CN114389972B - Packet loss detection method and device and storage medium - Google Patents
Packet loss detection method and device and storage medium Download PDFInfo
- Publication number
- CN114389972B CN114389972B CN202210171121.7A CN202210171121A CN114389972B CN 114389972 B CN114389972 B CN 114389972B CN 202210171121 A CN202210171121 A CN 202210171121A CN 114389972 B CN114389972 B CN 114389972B
- Authority
- CN
- China
- Prior art keywords
- storage unit
- data packet
- packet
- count value
- target storage
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 63
- 239000000872 buffer Substances 0.000 claims abstract description 82
- 238000013507 mapping Methods 0.000 claims abstract description 12
- 238000000034 method Methods 0.000 claims description 32
- 230000005540 biological transmission Effects 0.000 description 12
- 230000008569 process Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000001052 transient effect Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000008439 repair process Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000007723 transport mechanism Effects 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0823—Errors, e.g. transmission errors
- H04L43/0829—Packet loss
Abstract
A packet loss detection method applied to a source terminal provided with a first buffer zone, wherein the first buffer zone comprises at least one storage unit, the storage unit stores a count value, and the packet loss detection method comprises: the source end obtains a data packet to be sent, generates an index value for the data packet according to a preset generation mode, maps the data packet to a storage unit of the first buffer area according to the index value in a preset mapping mode, which is called a first target storage unit, and updates the count value of the first target storage unit: adding a count to the existing count value of the first target storage unit to serve as the updated count value of the first target storage unit; and the source end adds the count value updated by the first target storage unit to the packet head of the data packet and sends the data packet. According to the scheme provided by the embodiment, the count value is maintained at the source end and the count value is maintained at the destination end, and the count values of the source end and the count value are compared, so that packet loss detection is realized.
Description
Technical Field
The present disclosure relates to network technologies, and in particular, to a method and apparatus for packet loss detection, and a storage medium.
Background
In large computer network systems, packet loss and network failures due to packet loss are unlikely to be completely avoided due to persistent or transient route black holes, route loops, network congestion, hardware failures, and Gray failures (Gray failures). For example, a forwarding rule configuration error may create a routing black hole, such that packets with a certain source or destination IP address are systematically discarded. In a software defined network, the forwarding rules of the various switches need to be updated frequently, but their updates tend to be unsynchronized, which may create transient routing black holes in the network. For example, there is no forwarding rule for destination address x on a certain switch B, so it defaults to discard all packets with destination address x. Now the controller configures the switches in the network with new forwarding rules that require switch a to forward packets with destination address x to B, which then forwards these packets to switch C. At this time, if a forwards packets with destination address x to B before the forwarding rules of B are validated, B discards the packets, thereby creating a transient routing black hole. In addition, improper configuration of forwarding rules for routers and switches may create loops in the network, so that when a packet enters the routing loop, its TTL becomes smaller, and eventually, when the TTL of the packet decreases to 0, it is discarded.
Although the above packet loss types can be relieved by means of network configuration verification and the like, packet loss caused by network congestion cannot be completely avoided. For example, when packets entering the switch from multiple input ports are forwarded from one output port, if the total rate of the incoming packets continues to exceed the sending rate of the output port, the packets are continuously accumulated in the buffer area of the output port, and finally the buffer area of the output port overflows to cause packet loss. Increasing the buffer capacity of the output port is not useful in the face of this type of packet loss, since in the face of network rates up to hundreds of Gbps, the larger buffers are also rapidly filled. One solution is that the system can timely sense the situation when packet loss occurs, so that the traffic of an upstream node can be scheduled and split in real time, and the overflow of an output buffer zone of the node is avoided. Although many congestion control algorithms based on transmission control protocol (Transmission Control Protocol, TCP), such as data center transmission control protocol (Data Center Transmission Control Protocol, DCTCP), L2DCT, LPD, D2TCP, etc., can be used to solve the problem of packet loss caused by buffer overflow, in many scenarios, a network administrator does not have management authority to an end host, so that a corresponding TCP scheme cannot be deployed, and a congestion control mechanism of TCP cannot solve network congestion caused by non-TCP traffic (such as UDP traffic generated by online audio/video playing), so that a manner of detecting packet loss information in a network in real time to coordinate traffic of network devices inside the network to avoid packet loss still has important significance.
And the random packet loss and even complete link disconnection caused by hardware faults such as link breakage, optical module looseness or backboard circuit faults are also included. There is no software solution for such packet loss, and therefore, only the way to replace the optical fiber, reinsert the optical module, or restart the switch is possible. The key to dealing with such packet loss is to find out the packet loss in time and report the packet loss to network operation staff, so as to avoid the substantial influence of the packet loss on the network performance, and therefore it is important to detect the packet loss in real time.
Disclosure of Invention
The embodiment of the application provides a packet loss detection method and device, and a storage medium, and packet loss is detected in time.
The embodiment of the application provides a packet loss detection method, which is applied to a source end provided with a first buffer zone, wherein the first buffer zone comprises at least one storage unit, the storage unit stores a count value, and the packet loss detection method comprises the following steps:
the source end obtains a data packet to be sent, generates an index value for the data packet according to a preset generation mode, maps the data packet to a storage unit of the first buffer area according to the index value in a preset mapping mode, which is called a first target storage unit, and updates the count value of the first target storage unit, wherein the updating mode is as follows: adding a count to the existing count value of the first target storage unit to serve as the updated count value of the first target storage unit;
And the source end adds the count value updated by the first target storage unit to the packet head of the data packet and sends the data packet.
In an exemplary embodiment, the storage unit further stores a digest, and the digest is cleared initially;
before sending the data packet, the method further comprises: updating the digest of the first target storage unit as follows: performing exclusive OR operation on the existing abstract of the first target storage unit and the identification information of the data packet to obtain an updated abstract of the first target storage unit;
and adding the updated abstract of the first target storage unit into the packet head of the data packet.
In an exemplary embodiment, the source is provided with a counter;
the generating the index value for the data packet according to the preset generation mode includes: generating an index value for the data packet according to the count value of the counter;
after generating an index value for the data packet, further comprising, taking a count value of the counter as a count value of the counter by modulo m after increasing the count value of the counter once, wherein m is the number of storage units of the first buffer;
before sending the data packet, adding the index value to a packet head of the data packet.
