECMP scenario in the prior art
  • FIG. 3 is a schematic flowchart of a method for adjusting a parameter of a transport layer according to an embodiment of the present disclosure
  • FIG. 4 is a schematic diagram of a branch point of multiple forwarding paths according to an embodiment of the present invention.
  • FIG. 5 is a schematic diagram of a forwarding path of multiple streams according to an embodiment of the present invention.
  • FIG. 6 is a schematic structural diagram of a transmission layer in the prior art
  • FIG. 7 is a schematic structural diagram of a transport layer according to an embodiment of the present disclosure.
  • FIG. 8 is a schematic diagram of a SPUD message according to an embodiment of the present invention.
  • FIG. 9 is a schematic diagram of a process for applying for resource reservation and differential services according to an embodiment of the present invention.
  • FIG. 10 is a schematic diagram of a process of authentication and authentication according to an embodiment of the present invention.
  • FIG. 11 is a schematic structural diagram of a transmission layer parameter adjustment apparatus according to an embodiment of the present invention.
  • FIG. 12 is a schematic structural diagram of another transmission layer parameter adjustment apparatus according to an embodiment of the present invention.
  • the embodiment of the present invention provides a method and a device for adjusting a parameter of a transport layer.
  • the identifier of the network device included in the forwarding path of the flow is used as an index of the matching record, so that the sending end of the stream is in the process of creating the stream or in the process of sending the stream.
  • Adjusting the transport layer parameters of the flow according to the obtained Metrics record in particular, in the scenario where multiple different forwarding paths reach the same destination address in ECMP, the matching record of the obtained transport layer indicator can always be matched with the current stream.
  • the matching of the forwarding paths has a great reference value for the adjustment of the transport layer parameters of the stream, which improves the use effect of the Metrics records in the ECMP scenario.
  • an embodiment of the present invention provides a method for adjusting a transmission layer parameter, and the implementation process thereof is as follows:
  • Step 301 The first device determines an identifier of the network device included in the first forwarding path of the flow, and the flow flows from the first device to the second device by using the first forwarding path.
  • the determining, by the first device, the identifier of the network device that is included in the first forwarding path of the flow may be determining the identifier of all the network devices included in the forwarding path, or determining the identifier of the part of the network device included in the forwarding path.
  • the part of the network device may be one or more network devices.
  • the first device and the second device may be clients, such as a terminal device such as a computer or a mobile phone, or may be a cloud side, a data center, or a server on the network side.
  • the identifier of the network device may be an IP address of the network device, or a name or a character that can uniquely identify the network device, such as R1, R2, and the like in FIG.
  • the first device may implement step 301 by:
  • the first device determines a branch point in a plurality of forwarding paths from the first device to the second device, where the branch point is that the multiple forwarding paths starting from the first device meet A network device that has branches. Then, the first device acquires a quintuple of the stream, and sends the obtained quintuple to the branch point, and receives decision information determined according to the quintuple returned by the branch point, The determining information is used to indicate the identifier of the next network device that flows after the flow passes through the branch point; finally, the first device determines the first one of the multiple forwarding paths according to the decision information. Forwarding path and network included in the first forwarding path The identification of the network device.
  • the branch point is generally a network device with ECMP function, and each branch point may be connected to at least two branch forwarding paths, and the multiple forwarding paths may have one or more branch points.
  • the first device may obtain information about multiple forwarding paths of the second device by using an Internet Control Message Protocol (ICMP) or a centralized query mode. And determining a branch point of the plurality of forwarding paths according to the information of the multiple forwarding paths.
  • ICMP Internet Control Message Protocol
  • FIG 4 there are four forwarding paths for the flow from host H2 to host H1: R7>R5>R4>R2>R1, R7>R5>R4>R3>R1, R7>R6>R4>R2>R1 and R7 >R6>R4>R3>R1, R7 and R4 are the branch points of the four forwarding paths.
  • the branch points of these four forwarding paths are R1 and R4.
  • R1, R4, and R7 all have ECMP functions.
  • the first device does not know the forwarding path of the stream for a specific flow, but for the branch point, it can be a quintuple of a stream ( That is, the source IP address, the source port, the destination IP address, the destination port, and the transport layer protocol are hashed to determine which network device the stream flows to after passing through the branch point, and therefore the forwarding path of the packet of the same stream. It is ok, but the forwarding paths of different streams may be different.
  • the host H2 obtains in advance that the host H1 has two forwarding paths: R4>R2>R1 and R4>R3>R1, and H2 learns from the branch point R4 that the network device after the flow 1 passes the R4 and the next hop is R2, the network device of the next hop after R2 passes R4 is R3, then H2 can determine that the forwarding path of stream 1 is R4>R2>R1 according to the above information, and the forwarding path of stream 2 is R4>R3>R1.
  • Step 302 The first device acquires a matching record in the pre-stored transport layer indicator, and adjusts a transport layer parameter of the stream according to the content of the matching record, where the index of the matching record includes the second device.
  • the index of the pre-stored transport layer indicator includes an IP address of the receiving end of the stream and an identifier of the network device included in the forwarding path of the stream, where the content of the pre-stored transport layer indicator includes the flow direction receiving through the forwarding path.
  • Indicators at the end of the end of the stream such as link delay, packet loss rate, congestion window, and link bandwidth.
  • the pre-stored transport layer indicators may be recorded by means of a Metrics table, a database, or the like.
  • a new record may be created in the pre-stored transport layer indicator, and the new record is indexed.
  • the IP address of the second device and the identifier of the network device included in the first forwarding path are included.
  • the first device may further add other indicators, such as destination port information of the flow, to the index according to a specific use environment.
  • the adjustment of the parameters of BW, congestion window, MSS and c in formula 1 is mainly involved.
  • the congestion window of a newly established flow generally needs to undergo a slow start process, that is, exponentially grow from a relatively small window in the case of no packet loss, and shrink the window if a packet loss occurs until the window converges.
  • the convergence speed of the congestion window of the newly established flow can be accelerated.
  • the Metrics table shown in Table 1 is measured by the sender of the stream, and the link delay, packet loss rate, and bandwidth limitation of the stream, and the network device included in the forwarding path of the stream. It is also closely related. If the information such as delay, packet loss rate and bandwidth on the network device through which the flow passes can be obtained, it has a high reference significance for the sender of the stream.
  • the transport layer belongs to the fourth layer, and according to the OSI layered model, the three layers are transparent to the transport layer, as shown in Figure 6.
  • OSI Open System Interconnection
  • the transport layer of the host only sees and interacts with the transport layer of the peer host, and the intermediate network device is invisible and uninteresting to the transport layer.
  • the host may specifically apply for a piece of memory for communication with the network device, newly created Metrics records, and sent to the network. Messages from the device or messages received from the network device can be stored in this memory.
  • This is equivalent to establishing a new transport layer architecture, as shown in Figure 7, that is, a transport sublayer is added to the traditional transport layer, and the host communicates with the peer host through the original transport layer, through the transport sublayer and The intermediate network device communicates, so that real-time indicators of the network device can be obtained from the network device, and the transmission layer parameters of the flow are adjusted.
  • the specific implementation manner is as follows:
  • the indicator of the network device may be obtained from the network device corresponding to the determined network device identifier, and according to the content of the matching record, And/or the obtained indicator of the network device, adjusting the transport layer parameters of the stream.
  • the obtained metric of the network device may be used. More The newly described content in the matching record; if a new record is created, the obtained indicator of the network device may be added to the new record.
  • the indicator of the network device includes at least one of a packet loss rate, a delay, and a bandwidth, where the bandwidth may include at least two of currently used bandwidth, available bandwidth, and total bandwidth.
  • the Metrics table of the indicators of each network device on the forwarding path of the stream 1 is added to the table 1.
  • the first device may repeatedly obtain an indicator of each network device according to the service requirement of the flow, and the network indicator obtained in real time is beneficial to the adjustment of the transmission layer parameter.
  • the end-to-end link delay in the matching record of the obtained transport layer indicator is 0.1 second (unit: s)
  • the minimum of the available bandwidth of each network device acquired in real time is obtained.
  • the value is 20 megabytes (unit: M)
  • the technical means for obtaining and storing the indicators of the network device does not depend on the technical solution of the present invention, and is not limited to the scenario of the ECMP.
  • the technical means may be implemented in combination with other technical means, or may be implemented separately.
  • the transport layer parameters of the stream are all adjusted.
  • the communication between the application or the host and the network device generally needs to be implemented by the operator's controller, so-called network management.
  • the application or host needs to send the TCP request to the NMS first, and then the NMS forwards it to the network device. This method is cumbersome, especially when the flow is large, and the efficiency is low.
  • the embodiment of the present invention considers that a more flexible protocol, that is, a User Datagram Sublayer Protocol (English: Substrate Protocol for User Datagrams, SPUD).
  • a User Datagram Sublayer Protocol English: Substrate Protocol for User Datagrams, SPUD.
  • the SPUD message is encapsulated in UDP.
  • the header of the SPUD message is in binary format.
  • the message body is not limited to binary format or text format. It is more flexible when using the text format. It carries the application or host to interact with the network device.
  • the message which is more scalable, is equivalent to enhancing the ability of the application or host to interact directly with the network device.
  • the first device may obtain information about multiple forwarding paths that reach the second device by using the SPUD message, and may also send the obtained quintuple to the branch point through the SPUD message, and receive Said The decision information returned by the branch point through the SPUD message.
  • the first device may also obtain an indicator of the network device from the network device by using an SPUD message.
  • the embodiment of the present invention collects the indicators of the network device by using the SPUD message, so that the transport layer protocol can be used for reference, and the transport layer can also use the SPUD message to put forward some requirements for the network device.
  • different streams such as video streams, picture streams, common file streams, etc.
  • the first device can forward the stream for the high priority stream, request the forwarding of the stream.
  • a device on the path reserves more resources for the flow and provides a higher forwarding priority to ensure the service experience of the high-priority flow.
  • the first device may request, by using an SPUD message, at least one network device included in the first forwarding path to reserve resources for the flow, and/or the A device may request, by using an SPUD message, at least one network device included in the first forwarding path to set a forwarding priority of the flow to a high priority.
  • the host H2 when the host H2 requests the resource reservation and the differential service from the R2, the host sends the quintuple of the flow 1 to the SPUD message sent to the R2, and reserves 30 megabytes for the flow 1 application.
  • the first setting is performed before the information of the first forwarding path of the flow is determined, and the authentication and the authentication are required.
  • the specific process may be as shown in FIG. 10:
  • the first device first sends an authentication and authentication request to the authentication and authentication controller, such as process 1 in FIG. Second, the first device receives a key returned by the authentication and authentication controller after passing the authentication and authentication request, and the key is used to identify the identity and rights of the first device, as shown in the figure. Process 2 in 10.
  • the authentication and authentication controller sends the authenticated key of the first device and the authority of the first device to each network device, and each network device maintains an endpoint permission locally. Table, as shown in Process 3 in Figure 10.
  • the privilege includes at least one of the following: a privilege of querying a forwarding path information of a flow, a privilege of querying a network device indicator, a permission to apply for resource reservation, a permission to set a flow priority, and the like, for example,
  • the host H2 has the above four rights.
  • the first device when the first device sends a request to the network device included in the first forwarding path by using the SPUD message, the first device carries the key in the SPUD message, so that the network device that receives the request is maintained according to the maintenance.
  • the endpoint permission table and the key determine whether to respond to the request.
  • the request includes at least one of the following requests: request for forwarding path information of the query stream, and checking Requests for metrics for network devices, requests for resource reservations, and requests for prioritization of flows.
  • requests for forwarding path information of the query stream and checking Requests for metrics for network devices, requests for resource reservations, and requests for prioritization of flows.
  • the data and content carried in the SPUD message are also different. For example, if the request is the forwarding path information of the query flow or the indicator of the network device, the flow and the query to be queried need to be specified in the SPUD message. If the request is reserved for the application resource, the SPUD message needs to specify the flow that needs to be reserved, the reserved resource type, and the reserved value.
  • an embodiment of the present invention provides a transmission layer parameter adjustment apparatus, which is used to implement a transmission layer parameter adjustment method shown in FIG. 3 of the present invention.
  • the apparatus includes:
  • the processing unit 111 is configured to determine an identifier of the network device included in the first forwarding path of the flow, where the flow flows from the first device to the second device by using the first forwarding path.
  • the adjusting unit 112 is configured to obtain a matching record in the pre-stored transport layer indicator, and adjust a transport layer parameter of the stream according to the content of the matching record, where the index of the matching record includes an IP of the second device An address and an identifier of the network device included in the first forwarding path.
  • the processing unit 111 is specifically configured to: determine a branch point in the multiple forwarding paths that reach the second device from the first device, where the branch point is the multiple starting from the first device a network device that intersects and forwards a branch forwarding path; acquires a quintuple of the stream, and sends the obtained quintuple to the branch point, and receives the returned branch point and determines according to the quintuple
  • the decision information is used to indicate an identifier of the next network device that flows after the flow passes through the branch point; and the first forwarding path is determined in the multiple forwarding paths according to the decision information And determining an identifier of the network device included in the first forwarding path.
  • the device further includes:
  • An update unit configured to create a new record in the pre-stored transport layer indicator when the matching record is not obtained in the pre-stored transport layer indicator, where the index of the new record includes the second device The IP address and the identifier of the network device included in the first forwarding path.
  • the device further includes:
  • An obtaining unit configured to acquire an indicator of the network device from a network device corresponding to the determined network device identifier
  • the adjusting unit 112 is specifically configured to: adjust a transport layer parameter of the stream according to the content of the matching record, and/or the obtained indicator of the network device.
  • the updating unit is further configured to: when the adjusting unit 112 acquires the matching record, update the content in the matching record by using the obtained indicator of the network device; when creating a new record, Obtained network The indicator of the network device is added to the new record.
  • the processing unit 111 is specifically configured to: send the acquired quintuple through the SPUD message. Giving the branch point, and receiving decision information returned by the branch point through the SPUD message.
  • the acquiring unit is specifically configured to: obtain, by using the SPUD message, an indicator of the network device from the network device.
  • the indicator of the network device includes at least one of a packet loss rate, a delay, and a bandwidth; and the bandwidth includes at least two of currently used bandwidth, available bandwidth, and total bandwidth.
  • the device further includes:
  • a requesting unit configured to request, by using an SPUD message, at least one network device included in the first forwarding path to reserve resources for the flow; and/or requesting, by using an SPUD message, at least one network included in the first forwarding path
  • the device sets the forwarding priority of the stream to a high priority.
  • the device further includes:
  • An authentication and authentication unit configured to send an authentication and authentication request to the authentication and authentication controller; and receive a key returned by the authentication and authentication controller after passing the authentication and authentication request, the key
  • the identifier and the authority for identifying the first device when sending a request to the network device included in the first forwarding path by using an SPUD message, carrying the key in the SPUD message, so as to receive the The requesting network device determines whether to respond to the request based on the maintained endpoint permission table and the key.
  • the privilege includes at least one of the following: a privilege of querying the forwarding path information of the flow, a privilege of querying the metric of the network device, a permission to apply for resource reservation, and a permission to set the flow priority.
  • the request includes at least one of the following: a request for forwarding path information of the query stream, a request for querying an indicator of the network device, a request for resource reservation, and a request for setting a priority of the flow.
  • an embodiment of the present invention further provides a transmission layer parameter adjustment apparatus, where the apparatus includes:
  • the processor 121 is configured to determine an identifier of the network device included in the first forwarding path of the flow, where the flow flows from the first device to the second device by using the first forwarding path, and obtain a match in the pre-stored transport layer indicator. Recording, and adjusting the transport layer parameter of the stream according to the content of the matching record, where the index of the matching record includes an IP address of the second device and an identifier of the network device included in the first forwarding path.
  • the processor 121 is specifically configured to: determine a branch point in the multiple forwarding paths that reach the second device from the first device, where the branch point is the multiple starting from the first device Strip forwarding paths intersect and exist Branched network devices.
  • the device also includes:
  • the transceiver 122 is configured to acquire a quintuple of the stream, send the obtained quintuple to the branch point, and receive decision information determined according to the quintuple returned by the branch point, The decision information is used to indicate the identity of the next network device to which the flow flows after the branch point.
  • the processor 121 is configured to determine, according to the decision information, the first forwarding path, and the identifier of the network device included in the first forwarding path, in the multiple forwarding paths.
  • the processor 121 is further configured to: if the matching record is not obtained in the pre-stored transport layer indicator, create a new record in the pre-stored transport layer indicator, the new record
  • the index includes an IP address of the second device and an identifier of the network device included in the first forwarding path.
  • the transceiver 122 is further configured to obtain an indicator of the network device from the network device corresponding to the determined network device identifier.
  • the processor 121 is specifically configured to: adjust a transport layer parameter of the stream according to the content of the matching record, and/or the obtained indicator of the network device.
  • the processor 121 is further configured to: if the matching record is obtained, update the content in the matching record by using the obtained indicator of the network device; if the new record is created, the acquired network is obtained. The device's metrics are added to the new record.
  • the transceiver 122 is specifically configured to: send, by using an SPUD message, the acquired five-tuple to the branch point, and receive the decision information that the branch point returns through the SPUD message.
  • the transceiver 122 is configured to: obtain, by using the SPUD message, an indicator of the network device from the network device.
  • the indicator of the network device includes at least one of a packet loss rate, a delay, and a bandwidth; and the bandwidth includes at least two of currently used bandwidth, available bandwidth, and total bandwidth.
  • the transceiver 122 is further configured to: request, by using an SPUD message, at least one network device included in the first forwarding path to reserve resources for the flow; and/or request the first part by using an SPUD message.
  • At least one network device included in a forwarding path sets a forwarding priority of the flow to a high priority.
  • the transceiver 122 is further configured to: send an authentication and authentication request to the authentication and authentication controller; and receive a key that is returned by the authentication and authentication controller after passing the authentication and authentication request.
  • the key is used to identify the identity and authority of the first device; when the SPUD message is sent to the network device included in the first forwarding path, the key is carried in the SPUD message.
  • the network device is based on the dimension The endpoint permission table and the key determine whether to respond to the request.
  • the privilege includes at least one of the following: a privilege of querying the forwarding path information of the flow, a privilege of querying the metric of the network device, a permission to apply for resource reservation, and a permission to set the flow priority.
  • the request includes at least one of the following: a request for forwarding path information of the query stream, a request for querying an indicator of the network device, a request for resource reservation, and a request for setting a priority of the flow.
  • the technical solution provided by the embodiment of the present invention uses the identifier of the network device included in the forwarding path of the flow as the index of the matching record, and uses the indicator of the network device as the matching recorded content by communicating with the network device.
  • the transmitting end of the stream can adjust the transport layer parameters of the stream according to the matching record of the obtained transport layer indicator before the stream is created or during the process of sending the stream.
  • the embodiment of the present invention can also apply for resource reservation and differential service to the network device to ensure the service experience of the high priority flow.
  • embodiments of the present invention can be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware. Moreover, the invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.
  • computer-usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.
  • These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device.
  • the instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.