In an exemplary embodiment, the generating the index value for the data packet according to the preset generation manner includes: performing hash operation according to the information of the data packet to generate the index value; or determining the index value according to the checksum of the data packet.
In an exemplary embodiment, the information of the data packet includes at least one first type of information, wherein the first type of information of different data packets of the same data stream is different.
In an exemplary embodiment, the first type of information includes a payload of a data packet.
In an exemplary embodiment, the information of the data packet further includes at least one of: the source address of the data packet, the destination address of the data packet, the transport protocol, the source port number, and the destination port number.
In an exemplary embodiment, the adding, by the source end, the count value and the digest of the first target storage unit to the header of the data packet includes: the source end adds the count value and the abstract of the first target storage unit into the packet head of the data packet according to preset probability;
the method further comprises the steps of: the source end adds the index value to the packet header of the data packet with the probability of 1-the preset probability, or the source end keeps the packet header of the data packet unchanged with the probability of 1-the preset probability.
The embodiment of the disclosure provides a packet loss detection method, which is applied to a destination end provided with a second buffer zone, wherein the second buffer zone comprises at least one storage unit, the storage unit stores a count value, and the packet loss detection method comprises the following steps:
after the destination receives the data packet, an index value is obtained from a packet header of the data packet or the index value is generated for the data packet according to a preset generation mode, the data packet is mapped to one storage unit of the second buffer area according to the index value in a preset mapping mode, the storage unit is called a second target storage unit, and the count value of the second target storage unit is updated according to the following mode: adding a count to the existing count value of the second target storage unit to serve as the updated count value of the second target storage unit;
and judging that the packet loss exists when the packet head of the data packet carries the count value of the first target storage unit of the source end and the updated count value of the second target storage unit is different from the count value of the first target storage unit.
In an exemplary embodiment, the storage unit further stores a digest, and the digest is cleared initially;
The method further includes updating the digest of the second target storage unit as follows: performing exclusive OR operation on the existing abstract of the second target storage unit and the identification information of the data packet, and then using the exclusive OR operation as an updated abstract of the second target storage unit;
when judging that the packet loss exists and the count value of the second target storage unit is different from the count value of the first target storage unit by one time of counting, acquiring the abstract of the first target storage unit from the packet head of the data packet, performing exclusive OR operation on the abstract updated by the second target storage unit and the abstract of the first target storage unit, and determining the identification information of the lost data packet according to the obtained result.
In an exemplary embodiment, the method further comprises: and when judging that the packet loss exists, updating the count value of the second target storage unit by using the count value of the first target storage unit, and updating the abstract of the second target storage unit by using the abstract of the first target storage unit.
In an exemplary embodiment, the generating the index value for the data packet according to the preset generation manner includes:
performing hash operation according to the information of the data packet to generate the index value; or determining the index value according to the checksum carried by the data packet.
The embodiment of the disclosure provides a packet loss detection device, which comprises a memory and a processor, wherein the memory stores a program, and the program realizes the packet loss detection method according to any embodiment when being read and executed by the processor.
An embodiment of the present disclosure provides a computer readable storage medium, where one or more programs are stored, and the one or more programs may be executed by one or more processors to implement the packet loss detection method according to any one embodiment.
Compared with the related art, the embodiment of the application includes a packet loss detection method applied to a source end provided with a first buffer zone, wherein the first buffer zone includes at least one storage unit, the storage unit stores a count value, and the packet loss detection method includes: the source end obtains a data packet to be sent, generates an index value for the data packet according to a preset generation mode, maps the data packet to a storage unit of the first buffer area according to the index value in a preset mapping mode, which is called a first target storage unit, and updates the count value of the first target storage unit, wherein the updating mode is as follows: adding a count to the existing count value of the first target storage unit to serve as the updated count value of the first target storage unit; and the source end adds the count value updated by the first target storage unit to the packet head of the data packet and sends the data packet. According to the scheme provided by the embodiment, the count value is maintained at the source end and the count value is maintained at the destination end, and the count values of the source end and the count value are compared, so that packet loss detection is realized.
Additional features and advantages of the application will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the application. Other advantages of the present application may be realized and attained by the structure particularly pointed out in the written description and drawings.
Drawings
The accompanying drawings are included to provide an understanding of the technical aspects of the present application, and are incorporated in and constitute a part of this specification, illustrate the technical aspects of the present application and together with the examples of the present application, and not constitute a limitation of the technical aspects of the present application.
Fig. 1 is a schematic diagram of a packet loss detection system according to an exemplary embodiment;
fig. 2 is a flowchart of a packet loss detection method (source end) according to an exemplary embodiment;
fig. 3 is a flowchart of a packet loss detection method (destination end) according to an exemplary embodiment;
FIG. 4 is a schematic diagram of packet transmission according to an exemplary embodiment;
fig. 5 is a schematic diagram of a packet loss detection apparatus according to an exemplary embodiment.
Detailed Description
The present application describes a number of embodiments, but the description is illustrative and not limiting and it will be apparent to those of ordinary skill in the art that many more embodiments and implementations are possible within the scope of the embodiments described herein. Although many possible combinations of features are shown in the drawings and discussed in the detailed description, many other combinations of the disclosed features are possible. Any feature or element of any embodiment may be used in combination with, or in place of, any other feature or element of any other embodiment, except where expressly limited.
The present application includes and contemplates combinations of features and elements known to those of ordinary skill in the art. The embodiments, features and elements of the present disclosure may also be combined with any conventional features or elements to form a unique inventive arrangement as defined in the claims. Any feature or element of any embodiment may also be combined with features or elements from other inventive arrangements to form another unique inventive arrangement as defined in the claims. Thus, it should be understood that any of the features shown and/or discussed in this application may be implemented alone or in any suitable combination. Accordingly, the embodiments are not to be restricted except in light of the attached claims and their equivalents. Further, various modifications and changes may be made within the scope of the appended claims.
Furthermore, in describing representative embodiments, the specification may have presented the method and/or process as a particular sequence of steps. However, to the extent that the method or process does not rely on the particular order of steps set forth herein, the method or process should not be limited to the particular sequence of steps described. Other sequences of steps are possible as will be appreciated by those of ordinary skill in the art. Accordingly, the particular order of the steps set forth in the specification should not be construed as limitations on the claims. Furthermore, the claims directed to the method and/or process should not be limited to the performance of their steps in the order written, and one skilled in the art can readily appreciate that the sequences may be varied and still remain within the spirit and scope of the embodiments of the present application.
There is also a very common packet loss situation in real network scenarios, which is random packet loss caused by various reasons. For this type of packet loss, the network may occasionally lose some data packets at first, but generally still operates normally, and even the user cannot feel the degradation of the network performance, but if the packet loss cannot be corrected in time, various factors that cause the packet loss of the network are accumulated, so that the network performance is obviously degraded and even the system crashes, which is called gray Failure (Grey Failure).
In actual network operation and maintenance work, when a network user finds that the use experience is reduced, it is first determined whether the problem is caused by network failure, because hardware problems of application software, protocol stacks and end hosts may cause the use experience of the user to be reduced. After determining that the current problem is caused by a network failure, it is also necessary to locate the location where the network failure occurred, and the specific type of network failure. The above steps often require a longer time than repair of a particular network failure. In this time, if there is packet loss in the network, and when there is packet loss in the network, if it is able to find a specific location where packet loss occurs and obtain information such as a flow identifier of the packet loss, it is able to quickly locate a network failure and determine a type of the network failure according to the information such as the flow identifier of the packet loss, thereby greatly improving efficiency of repairing the network failure. Furthermore, if accurate real-time packet loss detection can be realized, the packet loss detection can be prevented from going gradually, when the packet loss occurs at first, the packet loss detection is reported to a network operation and maintenance system, and whether the packet loss is to be interfered is judged by the network operation and maintenance system, so that the network performance is repaired before the network performance is reduced, and the system can be ensured to run stably for a long time.
The embodiment of the disclosure provides a packet loss detection method, which is characterized in that a source end and a destination end maintain count values related to data packets according to the processed data packets, and the count values of the source end and the destination end are compared so as to realize packet loss detection. In another exemplary embodiment, digests related to the identification information of the data packets are also maintained at the source end and the destination end, and the identification information of the lost data packets is recovered through the digests, so that network faults can be rapidly located, the types of the network faults can be determined, and the efficiency of network fault repair is improved.
Fig. 1 is a schematic diagram of a packet loss detection system according to an exemplary embodiment. As shown in fig. 1, the packet loss detection system includes a source end and a destination end. When a data packet is transmitted from one end of a link to the other end, the end from which the data packet leaves is called the source end, and the end from which the data packet arrives is called the destination end. And respectively maintaining a buffer zone at the source end and the destination end, wherein the buffer zone at the source end is called a first buffer zone, the buffer zone at the destination end is called a second buffer zone, the first buffer zone comprises at least one storage unit, the second buffer zone comprises at least one storage unit, and each storage unit of the first buffer zone and the second buffer zone is provided with two fields of a summary and a count value. The buffer areas of the source end and the destination end are S and D respectively, S [ i ] is a storage unit with index i in S, and D [ i ] is a storage unit with index i in D. S and D have the same capacity m, i.e. S has m memory cells and D has m memory cells. In an exemplary implementation, the source may maintain a constant δ,0 < δ+.1.
In an exemplary embodiment, a packet loss detection method is provided, where the packet loss detection method may have two working modes: the normal working mode and the energy-saving working mode can be used according to a user instruction or system configuration, or can be dynamically switched between the two working modes, for example, the corresponding working mode is decided to be used according to the current network bandwidth resource. However, the embodiments of the present disclosure are not limited thereto, and only one of the operation modes may be used in the packet loss detection method. The implementation of the two modes of operation is described below separately.
In the normal working mode, the source end maintains a counter, the count value of the counter is n, and the initial value of the counter defaults to 0. Before a source end sends a data packet p, an index value is allocated to the data packet p and a counter is updated, and the value of the index value k and the updating mode of the counter are as follows:
k←n
n←(n+1)％m
that is, the current count value n of the counter is used as the index value k allocated to the packet p, and the count value is added with 1 and modulo m again to be used as the updated count value.
According to the index value k, the data packet p is mapped to a storage unit S [ k ] of the first buffer, and the source updates the first buffer as follows (specifically, updates the summary and count value of the storage unit S [ k ] mapped by the index value k):
S [ k ]. Count value ≡S [ k ]. Count value +1
That is, the memory cell S [ k ]]Existing digest value S [ k ]]Exclusive-or operation is performed on the summary and the identification information p of the data packet, and the result is used as a storage unit S k]The updated digest value is used to determine, among other things,the exclusive or operation is not described later;
the count value S [ k ] existing in the storage unit S [ k ] is increased by one count to be used as the updated count value of the storage unit S [ k ].
In the present embodiment, the count value may be incremented by 1 by one count value, but the embodiment of the present disclosure is not limited thereto, and the count value may be represented by another value.
The identification information of the data packet may include some information carried by the packet header, including but not limited to at least one of the following: source address of the data packet, destination address of the data packet, transport protocol, source port number, destination port number, etc.
Wherein, according to the index value k, the storage unit S [ k ] mapping the data packet to the first buffer is only an example, and other mapping rules may be set to map the index to k to another storage unit of the first buffer. The destination end and the source end follow the same mapping rule.
Then, the source adds the triplet (S [ k ] digest, S [ k ] count value, k) as a new header to the packet header of the packet with the probability of δ, and adds the index value k as a new header (excluding two fields of S [ k ] digest and S [ k ] count value) to the packet header with the probability of 1- δ, and transmits the packet.
One implementation is as follows:
generating a random number x, wherein x is more than or equal to 0 and less than or equal to 1;
when x is less than or equal to delta, the triplet (S [ k ]. Abstract, S [ k ]. Count value, k) is added as a new header to the header of the data packet, and when x > delta, the index value k is added as a new header to the header of the data packet.
The above implementation scheme of adding the triples (S k. Abstract, S k. Count value, k) as a new header to the header of the data packet with delta probability, because the bandwidth occupied by S k. Abstract, S k. Count value is large, the above triples are added only in part of the data packet, so that the consumed bandwidth can be reduced.