Abstract

本发明公开了一种传输层参数调整方法和装置，通过将流的转发路径包括的网络设备的标识作为匹配记录的索引，使得流的发送端在创建流之前或在发送流的过程中，根据获取到的传输层指标的匹配记录调整流的传输层参数。该方法为：第一设备确定流的第一转发路径包括的网络设备的标识；所述流通过所述第一转发路径从所述第一设备流向第二设备；所述第一设备在预先存储的传输层指标中获取一匹配记录，并根据所述匹配记录的内容调整所述流的传输层参数，所述匹配记录的索引包括所述第二设备的IP地址和所述第一转发路径包括的网络设备的标识。

Description

一种传输层参数调整方法和装置
本申请要求于2015年6月15日提交中国专利局、申请号为201510330207.X、发明名称为“一种传输层参数调整方法和装置”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，尤其涉及一种传输层参数调整方法和装置。
背景技术
在通信协议中，传输层协议用于为端到端连接提供传输服务。这种服务分为可靠和不可靠两种，其中传输控制协议(英文：Transmission Control Protocol，简称：TCP)是典型的可靠传输服务，而用户数据报协议(英文：User Datagram Protocol，简称：UDP)则是典型的不可靠传输服务。除TCP和UDP外，传输层协议还有数据报拥塞控制协议(英文：Datagram Congestion Control Protocol，简称：DCCP)和流控制传输协议(英文：Stream Control Transmission Protocol，简称：SCTP)等，但网络上绝大部分的流量都是通过TCP和UDP传输的。
虽然不同的传输层协议机制和原理不同，但本质上都是相通的，下面以TCP为例，介绍一下与TCP的吞吐率相关的关键要素。
TCP的吞吐率与多种因素有关，可以用公式描述如下：
Figure PCTCN2016085542-appb-000001
其中，BW为链路带宽，WindowSize为拥塞窗口，RTT为链路时延，MSS为最大分片报文，c为常量，ρ为系统的丢包率。上述公式表示，TCP的吞吐率小于等于三个表达式的最小值，其物理含义是当链路时延和丢包率不是特别大的时候，TCP的最大吞吐率受限于链路带宽；当链路时延较大而丢包率不是特别大的时候，TCP的最大吞吐率受限于发送端的拥塞窗口；当丢包率较大的时候，TCP的最大吞吐率受限于丢包率。
对于传输层协议来说，链路带宽、链路时延和丢包率是重要的外界因素，拥塞窗口的大小是重要的运行指标，这些数据对于传输层协议调整其算法行为，调整发送接收行为，减少拥塞，提高吞吐率都有重要的参考作用。
因此，目前的操作系统(比如Linux，FreeBSD等)都实现了记录指标(英文：metrics)的功能，即每当一条流结束时，就记录这条流的指标，并用这条流的接收端的互联网协议(英文：Internet Protocol，简称：IP)地址作为索引将该条流的指标保存下来。例如，在图1中，主机H2向主机H1发送TCP流时，在H2本地记录的TCP Metrics以H1的IP地址10.1.1.10为索引，记录了链路时延、丢包率、拥塞窗口和链路带宽等指标，其中H2和H1可以是客户端，如计算机、手机等终端设备，也可以是云侧、数据中心或者网络侧的服务器。后续当该IP地址10.1.1.10再和H2建立新的TCP流时，就可以使用之前保存下来的指标作为参考来设置TCP参数，从而使得新建立的TCP流可以避免一些盲目试探的过程，加快拥塞窗口的收敛速度，提高报文的发送效率。
但是，在等价多转发路径(英文：Equal-Cost Multipath Routing，简称：ECMP)的场景下，由于存在多条不同转发路径到达同一目的地址的情况，此时仅使用对端的IP地址作为Metrics记录的索引是不恰当的，因为每次流的转发路径都可能不同，而不同的转发路径，其链路时延、丢包率和链路带宽等都可能不一样。例如，在图2中，主机H2到主机H1的流存在两条转发路径，即R4>R2>R1和R4>R3>R1，H2建立的流具体走哪条转发路径，由具备ECMP能力的路由器R4决定，H2不知道也不关心，这将导致查询到的传输层指标的匹配记录对某条流而言可能完全没有参考价值，从而起不到调整流的传输层参数的作用。
发明内容
本发明实施例提供一种传输层参数调整方法和装置，用以解决仅使用对端的IP地址作为Metrics记录的索引时，流的发送端查询到的传输层指标的匹配记录可能起不到调整流的传输层参数的作用的问题。
第一方面，本发明提供了一种传输层参数调整方法，包括：
第一设备确定流的第一转发路径包括的网络设备的标识；所述流通过所述第一转发路径从所述第一设备流向第二设备；
所述第一设备在预先存储的传输层指标中获取一匹配记录，并根据所述匹配记录的内容调整所述流的传输层参数，所述匹配记录的索引包括所述第二设备的IP地址和所述第一转发路径包括的网络设备的标识。
结合第一方面，在第一方面的第一种可能的实现方式中，所述第一设备确定流的第一转发路径包括的网络设备的标识，包括：
所述第一设备确定从所述第一设备到达第二设备的多条转发路径中的分支点，所述分支点为从所述第一设备出发的所述多条转发路径相交汇且存在分支的网络设备；
所述第一设备获取所述流的五元组，并将获取的五元组发送给所述分支点，以及接收所述分支点返回的根据所述五元组确定的决策信息，所述决策信息用于指示所述流经过所述分支点后流向的下一个网络设备的标识；
所述第一设备所述决策信息，在所述多条转发路径中确定所述第一转发路径以及所述第一转发路径包括的网络设备的标识。
结合第一方面或第一方面的第一种可能的实现方式，在第一方面的第二种可能的实现方式中，所述方法还包括：
若在预先存储的传输层指标中未获取到所述匹配记录，所述第一设备在所述预先存储的传输层指标中创建一条新记录，所述新记录的索引包括所述第二设备的IP地址和所述第一转发路径包括的网络设备的标识。
结合第一方面的第二种可能的实现方式，在第一方面的第三种可能的实现方式中，所述第一设备确定所述第一转发路径包括的网络设备的标识之后，还包括：
所述第一设备从确定的网络设备标识对应的网络设备上获取所述网络设备的指标；
所述第一设备根据所述匹配记录的内容调整所述流的传输层参数，包括：
所述第一设备根据所述匹配记录的内容，和/或获取到的网络设备的指标，调整所述流的传输层参数。
结合第一方面的第三种可能的实现方式，在第一方面的第四种可能的实现方式中，所述第一设备从确定的网络设备标识对应的网络设备上获取所述网络设备的指标之后，所述方法还包括：
若获取到所述匹配记录，则使用获取到的网络设备的指标更新所述匹配记录中的内容；
若创建新记录，则将获取到的网络设备的指标添加到所述新记录中。
结合第一方面的第一种可能的实现方式，在第一方面的第五种可能的实现方式中，所述第一设备将获取的五元组发送给所述分支点，以及接收所述分支点返回的决策信息，包括：
所述第一设备通过SPUD消息，将获取的五元组发送给所述分支点，以及接收所述分支点通过SPUD消息返回的决策信息。
结合第一方面的第三种可能的实现方式，在第一方面的第六种可能的实现方式中， 所述第一设备获取网络设备的指标，包括：
所述第一设备通过SPUD消息，从所述网络设备获取所述网络设备的指标。
结合第一方面的第三种、第四种或第六种可能的实现方式，在第一方面的第七种可能的实现方式中，所述网络设备的指标包括以下至少之一：
丢包率，时延和带宽；
所述带宽包括当前正在使用带宽、可用带宽和总带宽中的至少两个。
结合第一方面和第一方面的第一种至第七种可能的实现方式中的任意一种，在第一方面的第八种可能的实现方式中，所述第一设备确定流的第一转发路径包括的网络设备的标识之后，所述方法还包括：
所述第一设备通过SPUD消息，请求所述第一转发路径包括的至少一个网络设备为所述流预留资源；和/或
所述第一设备通过SPUD消息，请求所述第一转发路径包括的至少一个网络设备将所述流的转发优先级设置为高优先级。
结合第一方面和第一方面的第一种至第八种可能的实现方式中的任意一种，在第一方面的第九种可能的实现方式中，所述第一设备确定流的第一转发路径包括的网络设备的标识之前，所述方法还包括：
所述第一设备向认证与鉴权控制器发送认证与鉴权请求；
所述第一设备接收所述认证与鉴权控制器在通过所述认证与鉴权请求后返回的密钥，所述密钥用于标识所述第一设备的身份和权限；
所述第一设备接收所述认证与鉴权控制器返回的密钥之后，还包括：
所述第一设备在通过SPUD消息向所述第一转发路径包括的网络设备发送请求时，将所述密钥携带在所述SPUD消息中，以便于接收所述请求的网络设备根据维护的端点权限表和所述密钥确定是否响应所述请求。
第二方面，本发明提供了一种传输层参数调整装置，包括：
处理单元，用于确定流的第一转发路径包括的网络设备的标识；所述流通过所述第一转发路径从第一设备流向第二设备；
调整单元，用于在预先存储的传输层指标中获取一匹配记录，并根据所述匹配记录的内容调整所述流的传输层参数，所述匹配记录的索引包括所述第二设备的IP地址和所述第一转发路径包括的网络设备的标识。
结合第二方面，在第二方面的第一种可能的实现方式中，所述处理单元具体用于：
确定从所述第一设备到达第二设备的多条转发路径中的分支点，所述分支点为从所述第一设备出发的所述多条转发路径相交汇且存在分支的网络设备；
获取所述流的五元组，并将获取的五元组发送给所述分支点，以及接收所述分支点返回的根据所述五元组确定的决策信息，所述决策信息用于指示所述流经过所述分支点后流向的下一个网络设备的标识；
根据所述决策信息，在所述多条转发路径中确定所述第一转发路径以及确定所述第一转发路径包括的网络设备的标识。
结合第二方面或第二方面的第一种可能的实现方式，在第二方面的第二种可能的实现方式中，所述装置还包括：
更新单元，用于在预先存储的传输层指标中未获取到所述匹配记录时，在所述预先存储的传输层指标中创建一条新记录，所述新记录的索引包括所述第二设备的IP地址和所述第一转发路径包括的网络设备的标识。
结合第二方面的第二种可能的实现方式，在第二方面的第三种可能的实现方式中，所述装置还包括：
获取单元，用于从确定的网络设备标识对应的网络设备上获取所述网络设备的指标；
所述调整单元具体用于：
根据所述匹配记录的内容，和/或获取到的网络设备的指标，调整所述流的传输层参数。
结合第二方面的第三种可能的实现方式，在第二方面的第四种可能的实现方式中，所述更新单元还用于：
在所述调整单元获取到所述匹配记录时，使用获取到的所述网络设备的指标更新所述匹配记录中的内容；
在创建新记录时，将获取到的所述网络设备的指标添加到所述新记录中。
结合第二方面的第一种可能的实现方式，在第二方面的第五种可能的实现方式中，在将获取的五元组发送给所述分支点，以及接收所述分支点返回的决策信息时，所述处理单元具体用于：
通过SPUD消息，将获取的五元组发送给所述分支点，以及接收所述分支点通过SPUD消息返回的决策信息。
结合第二方面的第三种可能的实现方式，在第二方面的第六种可能的实现方式中， 所述获取单元具体用于：
通过SPUD消息，从所述网络设备获取所述网络设备的指标。
结合第二方面的第三种、第四种或第六种可能的实现方式，在第二方面的第七种可能的实现方式中，所述网络设备的指标包括以下至少之一：
丢包率，时延和带宽；
所述带宽包括当前正在使用带宽、可用带宽和总带宽中的至少两个。
结合第二方面和第二方面的第一种至第七种可能的实现方式中的任意一种，在第二方面的第八种可能的实现方式中，所述装置还包括：
请求单元，用于通过SPUD消息，请求所述第一转发路径包括的至少一个网络设备为所述流预留资源；和/或，通过SPUD消息，请求所述第一转发路径包括的至少一个网络设备将所述流的转发优先级设置为高优先级。
结合第二方面和第二方面的第一种至第八种可能的实现方式中的任意一种，在第二方面的第九种可能的实现方式中，所述装置还包括：
认证与鉴权单元，用于向认证与鉴权控制器发送认证与鉴权请求；接收所述认证与鉴权控制器在通过所述认证与鉴权请求后返回的密钥，所述密钥用于标识所述第一设备的身份和权限；在通过SPUD消息向所述第一转发路径包括的网络设备发送请求时，将所述密钥携带在所述SPUD消息中，以便于接收所述请求的网络设备根据维护的端点权限表和所述密钥确定是否响应所述请求。
利用本发明实施例提供的方案，通过将流的转发路径包括的网络设备的标识作为匹配记录的索引，使得流的发送端在创建流之前或在发送流的过程中，根据获取到的传输层指标的匹配记录调整流的传输层参数。
附图说明
图1为现有技术下TCP Metrics示意图；
图2为现有技术下ECMP场景示意图；
图3为本发明实施例提供的一种传输层参数调整方法的流程示意图；
图4为本发明实施例中多条转发路径的分支点的示意图；
图5为本发明实施例中多条流的转发路径的示意图；
图6为现有技术中传输层的架构示意图；
图7为本发明实施例提供的一种传输层的架构示意图；
图8为本发明实施例提供的SPUD消息示意图；
图9为本发明实施例中申请资源预留与差分业务的过程示意图；
图10为本发明实施例中认证与鉴权的过程示意图；
图11为本发明实施例提供的一种传输层参数调整装置的结构示意图；
图12为本发明实施例提供的另一种传输层参数调整装置的结构示意图。
具体实施方式
本发明实施例提供了一种传输层参数调整方法和装置，通过将流的转发路径包括的网络设备的标识作为匹配记录的索引，使得流的发送端在创建流之前或在发送流的过程中，根据获取到的Metrics记录调整流的传输层参数，尤其是在ECMP即存在多条不同转发路径达到同一目的地址的场景下，能够使得获取到的传输层指标的匹配记录总是与当前的流的转发路径相匹配，对于流的传输层参数的调整有很大的参考价值，提升了Metrics记录在ECMP场景下的使用效果。
下面结合说明书附图和各实施例对本发明技术方案进行说明。
参阅图3所示，本发明实施例提供了一种传输层参数调整方法，其实施流程如下：
步骤301：第一设备确定流的第一转发路径包括的网络设备的标识；所述流通过所述第一转发路径从所述第一设备流向第二设备。
其中,第一设备确定流的第一转发路径包括的网络设备的标识，可以是确定所述转发路径包括的所有的网络设备的标识，也可以是确定所述转发路径包括的部分网络设备的标识，所述部分网络设备可以是一个或多个网络设备。
其中，所述第一设备和第二设备可以是客户端，如计算机、手机等终端设备，也可以是云侧、数据中心或者网络侧的服务器。所述网络设备的标识，可以是网络设备的IP地址，也可以是能唯一标识所述网络设备的名字或字符等，如图2中的R1、R2等。
可选的，在ECMP场景下，所述第一设备可以通过如下方式实现步骤301：
首先，所述第一设备确定从所述第一设备到达第二设备的多条转发路径中的分支点，所述分支点为从所述第一设备出发的所述多条转发路径相交汇且存在分支的网络设备。然后，所述第一设备获取所述流的五元组，并将获取的五元组发送给所述分支点，以及接收所述分支点返回的根据所述五元组确定的决策信息，所述决策信息用于指示所述流经过所述分支点后流向的下一个网络设备的标识；最后，所述第一设备根据所述决策信息，在所述多条转发路径中确定所述第一转发路径以及所述第一转发路径包括的网 络设备的标识。
所述分支点一般为具备ECMP功能的网络设备，每个分支点可以连接至少两条分支转发路径，所述多条转发路径可以有一个或多个分支点。
可选地，所述第一设备可以事先通过网间控制报文协议(英文：Internet Control Message Protocol，简称：ICMP)方式或者集中查询方式，获得到达所述第二设备的多条转发路径的信息，并根据所述多条转发路径的信息确定出多条转发路径的分支点。例如，在图4中，主机H2到主机H1的流存在四条转发路径：R7>R5>R4>R2>R1、R7>R5>R4>R3>R1、R7>R6>R4>R2>R1和R7>R6>R4>R3>R1，R7和R4即为这四条转发路径的分支点；同理，主机H1到主机H2的流也存在四条转发路径，这四条转发路径的分支点为R1和R4，R1、R4和R7均具备ECMP功能。