The above adding different headers to the header of the packet with the probability of δ and the probability of 1- δ, respectively, is merely an example, and in another exemplary embodiment, the triples (S [ k ]. Abstract, S [ k ]. Count value, k) may be added to the header of the packet as a new header for each packet without maintaining δ; or, every preset number of data packets, adding the triples (S [ k ]. Abstract, S [ k ]. Count value, k) as a new header to the packet header of the data packet, and adding the index value k as a new header to the packet header of the rest data packets; or, adding the triples (S [ k ]. Abstract, S [ k ]. Count value, k) as a new header to the packet header of the data packet at preset time intervals, and adding the index value k as a new header to the packet header of the rest of the data packets; etc.
In the above implementation, the 1 st to the m-th data packets are sequentially mapped to the storage units with the indexes of 0 to m-1 in the first buffer, the (m+1) -th data packets are sequentially mapped to the storage units with the indexes of 0 to m-1 in the first buffer, and the first buffer is recycled, so that the first buffer may be also referred to as a first ring buffer, and the second buffer may be also referred to as a second ring buffer. However, the present disclosure is not limited thereto and the 1 st to m-th packets may be otherwise mapped to the respective memory locations of the first buffer.
After the destination receives the data packet p, the header related to packet loss detection is parsed from the packet header of the data packet p. Since both the above-mentioned header portions contain the index value k, after the index value k is parsed, a storage unit of the second buffer mapped by the index value k is determined, and the second buffer is updated as follows (specifically, the digest and count value of the mapped storage unit are updated):
d [ k ] count value ≡Dk ] count value +1
That is, the digest value Dk existing in the storage unit Dk is exclusive-or operated with the identification information p of the packet p, and the result is used as the updated digest value of the storage unit Dk;
The count value D [ k ] existing in the storage unit D [ k ] is incremented by one count to be used as the count value updated by the storage unit D [ k ].
The destination end will determine whether the header of the data packet contains the digest and the count maintained by the source end. If the summary and the count value are included, packet loss detection is started, the summary and the count value resolved from the header of the data packet are respectively recorded as p-summary and p-count value, wherein p-summary is the summary S [ k ] summary of the mapped storage unit S [ k ] before the data packet p is sent at the source end, and p-count value is the count value S [ k ] of the mapped storage unit S [ k ] before the data packet p is sent at the source end. The destination end compares the D [ k ] count value with the p-count value and processes the two according to the following three conditions:
1. if p. count value = D [ k ]. Count value, no packet loss is detected;
2. if p. count value-D [ k ]. Count value = one count, say 1, indicates that exactly one packet was lost in the storage unit D [ k ], and the identification information of this packet can be recovered by:
the identification information of the lost data packet can be obtained by exclusive OR operation of the D < k >. Abstract and the p < abstract.
3. If the p-count value-D [ k ]. Count value > one count, for example, > 1, it means that the storage unit D [ k ] loses more than one packet than the storage unit S [ k ], so that the identification information of the lost packet cannot be recovered.
In the above case, as long as the p.count value is not equal to the D [ k ]. Count value, it is explained that the first buffer at the source end and the second buffer at the destination end do not coincide, and therefore S [ k ] and D [ k ] are synchronized in the following manner:
d [ k ]. Count value ≡p-
D [ k ] abstract ≡p ] abstract
In theory, the p-count value is greater than or equal to Dk-count value, so only the three cases need to be discussed, but if the p-count value is less than Dk-count value in the actual deployment process, only the third case is needed to be referred to, the possible packet loss information is not recovered, and the states of S [ k ] and Dk ] are synchronized.
According to the scheme provided by the embodiment, the packet loss can be detected rapidly, the packet loss position can be judged, the packet loss information can be recovered, the type of network faults and the occurrence position can be judged, and therefore the reliability of the network is improved.
In the power saving mode of operation, the source does not maintain the counter n. The source may perform a hash operation according to the information of the data packet to generate an index value k, and map the data packet to a storage unit S [ k ] in the first buffer according to the index value k. Updating the summary and count value of the storage unit S [ k ], the updating method referring to the updating method in the normal operation mode, namely:
S [ k ]. Count value ≡S [ k ]. Count value +1
The source end adds the S [ k ] count value and the S [ k ] abstract as a new header to the header of the data packet p with the probability of delta, and does not add any new header to the header of the data packet p with the probability of 1-delta, so as to send the data packet p. In the energy-saving operation mode, the index value k is not transferred between the source terminal and the destination terminal, and thus bandwidth consumption can be significantly reduced.
In an exemplary embodiment, the source may use an MD5 (MD 5 Message-Digest Algorithm) or a CRC (Cyclic Redundancy Check ) Algorithm for the hash operation. The hash operation is performed according to the information of the data packet to generate the index value k, which may be obtained by modulo m of the value obtained by the hash operation, where m is the number of storage units in the second buffer.
In an exemplary embodiment, the information of the data packet includes at least one first type of information, wherein the first type of information of different data packets of the same data stream is different. The first type of information may include a payload of a data packet. The first type of information can distinguish between different packets of the same data flow, because in network failures such as routing black holes, packets with specific IP addresses are discarded, and if there is no first type of information in the input to perform the hash operation, multiple lost packets are mapped to the same storage unit, which results in failure to recover the information such as the flow identifier of the lost packet. By taking the first type of information as an input to the hash operation, the lost packets with the same five-tuple can also be dispersed in different memory cells, so that their identification information can be recovered conveniently.
In an exemplary embodiment, the information of the data packet may further include, but is not limited to, at least one of the following: the source address of the packet (e.g., source IP address), the destination address of the packet (e.g., destination IP address), the transport protocol, the source port number, the destination port number. The information of the data packet does not include fields such as TTL (Time To Live) that may change during the transmission of the data packet, which may cause inconsistent index values generated by the source and destination ends.
After receiving a data packet p, the destination end performs hash operation according to the information of the data packet p in the same calculation manner as the source end to generate an index value k, determines a storage unit D [ k ] of the second buffer area mapped by the index value k, and updates the summary and the count value of the storage unit D [ k ] of the second buffer area as follows:
d [ k ] count value ≡Dk ] count value +1
If the packet p carries a count value (called p-count value) and a digest (p-digest) of the source, the destination uses it to detect possible packet loss and recover the identification information of the packet, that is,
1. if p. count value = D [ k ]. Count value, no packet loss is detected;
2. if p. count value-D [ k ]. Count value = one count, say 1, indicates that exactly one packet was lost in the storage unit D [ k ], and the identification information of this packet can be recovered by:
3. If the p-count value-Dk-count value > one count, for example, > 1, it means that the storage unit Dk has lost more than one packet than the storage unit SK, so that the identification information of the lost packet cannot be recovered.
When the count value of the source end carried in the data packet is inconsistent with the count value in D [ K ], the count value of the source end carried in the data packet p is used for synchronizing the count value and the abstract in D [ K ].
During the transmission of the data packet, the payload of the data packet, even the source and destination IP addresses, etc. are changed intentionally or unintentionally, which also causes the index values calculated by the source and destination to be inconsistent, thereby causing errors in packet loss detection. In an exemplary embodiment, packet loss detection may be performed only on packets such as TCP/UDP that contain a checksum, because TCP/UDP can quickly detect packets that are accidentally modified through the checksum, and when the checksum verification of a packet fails, the destination peer regards it as a packet loss without using the packet to update the second buffer of the destination peer. When a subsequent packet arrives, the information of the packet may be recovered. That is, before the source end sends the data packet, it judges whether the data packet is TCP or UDP data packet,
In fact, because the checksum of the TCP/UDP packets does not change during packet transmission and has a strong randomness, their checksum generation index values can be used as an index to the storage locations in the buffer to which the packets are mapped. Although the scheme can not detect the packet loss of the data packets except the TCP/UDP data packets, the scheme can still achieve a very good effect in the packet loss detection of the real network scene by considering that most of the data packets in the real network are TCP/UDP data packets. That is, in an exemplary embodiment, the source and destination may take the checksum modulo m of the data packet as the index value k. The remaining processes may refer to processes in the power saving mode of operation. Taking packet p as an example. The processing of the source end comprises the following steps: the source end takes the checksum modulo m of the data packet as an index value k, maps the index value k to a storage unit of the first buffer area, updates the abstract and the count value of the mapped storage unit, adds the updated abstract and count value to the packet header of the data packet p as a new header with the probability of delta, does not add any new header to the packet header of the data packet p with the probability of 1-delta, and sends the data packet p. The processing of the destination comprises the following steps: the destination receives the data packet p, takes the checksum modulo m of the data packet as an index value k, maps to a storage unit of the second buffer area according to the index value k, updates the abstract and the count value of the mapped storage unit, compares the count value of the mapped storage unit of the destination with the count value carried in the received data packet, judges whether the packet is lost or not, and when the packet is lost, restores the identification information of the packet loss according to the abstract of the mapped storage unit of the destination and the abstract carried in the received data packet, and when the packet is lost, respectively updates the abstract and the count value of the mapped storage unit of the destination according to the abstract and the count value carried in the data packet.
Fig. 2 is a flowchart of a packet loss detection method according to an embodiment of the present disclosure. As shown in fig. 2, an embodiment of the present disclosure provides a packet loss detection method, which is applied to a source end provided with a first buffer, where the first buffer includes at least one storage unit, and the storage unit stores a count value, and the packet loss detection method includes:
step 201, the source end obtains a data packet to be sent, generates an index value for the data packet according to a preset generation mode, maps the data packet to a storage unit of the first buffer area according to the index value in a preset mapping mode, which is called a first target storage unit, and updates a count value of the first target storage unit, wherein the updating mode is as follows: adding a count to the existing count value of the first target storage unit to serve as the updated count value of the first target storage unit;
step 202, the source end adds the count value updated by the first target storage unit to the header of the data packet, and sends the data packet.
According to the scheme provided by the embodiment, the index value can be generated for the data packet at the source end, the count value corresponding to the index value is maintained and sent to the destination end, so that the destination end can determine the index value corresponding to the data packet after receiving the data packet, and whether the packet is lost can be judged by comparing the count values of the destination end and the source end according to the count value of the storage unit of the destination end obtained by the index value.
In an exemplary embodiment, the storage unit further stores a digest, and the digest is cleared initially;
before sending the data packet, the method further comprises: updating the digest of the first target storage unit as follows: performing exclusive OR operation on the existing abstract of the first target storage unit and the identification information of the data packet to obtain an updated abstract of the first target storage unit;
and adding the updated abstract of the first target storage unit into the packet head of the data packet.
According to the scheme provided by the embodiment, the abstract information related to the identification information of the data packet is maintained, and the abstract is sent to the destination end, so that the destination end can recover the identification information of the lost packet according to the abstract when the lost packet exists, thereby locating the network fault in time and determining the type of the network fault, and further repairing the network fault in time, improving the network reliability.
In an exemplary embodiment, the source is provided with a counter;
the generating an index value for the data packet includes: generating an index value for the data packet according to the count value of the counter;
After generating an index value for the data packet, further comprising, taking a count value of the counter as a count value of the counter by modulo m after increasing the count value of the counter once, wherein m is the number of storage units of the first buffer;
before sending the data packet, adding the index value to a packet head of the data packet.
The scheme provided by the embodiment can generate the index value according to the counter value by setting the counter.
In an exemplary embodiment, the generating the index value for the data packet may include: performing hash operation according to the information of the data packet to generate the index value; or determining the index value according to the checksum of the data packet. The information of the data packet may include: at least one of a payload of a data packet, a source address of the data packet, a destination address of the data packet, a transport protocol, a source port number, a destination port number, and the like. And carrying out hash operation according to the information of the data packet to generate a hash value, wherein m is the number of storage units of the first buffer area, and the hash value is taken as an index value. The determining the index value according to the checksum of the data packet includes: and taking the checksum modulo m of the data packet as an index value.
In an exemplary embodiment, the adding, by the source end, the count value and the digest of the first target storage unit to the header of the data packet includes: the source end adds the count value and the abstract of the first target storage unit into the packet head of the data packet according to preset probability;
the method further comprises the steps of: the source end adds the index value to the packet header of the data packet with the probability of 1-the preset probability, or the source end keeps the packet header of the data packet unchanged with the probability of 1-the preset probability.
According to the scheme provided by the embodiment, the count value and the abstract are added to the packet header of the data packet with the preset probability, so that occupied network resources can be reduced. However, the embodiments of the present disclosure are not limited thereto, and the count value and the digest may be added to the packet header of all the data packets.