由于所述分支点的ECMP功能，对于某一条具体的流，所述第一设备并不清楚这条流的转发路径，但对于所述分支点而言，其可以对一条流的五元组(即源IP地址，源端口，目的IP地址，目的端口和传输层协议)进行哈希运算，从而决定该条流经过该分支点后流向哪一个网络设备，因此同一条流的报文的转发路径是确定的，但不同的流的转发路径可能不同。
例如，在图5中，主机H2事先获得达到主机H1有R4>R2>R1和R4>R3>R1两条转发路径，并且H2从分支点R4获知流1经过R4后下一跳的网络设备为R2，流2经过R4后的下一跳的网络设备为R3，那么H2根据以上信息可以确定流1的转发路径为R4>R2>R1，流2的转发路径为R4>R3>R1。
步骤302：所述第一设备在预先存储的传输层指标中获取一匹配记录，并根据所述匹配记录的内容调整所述流的传输层参数，所述匹配记录的索引包括所述第二设备的IP地址和所述第一转发路径包括的网络设备的标识。
其中，所述预先存储的传输层指标的索引包括了流的接收端的IP地址和流的转发路径包括的网络设备的标识，所述预先存储的传输层指标的内容包括了通过该转发路径流向接收端的流结束时的指标，如链路时延、丢包率、拥塞窗口和链路带宽等。所述预先存储的传输层指标可采用Metrics表、数据库等方式进行记录。
可选的，若步骤302中所述第一设备在预先存储的传输层指标中未获取到所述匹配记录，则可以在所述预先存储的传输层指标中创建一条新记录，新记录的索引包括所述第二设备的IP地址和所述第一转发路径包括的网络设备的标识。
例如，假设图5中的主机H1的IP地址是10.1.1.10，则对于流1而言，在主机H2 上获取到的匹配记录或新建的记录可能如下表1所示：
表1 Metrics表
索引：(10.1.1.10,R4,R2,R1)
内容：链路时延，丢包率，拥塞窗口，链路带宽，……
此外，所述第一设备还可以根据具体的使用环境在索引中增添其他指标，如流的目的端口信息等。
在根据所述匹配记录中的内容调整流的传输层参数时，主要涉及对公式1中的BW、拥塞窗口、MSS和c这些参数的调整。例如，一般情况下新建立的流的拥塞窗口一般需要经历一个慢启动的过程，即在不丢包的情况从一个比较小的窗口开始按指数增长，若发生丢包则缩小窗口，直至窗口收敛到一个合适的大小，而根据所述匹配记录中的拥塞窗口，可以加快新建立的流的拥塞窗口的收敛速度。
需要注意的是，如表1所示的Metrics表，是由流的发送端测量到的，而流的链路时延、丢包率以及带宽限制等因素，与流的转发路径包括的网络设备也密切相关，如果能够获得流经过的网络设备上的时延、丢包率和带宽等信息，对于流的发送端具有很高的参考意义。
但在开放式系统互联(英文：Open System Interconnection，简称：OSI)模型中，传输层属于第四层，而根据OSI的分层模型，三层以下对于传输层是透明的，如在图6所示的传统的传输层架构中，主机的传输层只看到对端主机的传输层并与之交互，中间的网络设备对于传输层而言是不可见也不关心的。
而本发明实施例中，为了获得流经过的网络设备上的时延、丢包率和带宽等信息，主机可以专门申请一块内存用于和网络设备的通信，新建立的Metrics记录以及发送给网络设备的消息或从网络设备接收的消息都可以存储在这块内存中。这相当于建立了一种新的传输层架构，如图7所示，即在传统的传输层上增加了一个传输子层，主机通过原来的传输层与对端主机通信，通过传输子层与中间的网络设备通信，从而可以从网络设备上获取网络设备实时的指标，用于调整流的传输层参数，具体实施方式如下：
在所述第一设备确定所述第一转发路径包括的网络设备的标识之后，可以从确定的网络设备标识对应的网络设备上获取所述网络设备的指标，并根据所述匹配记录的内容，和/或获取到的网络设备的指标，调整所述流的传输层参数。
可选的，在从确定的网络设备标识对应的网络设备上获取所述网络设备的指标之后，若所述第一设备获取到所述匹配记录，则可以使用获取到的所述网络设备的指标更 新所述匹配记录中的内容；若创建新记录，则可以将获取到的网络设备的指标添加到所述新记录中。
可选的，所述网络设备的指标包括丢包率，时延和带宽中的至少之一，所述带宽可以包括当前正在使用带宽、可用带宽和总带宽中的至少两个。
如表2所示，为在表1的基础上增加了流1的转发路径上各网络设备的指标的Metrics表。
表2 Metrics表
索引：(10.1.1.10,R4,R2,R1)
内容：链路时延，丢包率，拥塞窗口，链路带宽，……
R4：丢包率，时延，带宽，......
R2：丢包率，时延，带宽，......
R1：丢包率，时延，带宽，......
所述第一设备可以根据所述流的业务需求，重复向各网络设备获取各网络设备自身的指标，这种实时获取的网络指标，有利于传输层参数的调整。以拥塞窗口的实时调整为例，假设获取到的传输层指标的匹配记录中的端到端的链路时延为0.1秒(单位：s)，实时获取到的各网络设备的可用带宽中的最小值为20兆(单位：M)，那么可知当前的流的拥塞窗口的最大值不能超过20*0.1＝2M比特(单位：bit)。
需要说明的是，上述获取并保存网络设备的指标的技术手段并不依赖于本发明的技术方案，也不限定于ECMP的场景，该技术手段可以与其它技术手段配合实施，也可以单独实施，均调整流的传输层参数。
目前应用或主机与网络设备的通信，一般都需要通过运营商的控制器，即所谓的网管来实现。应用或主机需要先将TCP请求发送给网管，再由网管转发给网络设备。这种方式比较麻烦，尤其在流很多的时候，效率较低。
为了解决这个问题，本发明实施例认为可以采用一种更灵活的协议，即用户数据报子层协议(英文：Substrate Protocol for User Datagrams，简称：SPUD)。如图8所示，SPUD消息使用UDP封装，SPUD消息的消息头采用二进制格式，消息体没有限定是二进制格式还是文本格式，使用文本格式时更加灵活，里面承载了应用或主机与网络设备进行交互的消息，它的扩展性较强，相当于增强了应用或主机与网络设备的直接交互的能力。
可选地，所述第一设备可以通过SPUD消息，获得到达所述第二设备的多条转发路径的信息，也可以通过SPUD消息，将获取的五元组发送给所述分支点，以及接收所述 分支点通过SPUD消息返回的决策信息。
可选地，所述第一设备也可以通过SPUD消息，从所述网络设备获取所述网络设备的指标。
本发明实施例利用SPUD消息收集网络设备的指标，以便于传输层协议参考使用，同时传输层也可以利用SPUD消息对网络设备提出某种需求。因第一设备在发送流时，不同的流(如视频流，图片流，普通文件流等)可能有不同的优先级，若第一设备能够针对高优先级的流，请求该条流的转发路径上的设备为该条流预留更多的资源，提供更高的转发优先级，则能够保证高优先级的流的业务体验。具体实施方式如下：
所述第一设备在确定流的第一转发路径的信息之后，可以通过SPUD消息，请求所述第一转发路径包括的至少一个网络设备为所述流预留资源，和/或，所述第一设备可以通过SPUD消息，请求所述第一转发路径包括的至少一个网络设备将所述流的转发优先级设置为高优先级。
例如，如图9所示，主机H2为流1向R2申请资源预留与差分业务时，在发送给R2的SPUD消息中携带流1的五元组，并为流1申请预留30兆(单位：M)的带宽资源，以及将流1在R2上的转发优先级设置为高优先级。相应地，R2可以返回SPUD应答。
可选地，所述第一设在确定流的第一转发路径的信息之前，需要通过认证与鉴权，具体过程可以如图10所示：
所述第一设备首先向认证与鉴权控制器发送认证与鉴权请求，如图10中的过程1。其次，所述第一设备接收所述认证与鉴权控制器在通过所述认证与鉴权请求后返回的密钥，所述密钥用于标识所述第一设备的身份和权限，如图10中的过程2。此外，所述认证与鉴权控制器将认证过的所述第一设备的密钥，以及所述第一设备的权限下发到每个网络设备中，每个网络设备在本地维护一个端点权限表，如图10中的过程3。
可选的，所述权限包括下述权限中的至少一种：查询流的转发路径信息的权限，查询网络设备指标的权限，申请资源预留的权限，设置流优先级的权限等，例如，在图10中，主机H2具备以上四种权限。
后续，所述第一设备在通过SPUD消息向所述第一转发路径包括的网络设备发送请求时，将所述密钥携带在所述SPUD消息中，以便于接收所述请求的网络设备根据维护的端点权限表和所述密钥确定是否响应所述请求。