Fig. 3 is a flowchart of a packet loss detection method according to an embodiment of the present disclosure. As shown in fig. 3, an embodiment of the present disclosure provides a packet loss detection method applied to a destination terminal provided with a second buffer area, where the second buffer area includes at least one storage unit, the storage unit stores a count value, and the packet loss detection method includes:
step 301, after the destination receives the data packet, obtaining an index value from a header of the data packet or generating the index value for the data packet according to a preset generation mode, mapping the data packet to a storage unit of the second buffer area according to the index value in a preset mapping mode, which is called a second target storage unit, and updating a count value of the second target storage unit according to the following mode: adding a count to the existing count value of the second target storage unit to serve as the updated count value of the second target storage unit;
Step 302, when the packet header of the data packet carries a count value of a first target storage unit of the source end, and the count value updated by the second target storage unit is different from the count value of the first target storage unit, determining that there is a packet loss.
According to the packet loss detection method provided by the embodiment, the count value of the destination end is compared with the received count value, and the packet loss is judged to exist when the count values are different, so that the packet loss is detected in time.
In an exemplary embodiment, the storage unit further stores a digest, and the digest is cleared initially;
the method further includes updating the digest of the second target storage unit as follows: performing exclusive OR operation on the existing abstract of the second target storage unit and the identification information of the data packet, and then using the exclusive OR operation as an updated abstract of the second target storage unit;
when judging that the packet loss exists and the count value of the second target storage unit is different from the count value of the first target storage unit by one time of counting, acquiring the abstract of the first target storage unit from the packet head of the data packet, performing exclusive OR operation on the abstract updated by the second target storage unit and the abstract of the first target storage unit, and determining the identification information of the lost data packet according to the obtained result.
According to the scheme provided by the embodiment, the identification information of the lost packet can be recovered through the abstract related to the identification information of the data packet, so that the network fault is rapidly positioned, the network fault is solved, the network reliability is improved, and the user experience is improved.
In an exemplary implementation, the method further comprises: and when judging that the packet loss exists, updating the count value of the second target storage unit by using the count value of the first target storage unit, and updating the abstract of the second target storage unit by using the abstract of the first target storage unit.
In an exemplary implementation, the generating the index value for the data packet according to a preset generation manner includes:
performing hash operation according to the information of the data packet to generate the index value; or determining the index value according to the checksum carried by the data packet. The destination and the source generate the index value in the same manner, and reference may be made to the description of the source, which is not repeated herein.
The technical scheme of the present disclosure is further described below by an exemplary embodiment.
Fig. 4 is a schematic diagram of a packet loss detection method according to an exemplary embodiment. Since the principle of the energy saving operation mode and the normal operation mode is similar, the normal operation mode is only described as an example. And δ=1.0 for convenience. In the figure, the first buffer at the source end has 3 storage units, the index is divided into 0,1,2, the second buffer at the destination end has 3 storage units, and the index is divided into 0,1,2, the count value and abstract of the memory cell are cleared. For any data packet p i The identification information (also called as stream identifier) is f i The object of packet loss detection is to recover the flow identifier of the lost packet. In this embodiment, a counter may be maintained, and the initial value of the count value n of the counter may be 0. As shown in fig. 4, the packet transmission and reception process in this embodiment is as follows:
example 1: data packet p 1 When reaching the source end, the count value n of the counter is 0 and is the data packet p 1 Assigning index value k=n=0, updating n=1, packet p 1 The memory cell mapped to index 0 updates the count value of the memory cell with index 0 to 1 and updates the abstract to f 1 Count value 1, abstract f 1 And an index value k (0 at this time) is added to packet p 1 In the packet header, a data packet p is transmitted 1 . But packet p 1 Is lost during transmission and therefore the state of the buffer of the destination is not changed.
Example 2: data packet p 2 When reaching the source end, data packet p 2 Assigning index value k=n=1, updating n=2, packet p 2 The storage unit mapped with the index 1 updates the count value of the storage unit with the index 1 to 1 and updates the abstract to f 2 Count value 1, abstract f 2 And an index value k (1 in this case) is added to packet p 2 In the packet header, a data packet is transmitted. Data packet p 2 Is lost during transmission as well, so the buffer status of the destination is not changed.
Example 3: data packet p 3 When reaching the source end, it is a data packet p 3 Assigning index value k=n=2, updating n= 3%3 =0, packet p 3 The storage unit mapped with index 2 updates the count value of the storage unit with index 2 by 1 and updates the abstract by f 3 Count value 1, abstract f 3 And an index value k (2 in this case) is added to packet p 3 In the packet header, a data packet p is transmitted 3 . Data packet p 3 After reaching the destination, the destination transmits the data packet p 3 The head of the server analyzes the index value 2, and updates the count value of the storage unit with the destination index of 2 to 1 and the abstract to f 3 . Because of the data packet p 3 The count value (i.e., 1) analyzed by the header is the same as the count value of the storage unit with the index of 2 at the destination, so that no packet loss occurs in the storage unit with the index of 2.
Example 4: data packet p 4 When reaching the source end, it is a data packet p 4 Assigning index value k=n=0, updating n=1, packet p 4 The memory cell mapped with index 0 updates the count value of the memory cell with index 0 to 2, and the abstract to 2 Index value (0 at this time), count value 2 and summary +.>Adding to packet p 4 Is to make the data packet p 4 The data packet is sent to the destination but is lost during transmission so the contents of the destination buffer are not changed.
Example 5: data packet p 5 When reaching the source end, it is a data packet p 5 Assigning index value k=n=1, updating n=2, packet p 5 The storage unit mapped with the index 1 updates the count value of the storage unit with the index 1 to 2, and updates the digest to 2The source terminal adds index value 1, count value 2 and abstract +.>Respectively add to data packet p 5 Header and then packet p 5 And sending the message to the destination. The destination receives the data packet p 5 Later, the packet p is parsed 5 And updates its own storage unit with index 1. The count value and abstract of the storage unit with the index of 1 at the destination end after updating are respectively 1 and f 5 From data packetsp 5 The count value and abstract resolved in (a) are 2 and +.>So the destination knows that there is a packet loss in the storage unit with index 1, the flow identifier of the packet loss is +.>Subsequently, the destination end uses the data packet p 5 The count value and the abstract of the storage unit with the destination index of 1 are respectively updated to 2 and +. >/>
Example 6: data packet p 6 When reaching the source end, it is a data packet p 6 Assigning index value k=n=2, updating n= 3%3 =0, packet p 6 The count value and the abstract of the storage unit with the index of 2 are respectively updated to 2 and 2 in the storage unit with the index of 2 mapped to the first buffer areaThe source end adds index value 2, count value 2 and abstract +.>Adding to packet p 6 Header and packet p 6 And sending the message to the destination. Destination slave data packet p 6 The index value 2, the count value 2 and the abstract +.>And updates the memory location with index 2 in the second buffer. The count value and digest of the updated storage unit with the second buffer index of 2 are 2 and +.>Because the count value of the storage unit with the second buffer index of 2 and the slave packet p 6 Counts resolved from (a)The values are the same so there are no missing data packets in this memory location.