可选的，所述请求包括下述请求中的至少一种：查询流的转发路径信息的请求、查 询网络设备的指标的请求、申请资源预留和设置流的优先级的请求。发送不同的请求时，SPUD消息中携带的数据和内容也不同，例如，若该请求为查询流的转发路径信息或查询网络设备的指标时，需要在SPUD消息中指定要查询的流和查询的项目，若该请求为申请资源预留时，需要在SPUD消息中指定需要预留资源的流、预留的资源类型和预留的值等。
参阅图11所示，本发明实施例提供了一种传输层参数调整装置，用于实现本发明图3所示的一种传输层参数调整方法，所述装置包括：
处理单元111，用于确定流的第一转发路径包括的网络设备的标识；所述流通过所述第一转发路径从第一设备流向第二设备。
调整单元112，用于在预先存储的传输层指标中获取一匹配记录，并根据所述匹配记录的内容调整所述流的传输层参数，所述匹配记录的索引包括所述第二设备的IP地址和所述第一转发路径包括的网络设备的标识。
其中，
可选的，所述处理单元111具体用于：确定从所述第一设备到达第二设备的多条转发路径中的分支点，所述分支点为从所述第一设备出发的所述多条转发路径相交汇且存在分支的网络设备；获取所述流的五元组，并将获取的五元组发送给所述分支点，以及接收所述分支点返回的根据所述五元组确定的决策信息，所述决策信息用于指示所述流经过所述分支点后流向的下一个网络设备的标识；根据所述决策信息，在所述多条转发路径中确定所述第一转发路径以及确定所述第一转发路径包括的网络设备的标识。
可选的，所述装置还包括：
更新单元，用于在预先存储的传输层指标中未获取到所述匹配记录时，在所述预先存储的传输层指标中创建一条新记录，所述新记录的索引包括所述第二设备的IP地址和所述第一转发路径包括的网络设备的标识。
可选的，所述装置还包括：
获取单元，用于从确定的网络设备标识对应的网络设备上获取所述网络设备的指标；
所述调整单元112具体用于：根据所述匹配记录的内容，和/或获取到的网络设备的指标，调整所述流的传输层参数。
可选的，所述更新单元还用于：在所述调整单元112获取到所述匹配记录时，使用获取到的网络设备的指标更新所述匹配记录中的内容；在创建新记录时，将获取到的网 络设备的指标添加到所述新记录中。
可选的，在将获取的五元组发送给所述分支点，以及接收所述分支点返回的决策信息时，所述处理单元111具体用于：通过SPUD消息，将获取的五元组发送给所述分支点，以及接收所述分支点通过SPUD消息返回的决策信息。
可选的，所述获取单元具体用于：通过SPUD消息，从所述网络设备获取所述网络设备的指标。
可选的，所述网络设备的指标包括丢包率，时延和带宽中的至少之一；所述带宽包括当前正在使用带宽、可用带宽和总带宽中的至少两个。
可选的，所述装置还包括：
请求单元，用于通过SPUD消息，请求所述第一转发路径包括的至少一个网络设备为所述流预留资源；和/或，通过SPUD消息，请求所述第一转发路径包括的至少一个网络设备将所述流的转发优先级设置为高优先级。
可选的，所述装置还包括：
认证与鉴权单元，用于向认证与鉴权控制器发送认证与鉴权请求；接收所述认证与鉴权控制器在通过所述认证与鉴权请求后返回的密钥，所述密钥用于标识所述第一设备的身份和权限；在通过SPUD消息向所述第一转发路径包括的网络设备发送请求时，将所述密钥携带在所述SPUD消息中，以便于接收所述请求的网络设备根据维护的端点权限表和所述密钥确定是否响应所述请求。
可选的，所述权限包括下述权限中的至少一种：查询流的转发路径信息的权限，查询网络设备指标的权限，申请资源预留的权限，设置流优先级的权限。
可选的，所述请求包括下述请求中的至少一种：查询流的转发路径信息的请求、查询网络设备的指标的请求、申请资源预留和设置流的优先级的请求。
参阅图12所示，本发明实施例还提供了一种传输层参数调整装置，所述装置包括：
处理器121，用于确定流的第一转发路径包括的网络设备的标识；所述流通过所述第一转发路径从第一设备流向第二设备；在预先存储的传输层指标中获取一匹配记录，并根据所述匹配记录的内容调整所述流的传输层参数，所述匹配记录的索引包括所述第二设备的IP地址和所述第一转发路径包括的网络设备的标识。
其中，
可选的，所述处理器121具体用于：确定从所述第一设备到达第二设备的多条转发路径中的分支点，所述分支点为从所述第一设备出发的所述多条转发路径相交汇且存在 分支的网络设备。
所述装置还包括：
收发器122，用于获取所述流的五元组，并将获取的五元组发送给所述分支点，以及接收所述分支点返回的根据所述五元组确定的决策信息，所述决策信息用于指示所述流经过所述分支点后流向的下一个网络设备的标识。
所述处理器121具体用于，根据所述决策信息，在所述多条转发路径中确定所述第一转发路径以及确定所述第一转发路径包括的网络设备的标识。
可选的，所述处理器121还用于，若在预先存储的传输层指标中未获取到所述匹配记录，则在所述预先存储的传输层指标中创建一条新记录，所述新记录的索引包括所述第二设备的IP地址和所述第一转发路径包括的网络设备的标识。
可选的，所述收发器122还用于，从确定的网络设备标识对应的网络设备上获取所述网络设备的指标。
相应的，所述处理器121具体用于：根据所述匹配记录的内容，和/或获取到的网络设备的指标，调整所述流的传输层参数。
可选的，所述处理器121还用于：若获取到所述匹配记录，则使用获取到的网络设备的指标更新所述匹配记录中的内容；若创建新记录，则将获取到的网络设备的指标添加到所述新记录中。
可选的，所述收发器122具体用于：通过SPUD消息，将获取的五元组发送给所述分支点，以及接收所述分支点通过SPUD消息返回的决策信息。
可选的，所述收发器122具体用于：通过SPUD消息，从所述网络设备获取所述网络设备的指标。
可选的，所述网络设备的指标包括丢包率，时延和带宽中的至少之一；所述带宽包括当前正在使用带宽、可用带宽和总带宽中的至少两个。
可选的，所述收发器122还用于，通过SPUD消息，请求所述第一转发路径包括的至少一个网络设备为所述流预留资源；和/或，通过SPUD消息，请求所述第一转发路径包括的至少一个网络设备将所述流的转发优先级设置为高优先级。
可选的，所述收发器122还用于，向认证与鉴权控制器发送认证与鉴权请求；接收所述认证与鉴权控制器在通过所述认证与鉴权请求后返回的密钥，所述密钥用于标识所述第一设备的身份和权限；在通过SPUD消息向所述第一转发路径包括的网络设备发送请求时，将所述密钥携带在所述SPUD消息中，以便于接收所述请求的网络设备根据维 护的端点权限表和所述密钥确定是否响应所述请求。
可选的，所述权限包括下述权限中的至少一种：查询流的转发路径信息的权限，查询网络设备指标的权限，申请资源预留的权限，设置流优先级的权限。
可选的，所述请求包括下述请求中的至少一种：查询流的转发路径信息的请求、查询网络设备的指标的请求、申请资源预留和设置流的优先级的请求。
综上所述，采用本发明实施例提供的技术方案，通过将流的转发路径包括的网络设备的标识作为匹配记录的索引，以及通过与网络设备的通信将网络设备的指标作为匹配记录的内容，便于流的发送端在创建流之前或在发送流的过程中，可以根据获取到的传输层指标的匹配记录调整流的传输层参数。并且本发明实施例还可以向网络设备申请资源预留和差分服务，保证高优先级流的业务体验。
本领域内的技术人员应明白，本发明的实施例可提供为方法、系统、或计算机程序产品。因此，本发明可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本发明可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本发明是参照根据本发明实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
显然，本领域的技术人员可以对本发明实施例进行各种改动和变型而不脱离本发明实施例的范围。这样，倘若本发明实施例的这些修改和变型属于本发明权利要求及其等同技术的范围之内，则本发明也意图包含这些改动和变型在内。

Claims (20)

  1. 一种传输层参数调整方法，其特征在于，包括：
    第一设备确定流的第一转发路径包括的网络设备的标识；所述流通过所述第一转发路径从所述第一设备流向第二设备；
    所述第一设备在预先存储的传输层指标中获取一匹配记录，并根据所述匹配记录的内容调整所述流的传输层参数，所述匹配记录的索引包括所述第二设备的互联网协议IP地址和所述第一转发路径包括的网络设备的标识。
  2. 如权利要求1所述的方法，其特征在于，所述第一设备确定流的第一转发路径包括的网络设备的标识，包括：
    所述第一设备确定从所述第一设备到达第二设备的多条转发路径中的分支点，所述分支点为从所述第一设备出发的所述多条转发路径相交汇且存在分支的网络设备；
    所述第一设备获取所述流的五元组，并将获取的五元组发送给所述分支点，以及接收所述分支点返回的根据所述五元组确定的决策信息，所述决策信息用于指示所述流经过所述分支点后流向的下一个网络设备的标识；
    所述第一设备根据所述决策信息，在所述多条转发路径中确定所述第一转发路径以及所述第一转发路径包括的网络设备的标识。
  3. 如权利要求1或2所述的方法，其特征在于，所述方法还包括：
    若在预先存储的传输层指标中未获取到所述匹配记录，所述第一设备在所述预先存储的传输层指标中创建一条新记录，所述新记录的索引包括所述第二设备的IP地址和所述第一转发路径包括的网络设备的标识。
  4. 如权利要求3所述的方法，其特征在于，所述第一设备确定所述第一转发路径包括的网络设备的标识之后，还包括：
    所述第一设备从确定的网络设备标识对应的网络设备上获取所述网络设备的指标；
    所述第一设备根据所述匹配记录的内容调整所述流的传输层参数，包括：
    所述第一设备根据所述匹配记录的内容，和/或获取到的网络设备的指标，调整所述流的传输层参数。
  5. 如权利要求4所述的方法，其特征在于，所述第一设备从确定的网络设备标识对应的网络设备上获取所述网络设备的指标之后，所述方法还包括：
    若获取到所述匹配记录，则使用获取到的所述网络设备的指标更新所述匹配记录中的内容；
    若创建新记录，则将获取到的所述网络设备的指标添加到所述新记录中。
  6. 如权利要求2所述的方法，其特征在于，所述第一设备将获取的五元组发送给所述分支点，以及接收所述分支点返回的决策信息，包括：
    所述第一设备通过用户数据报子层协议SPUD消息，将获取的五元组发送给所述分支点，以及接收所述分支点通过SPUD消息返回的决策信息。
  7. 如权利要求4所述的方法，其特征在于，所述第一设备从网络设备获取所述网络设备的指标，包括：
    所述第一设备通过SPUD消息，从所述网络设备获取所述网络设备的指标。
  8. 如权利要求4、5或7所述的方法，其特征在于，所述网络设备的指标包括以下至少之一：
    丢包率，时延和带宽；
    所述带宽包括当前正在使用带宽、可用带宽和总带宽中的至少两个。
  9. 如权利要求1-8中任意一项所述的方法，其特征在于，所述第一设备确定流的第一转发路径包括的网络设备的标识之后，所述方法还包括：
    所述第一设备通过SPUD消息，请求所述第一转发路径包括的至少一个网络设备为所述流预留资源；和/或
    所述第一设备通过SPUD消息，请求所述第一转发路径包括的至少一个网络设备将所述流的转发优先级设置为高优先级。
  10. 如权利要求1-9中任意一项所述的方法，其特征在于，所述第一设备确定流的第一转发路径包括的网络设备的标识之前，所述方法还包括：
    所述第一设备向认证与鉴权控制器发送认证与鉴权请求；
    所述第一设备接收所述认证与鉴权控制器在通过所述认证与鉴权请求后返回的密钥，所述密钥用于标识所述第一设备的身份和权限；
    所述第一设备接收所述认证与鉴权控制器返回的密钥之后，还包括：
    所述第一设备在通过SPUD消息向所述第一转发路径包括的网络设备发送请求时，将所述密钥携带在所述SPUD消息中，以便于接收所述请求的网络设备根据维护的端点权限表和所述密钥确定是否响应所述请求。
  11. 一种传输层参数调整装置，其特征在于，包括：
    处理单元，用于确定流的第一转发路径包括的网络设备的标识；所述流通过所述第一转发路径从第一设备流向第二设备；
    调整单元，用于在预先存储的传输层指标中获取一匹配记录，并根据所述匹配记录的内容调整所述流的传输层参数，所述匹配记录的索引包括所述第二设备的互联网协议IP地址和所述第一转发路径包括的网络设备的标识。
  12. 如权利要求11所述的装置，其特征在于，所述处理单元具体用于：
    确定从所述第一设备到达第二设备的多条转发路径中的分支点，所述分支点为从所述第一设备出发的所述多条转发路径相交汇且存在分支的网络设备；
    获取所述流的五元组，并将获取的五元组发送给所述分支点，以及接收所述分支点返回的根据所述五元组确定的决策信息，所述决策信息用于指示所述流经过所述分支点后流向的下一个网络设备的标识；
    根据所述决策信息，在所述多条转发路径中确定所述第一转发路径以及确定所述第一转发路径包括的网络设备的标识。
  13. 如权利要求11或12所述的装置，其特征在于，所述装置还包括：
    更新单元，用于在预先存储的传输层指标中未获取到所述匹配记录时，在所述预先存储的传输层指标中创建一条新记录，所述新记录的索引包括所述第二设备的IP地址和所述第一转发路径包括的网络设备的标识。
  14. 如权利要求13所述的装置，其特征在于，所述装置还包括：
    获取单元，用于从确定的网络设备标识对应的网络设备上获取所述网络设备的指标；
    所述调整单元具体用于：
    根据所述匹配记录的内容，和/或获取到的网络设备的指标，调整所述流的传输层参数。
  15. 如权利要求14所述的装置，其特征在于，所述更新单元还用于：
    在所述调整单元获取到所述匹配记录时，使用获取到的所述网络设备的指标更新所述匹配记录中的内容；
    在创建新记录时，将获取到的所述网络设备的指标添加到所述新记录中。
  16. 如权利要求12所述的装置，其特征在于，在将获取的五元组发送给所述分支点，以及接收所述分支点返回的决策信息时，所述处理单元具体用于：
    通过用户数据报子层协议SPUD消息，将获取的五元组发送给所述分支点，以及接收所述分支点通过SPUD消息返回的决策信息。
  17. 如权利要求14所述的装置，其特征在于，所述获取单元具体用于：
    通过SPUD消息，从所述网络设备获取所述网络设备的指标。
  18. 如权利要求14、15或17所述的装置，其特征在于，所述网络设备的指标包括以下至少之一：
    丢包率，时延和带宽；
    所述带宽包括当前正在使用带宽、可用带宽和总带宽中的至少两个。
  19. 如权利要求11-18中任意一项所述的装置，其特征在于，所述装置还包括：
    请求单元，用于通过SPUD消息，请求所述第一转发路径包括的至少一个网络设备为所述流预留资源；和/或，通过SPUD消息，请求所述第一转发路径包括的至少一个网络设备将所述流的转发优先级设置为高优先级。
  20. 如权利要求11-19中任意一项所述的装置，其特征在于，所述装置还包括：
    认证与鉴权单元，用于向认证与鉴权控制器发送认证与鉴权请求；接收所述认证与鉴权控制器在通过所述认证与鉴权请求后返回的密钥，所述密钥用于标识所述第一设备的身份和权限；在通过SPUD消息向所述第一转发路径包括的网络设备发送请求时，将所述密钥携带在所述SPUD消息中，以便于接收所述请求的网络设备根据维护的端点权限表和所述密钥确定是否响应所述请求。
PCT/CN2016/085542 2015-06-15 2016-06-13 一种传输层参数调整方法和装置 WO2016202224A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510330207.X 2015-06-15
CN201510330207.XA CN106330386B (zh) 2015-06-15 2015-06-15 一种传输层参数调整方法和装置