Example 7: data packet p 7 When reaching the source end, it is a data packet p 7 Assigning index value k=n=0, updating n=1, packet p 7 The count value and the abstract of the storage unit with the index of 0 are respectively changed into 3 and 3 after being updated in the storage unit with the index of 0 mapped to the first buffer areaThe source end will index value 0, count value 3 and abstract +.>Adding to packet p 7 Then packet p 7 And sending the message to the destination. The destination receives the data packet p 7 From which index value 0, count value 3 and abstract +.>And then updating the storage unit with the index of 0 in the second buffer area according to the index value of 0. After updating, the count value and the abstract of the storage unit with the index of 0 in the second buffer are respectively 1 and f 7 . Because of the slave packet p 7 The difference between the count value resolved in (a) and the count value of the storage unit with the index of 0 in the second buffer (i.e. 2) is greater than 1, so that it cannot recover the specific packet loss information. The destination end then updates the count value and the abstract of the storage unit with the index of 0 in the second buffer into the data packet p 7 Count value 3 and abstract +.>
As shown in fig. 5, an embodiment of the present disclosure provides a packet loss detection apparatus 70, including a memory 710 and a processor 720, where the memory 710 stores a program, and the program implements the packet loss detection method described in any one of the embodiments when read and executed by the processor 720.
In an exemplary embodiment, the packet loss detection device can be widely deployed at each network location, so that packet loss can be rapidly detected, the location of the packet loss can be judged, and meanwhile, the information of the packet loss can be recovered, and further, the type of network failure and the occurrence location can be judged, thereby improving the reliability of the network.
The embodiment of the disclosure provides a computer readable storage medium storing one or more programs executable by one or more processors to implement the packet loss detection method according to any of the embodiments above.
Those of ordinary skill in the art will appreciate that all or some of the steps, systems, functional modules/units in the apparatus, and methods disclosed above may be implemented as software, firmware, hardware, and suitable combinations thereof. In a hardware implementation, the division between the functional modules/units mentioned in the above description does not necessarily correspond to the division of physical components; for example, one physical component may have multiple functions, or one function or step may be performed cooperatively by several physical components. Some or all of the components may be implemented as software executed by a processor, such as a digital signal processor or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on computer readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). The term computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data, as known to those skilled in the art. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by a computer. Furthermore, as is well known to those of ordinary skill in the art, communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media.
Claims (14)
1. The packet loss detection method is characterized by being applied to a source end provided with a first buffer zone, wherein the first buffer zone comprises at least one storage unit, the storage unit stores a count value, and the packet loss detection method comprises the following steps:
the source end obtains a data packet to be sent, generates an index value for the data packet according to a preset generation mode, maps the data packet to a storage unit of the first buffer area according to the index value in a preset mapping mode, which is called a first target storage unit, and updates the count value of the first target storage unit, wherein the updating mode is as follows: adding a count to the existing count value of the first target storage unit to serve as the updated count value of the first target storage unit;
and the source end adds the count value updated by the first target storage unit to the packet head of the data packet and sends the data packet.
2. The packet loss detection method according to claim 1, wherein the storage unit further stores a digest, and the digest is cleared initially;
before sending the data packet, the method further comprises: updating the digest of the first target storage unit as follows: performing exclusive OR operation on the existing abstract of the first target storage unit and the identification information of the data packet to obtain an updated abstract of the first target storage unit;
And adding the updated abstract of the first target storage unit into the packet head of the data packet.
3. The packet loss detection method according to claim 2, wherein the source end is provided with a counter;
the generating the index value for the data packet according to the preset generation mode includes: generating an index value for the data packet according to the count value of the counter;
after generating an index value for the data packet, further comprising, taking a count value of the counter as a count value of the counter by modulo m after increasing the count value of the counter once, wherein m is the number of storage units of the first buffer;
before sending the data packet, adding the index value to a packet head of the data packet.
4. The method for detecting packet loss according to claim 2, wherein the generating an index value for the data packet according to a preset generation manner includes: performing hash operation according to the information of the data packet to generate the index value; or determining the index value according to the checksum of the data packet.
5. The method of claim 4, wherein the information of the data packet includes at least one first type of information, wherein the first type of information of different data packets of the same data stream is different.
6. The packet loss detection method of claim 5, wherein the first type of information comprises a payload of a data packet.
7. The packet loss detection method according to claim 4, wherein the information of the data packet further includes at least one of: the source address of the data packet, the destination address of the data packet, the transport protocol, the source port number, and the destination port number.
8. The packet loss detection method according to any one of claims 2 to 7, wherein the adding, by the source end, the count value and the digest of the first target storage unit to the header of the data packet includes: the source end adds the count value and the abstract of the first target storage unit into the packet head of the data packet according to preset probability;
the method further comprises the steps of: the source end adds the index value to the packet header of the data packet with the probability of 1-the preset probability, or the source end keeps the packet header of the data packet unchanged with the probability of 1-the preset probability.
9. The packet loss detection method is characterized by being applied to a destination terminal provided with a second buffer zone, wherein the second buffer zone comprises at least one storage unit, the storage unit stores a count value, and the packet loss detection method comprises the following steps:
After the destination receives the data packet, an index value is obtained from a packet header of the data packet or the index value is generated for the data packet according to a preset generation mode, the data packet is mapped to one storage unit of the second buffer area according to the index value in a preset mapping mode, the storage unit is called a second target storage unit, and the count value of the second target storage unit is updated according to the following mode: adding a count to the existing count value of the second target storage unit to serve as the updated count value of the second target storage unit;
and judging that the packet loss exists when the packet head of the data packet carries the count value of the first target storage unit of the source end and the updated count value of the second target storage unit is different from the count value of the first target storage unit.