Publications (1)

Publication Number Publication Date
WO2016202224A1 true WO2016202224A1 (zh) 2016-12-22

Family

ID=57545009

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/085542 WO2016202224A1 (zh) 2015-06-15 2016-06-13 一种传输层参数调整方法和装置

Country Status (2)

Country Link
CN (1) CN106330386B (zh)
WO (1) WO2016202224A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112751763A (zh) * 2019-10-30 2021-05-04 北京华为数字技术有限公司 一种报文转发方法、设备、存储介质及系统

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102139378B1 (ko) * 2018-11-20 2020-07-29 울산과학기술원 혼잡 제어 방법 및 장치
WO2021000238A1 (zh) * 2019-07-01 2021-01-07 深圳市大疆创新科技有限公司 消息流显示方法和装置、无人驾驶系统以及可移动平台
CN111600796B (zh) * 2020-05-20 2021-10-01 中国电子科技集团公司第五十四研究所 一种基于可配置解析字段的流识别装置及方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020118649A1 (en) * 2001-02-05 2002-08-29 Tantivy Communications, Inc. Application specific traffic optimization in a wireless link
CN101742366A (zh) * 2009-12-18 2010-06-16 北京邮电大学 一种光分组网络中qos保证的参数化标签转发方法
CN103379042A (zh) * 2012-04-25 2013-10-30 丛林网络公司 路径加权的等价多路径

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE10301265A1 (de) * 2003-01-15 2004-07-29 Siemens Ag Verfahren und Anordnung zum Routing von Datenpaketen in einem paketvermittelnden Datennetz
CN101662842B (zh) * 2009-10-09 2012-01-11 天津大学 一种应用于无线自组织网络的tcp参数调节方法

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020118649A1 (en) * 2001-02-05 2002-08-29 Tantivy Communications, Inc. Application specific traffic optimization in a wireless link
CN101742366A (zh) * 2009-12-18 2010-06-16 北京邮电大学 一种光分组网络中qos保证的参数化标签转发方法
CN103379042A (zh) * 2012-04-25 2013-10-30 丛林网络公司 路径加权的等价多路径

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
"Draft Rel-6 PSS Quality Metrics Permanent Document", 3GPP TSG-SA4 MEETING #30, TDOC S4-040073 ., 27 February 2004 (2004-02-27), XP050637861 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112751763A (zh) * 2019-10-30 2021-05-04 北京华为数字技术有限公司 一种报文转发方法、设备、存储介质及系统

Also Published As

Publication number Publication date
CN106330386B (zh) 2019-10-25
CN106330386A (zh) 2017-01-11

Similar Documents

Publication Publication Date Title
US10574763B2 (en) Session-identifer based TWAMP data session provisioning in computer networks
US10244082B2 (en) Selecting and monitoring a plurality of services key performance indicators using TWAMP
CN108234235B (zh) 用于数据监控的方法、网络设备以及计算机可读存储介质
US10498652B2 (en) Method and system of application-aware routing with crowdsourcing
US10425382B2 (en) Method and system of a cloud-based multipath routing protocol
US20200296011A1 (en) Satisfying service level agreement metrics for unknown applications
CN107948076B (zh) 一种转发报文的方法及装置
US7835285B2 (en) Quality of service, policy enhanced hierarchical disruption tolerant networking system and method
EP3761591B1 (en) Tcp link configuration method, apparatus, and computer program product
US9800551B2 (en) AVC Bi-directional correlation using an overlay fabric header
US9509631B2 (en) Quality of service (QoS) for information centric networks
WO2017113238A1 (zh) 一种建立路由表的方法、电子设备及网络
WO2016202224A1 (zh) 一种传输层参数调整方法和装置
WO2016034029A1 (zh) 业务流量的处理方法和装置
Pana et al. A Survey on the Evolution of RSVP
US20220200915A1 (en) Network policy application based on session state
US20130275552A1 (en) Virtual desktop system
CN102480503B (zh) P2p流量识别方法和装置
US10374944B2 (en) Quality of service for data transmission
WO2015096734A1 (zh) 一种业务数据的下行传输方法及分组数据网关
JP4705619B2 (ja) 受付制御方法、無線ｌａｎ基地局装置およびプログラム
US10708188B2 (en) Application service virtual circuit
WO2024067640A1 (zh) 一种协议数据单元集合传输方法及装置
US20240114323A1 (en) Apparatus and method for providing service function chaining service exposure in wireless communication system
WO2023202467A1 (zh) 网络配置方法及装置

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: 16810965

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: 16810965

Country of ref document: EP

Kind code of ref document: A1