10. The packet loss detection method according to claim 9, wherein the storage unit further stores a digest, and the digest is cleared initially;
the method further includes updating the digest of the second target storage unit as follows: performing exclusive OR operation on the existing abstract of the second target storage unit and the identification information of the data packet, and then using the exclusive OR operation as an updated abstract of the second target storage unit;
When judging that the packet loss exists and the count value of the second target storage unit is different from the count value of the first target storage unit by one time of counting, acquiring the abstract of the first target storage unit from the packet head of the data packet, performing exclusive OR operation on the abstract updated by the second target storage unit and the abstract of the first target storage unit, and determining the identification information of the lost data packet according to the obtained result.
11. The packet loss detection method according to claim 10, further comprising: and when judging that the packet loss exists, updating the count value of the second target storage unit by using the count value of the first target storage unit, and updating the abstract of the second target storage unit by using the abstract of the first target storage unit.
12. The method of claim 10, wherein the generating the index value for the data packet according to the preset generation manner includes:
performing hash operation according to the information of the data packet to generate the index value; or determining the index value according to the checksum carried by the data packet.
13. A packet loss detection apparatus comprising a memory and a processor, the memory storing a program which, when read and executed by the processor, implements the packet loss detection method according to any one of claims 1 to 12.
14. A computer-readable storage medium storing one or more programs executable by one or more processors to implement the packet loss detection method of any of claims 1-12.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210171121.7A CN114389972B (en) | 2022-02-22 | 2022-02-22 | Packet loss detection method and device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210171121.7A CN114389972B (en) | 2022-02-22 | 2022-02-22 | Packet loss detection method and device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114389972A CN114389972A (en) | 2022-04-22 |
| CN114389972B true CN114389972B (en) | 2024-03-26 |
Family
ID=81206252
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210171121.7A Active CN114389972B (en) | 2022-02-22 | 2022-02-22 | Packet loss detection method and device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114389972B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114979707B (en) * | 2022-05-13 | 2023-12-19 | 深圳智慧林网络科技有限公司 | Data transmission method and device |
| CN114979794B (en) * | 2022-05-13 | 2023-11-14 | 深圳智慧林网络科技有限公司 | Data transmission method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6775293B1 (en) * | 2000-06-30 | 2004-08-10 | Alcatel Canada Inc. | Method and apparatus for monitoring buffer contents in a data communication system |
| CN110661716A (en) * | 2019-09-16 | 2020-01-07 | 锐捷网络股份有限公司 | Network packet loss notification method, monitoring device, switch and storage medium |
| CN111786859A (en) * | 2020-07-17 | 2020-10-16 | 中国联合网络通信集团有限公司 | Packet loss detection method, device and storage medium |
| CN112995059A (en) * | 2019-12-12 | 2021-06-18 | 阿里巴巴集团控股有限公司 | Flow control method, flow control device, terminal equipment and computer storage medium |
| WO2021155637A1 (en) * | 2020-02-07 | 2021-08-12 | 阿里巴巴集团控股有限公司 | Information processing method, device, and system, and storage medium |
-
2022
- 2022-02-22 CN CN202210171121.7A patent/CN114389972B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6775293B1 (en) * | 2000-06-30 | 2004-08-10 | Alcatel Canada Inc. | Method and apparatus for monitoring buffer contents in a data communication system |
| CN110661716A (en) * | 2019-09-16 | 2020-01-07 | 锐捷网络股份有限公司 | Network packet loss notification method, monitoring device, switch and storage medium |
| CN112995059A (en) * | 2019-12-12 | 2021-06-18 | 阿里巴巴集团控股有限公司 | Flow control method, flow control device, terminal equipment and computer storage medium |
| WO2021155637A1 (en) * | 2020-02-07 | 2021-08-12 | 阿里巴巴集团控股有限公司 | Information processing method, device, and system, and storage medium |
| CN111786859A (en) * | 2020-07-17 | 2020-10-16 | 中国联合网络通信集团有限公司 | Packet loss detection method, device and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 一种提高TCP性能的移动自组网E-OAODV协议的实现;邬春学;杨朝梁;;计算机科学(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114389972A (en) | 2022-04-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10972391B2 (en) | Full-path validation in segment routing | |
| CN114389972B (en) | Packet loss detection method and device and storage medium | |
| CN109391560B (en) | Network congestion notification method, proxy node and computer equipment | |
| US20200145331A1 (en) | Using In-Band Operations Data to Signal Packet Processing Departures in a Network | |
| JP4860949B2 (en) | Method and apparatus for providing fast end-to-end failover in a packet switched communication network | |
| AU2015271104B2 (en) | Systems and methods for controlling network switches using a switch modeling interface at a controller | |
| US8619546B2 (en) | Method and apparatus for coping with link failures in central control plane architectures | |
| US9270536B2 (en) | BGP slow peer detection | |
| US9001656B2 (en) | Dynamic route branching system and dynamic route branching method | |
| US9876700B2 (en) | Method and system for active fabric management using unicast reachability monitoring | |
| US20140341226A1 (en) | Routing control system for l3vpn service network | |
| US20130033978A1 (en) | Source Alive Route Injection | |
| CN112311580B (en) | Message transmission path determining method, device and system and computer storage medium | |
| US9425893B1 (en) | Methods and apparatus for implementing optical integrated routing with traffic protection | |
| CN106559246B (en) | Cluster implementation method and server | |
| US10110421B2 (en) | Methods, systems, and computer readable media for using link aggregation group (LAG) status information | |
| US9210067B1 (en) | Method and apparatus for exchanging routing information | |
| US20130250797A1 (en) | Communication control system, control device, communication control method, and communication control program | |
| US8477655B2 (en) | Method, device, and system for establishing label switching path in fast rerouting switching | |
| US8665754B2 (en) | Path generating method, relay device, and computer product | |
| US20230171178A1 (en) | Systems and methods for data plane validation of multiple paths in a network | |
| CN113615132A (en) | Fast flooding topology protection | |
| US11552882B2 (en) | Efficient propagation of fault routing notifications | |
| CN115152192A (en) | PCE controlled network reliability | |
| US20170012869A1 (en) | Forwarding table management in computer networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |