CN115883469A - Data flow load balancing method and device, network topology and data center - Google Patents
Data flow load balancing method and device, network topology and data center Download PDFInfo
- Publication number
- CN115883469A CN115883469A CN202310005574.7A CN202310005574A CN115883469A CN 115883469 A CN115883469 A CN 115883469A CN 202310005574 A CN202310005574 A CN 202310005574A CN 115883469 A CN115883469 A CN 115883469A
- Authority
- CN
- China
- Prior art keywords
- data flow
- hash seed
- hash
- seed value
- data
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 77
- 238000001514 detection method Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 7
- 239000000523 sample Substances 0.000 claims description 6
- 238000007405 data analysis Methods 0.000 claims description 3
- 238000012856 packing Methods 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 3
- 230000010287 polarization Effects 0.000 abstract description 14
- 238000010586 diagram Methods 0.000 description 11
- 238000005516 engineering process Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 235000008694 Humulus lupulus Nutrition 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate
Abstract
The invention relates to a data flow load balancing method, a device, a network topology and a data center, wherein the method comprises the following steps: acquiring data stream switching equipment to be adjusted; setting the hierarchy of the data stream switching equipment to be adjusted as an Nth hierarchy, and acquiring a first Hash seed value of the data stream switching equipment to be adjusted and a first Hash seed value set of the Nth hierarchy; acquiring second hash seed value sets of all data stream switching equipment of the (N-1) th level and/or the (N + 1) th level; judging whether the first hash seed value and the second hash seed value set have the same value or not; and if the number of the Nth-level data stream switching equipment is greater than 1, randomly selecting an adjusting value from the first hash seed value set for updating the first hash seed value. By the technical scheme, the problems that data flow loads in a plurality of data flow paths in the conventional network topology are unbalanced and hash polarization is easy to occur can be solved.
Description
Technical Field
The present invention relates to the field of data flow technologies, and in particular, to a data flow load balancing method and apparatus, a network topology, and a data center.
Background
As shown in fig. 1, in a structure adopted by a network topology of a current data center, multiple paths generally exist between hosts; in order to meet the demand of throughput, a data center provides a large amount of bandwidth resources, and the demands on the reliability and resource utilization rate of network equipment are increasing day by day, wherein data streams are distributed to different paths for data transmission, so that congestion can be avoided, and the resource utilization rate in the data center can be improved.
The common technology is to use ECMP (equal cost multi-path routing), which has the advantages of improving network redundancy and reliability, and improving network resource utilization. When a router finds that multiple optimal paths occur for the same destination address, the routing table is updated, and multiple rules are added to the address for this purpose, corresponding to multiple next hops.
These paths can be used simultaneously to forward data, increasing bandwidth. The hash operation is a path selection strategy of ECMP (equal cost multi-path routing technology), and a next hop routing can be calculated according to a hash factor; common hash operation factors can be classified into the following two types:
1) Five-element group: source address (sip), destination address (dip), source port (port), destination port (dport), and IP protocol number (protocol).
2) Hash seed value: in addition to the conventional quintuple described above, a perturbation factor may be added.
In a multiple Hash scene (namely a Hash operation path selection scene) of the cross-equipment, the phenomenon of uneven load sharing after the common flow passes through Hash for 2 times or more (namely Hash operation path selection) is called Hash polarization (namely Hash polarization) and also called Hash unevenness (namely Hash unevenness); when a specific device of the data center needs to be replaced, the hash polarization phenomenon is easy to occur.
A common solution is to solve the problem of polarization by manually adjusting the hash seed value (i.e., hash seed value) on the device. However, this adjustment method not only excessively depends on the configuration capability of the network administrator, but also may cause a configuration error, which increases the risk of the error and causes a large amount of labor cost to be consumed.
Disclosure of Invention
In order to solve the technical problems, the invention provides a data flow load balancing method, a data flow load balancing device, a network topology and a data center.
In order to achieve the above object, the present invention provides a data stream load balancing method, configured to perform data stream load balancing adjustment on multiple data stream paths in a network topology, where the network topology allocates data streams to each data stream path through a hash operation path selection policy; the balancing method comprises the following steps:
acquiring data flow switching equipment to be adjusted, which needs to adjust the load in the network topology framework, and adjusting the load by adjusting the hash seed value;
setting the hierarchy of the data flow switching equipment to be adjusted in the network topology structure as an Nth hierarchy, and acquiring a first Hash seed value of the data flow switching equipment to be adjusted at the current moment and first Hash seed value sets of all the data flow switching equipment in the Nth hierarchy; wherein N is a natural number;
acquiring a second hash seed value set of all data flow switching equipment of the (N-1) th level and/or the (N + 1) th level in the network topology architecture;
judging whether the first hash seed value and the second hash seed value set have the same value or not;
if the same value exists and the number of the data stream switching devices of the Nth level is greater than 1, randomly selecting an adjusting value from the first hash seed value set for updating the first hash seed value; wherein the adjustment value and the first hash seed value are different.
Further, before the level of the data stream switching device to be adjusted in the network topology is set as the nth level, the balancing method further includes:
and entering all data stream switching equipment in the Nth level through an in-band telemetering detection message, and acquiring all hash seed values of all the data stream switching equipment for sending all the hash seed values to a collector.
Further, entering all data stream switching devices in the nth level through an in-band telemetry detection packet, and acquiring all hash seed values of all data stream switching devices specifically includes:
and sequentially entering all data stream switching equipment in the Nth level through the in-band telemetry detection message, acquiring all hash seed values of all data stream switching equipment, and storing the hash seed values in-band telemetry metadata.
Further, acquiring the to-be-adjusted data stream switching device requiring load adjustment in the network topology architecture specifically includes:
and acquiring the Nth layer from a layer close to a core network in the network topology structure, wherein the Nth layer is used for acquiring the data stream switching equipment to be adjusted in the Nth layer.
Further, the balancing method further comprises:
if the first hash seed value and the second hash seed value set do not have the same value, sequentially performing data flow load balance adjustment on the high-load data flow switching devices in all the levels from the Nth level according to a descending order of the levels.
Further, the balancing method further comprises the following steps:
if the same value exists and the number of the data stream switching devices of the Nth level is 1, randomly selecting an adjusting value from the second hash seed value set for updating the first hash seed value; wherein the adjustment value and the first hash seed value are different.
Further, the balancing method further comprises the following steps:
and judging whether the hash seed values of the data stream switching devices in the same level are the same value and/or whether the hash seed values of the data stream switching devices in different levels are different values, wherein the hash seed values are used as the basis for sequentially adjusting the load balance of the data streams.
Further, acquiring the nth hierarchy from a hierarchy close to a core network in the network topology framework specifically includes:
and acquiring the Nth layer from the layer close to the core network in the network topology architecture, and acquiring a preset type device from the Nth layer as the data stream switching device to be adjusted.
Further, acquiring the to-be-adjusted data stream switching device requiring load adjustment in the network topology architecture specifically includes:
and acquiring the Nth layer from a layer close to the core network in a leaf ridge network topology structure, wherein the Nth layer is used for acquiring the data stream switching equipment to be adjusted in the Nth layer.
Further, acquiring all hash seed values of all data stream switching devices, and storing the hash seed values in-band telemetry metadata, specifically including:
and acquiring all hash seed values and all device IDs of all data stream switching devices in the Nth level and storing the hash seed values and all device IDs in the in-band telemetry metadata.
Further, the balancing method further comprises:
and uniformly acquiring all hash seed values of all levels through the collector, and judging the network operation condition and checking fault points.
Further, the balancing method further comprises the following steps:
acquiring system performance data of each data stream switching device, and performing packing operation according to a preset format for sending out from a specified port;
and receiving and decoding the packed data through the collector, and performing data analysis and data visual display.
Further, the balancing method further comprises the following steps:
acquiring current real flow information of each data stream switching device in the network topology architecture;
and carrying out data flow load balance adjustment on the data flow switching equipment to be adjusted according to the current real flow information and the first Hash seed value.
Further, the balancing method further comprises:
and setting a disturbance factor for the first Hash seed value, generating a first Hash operation factor, and performing data flow load balance adjustment through Hash operation.
The invention also provides a data flow load balancing device, which is used for carrying out data flow load balancing adjustment on a plurality of data flow paths in a network topology framework, wherein the network topology framework distributes data flows for each data flow path through a Hash operation path selection strategy; the balancing device comprises an acquisition unit of the data stream switching equipment to be adjusted and a first hash seed value calculation updating unit;
the data flow switching equipment to be adjusted acquiring unit is used for acquiring the data flow switching equipment to be adjusted needing load adjustment in the network topology framework and adjusting the load by adjusting the hash seed value;
the first hash seed value calculation updating unit is configured to:
setting the hierarchy of the data stream switching equipment to be adjusted in the network topology architecture as an Nth hierarchy, and acquiring a first hash seed value of the data stream switching equipment to be adjusted at the current moment and first hash seed value sets of all the data stream switching equipment in the Nth hierarchy; wherein N is a natural number;
acquiring a second hash seed value set of all data flow switching equipment of the (N-1) th level and/or the (N + 1) th level in the network topology architecture;
judging whether the first hash seed value and the second hash seed value set have the same value or not;
if the same value exists and the number of the data stream switching equipment of the Nth level is more than 1, randomly selecting an adjusting value from the first hash seed value set for updating the first hash seed value; wherein the adjustment value and the first hash seed value are different.
Further, the balancing apparatus further comprises an in-band telemetry management unit for:
and entering all data stream switching equipment in the Nth level through an in-band telemetering detection message, and acquiring all hash seed values of all the data stream switching equipment for sending all the hash seed values to a collector.
The present invention further provides a computer device comprising a memory, a processor and a computer program, the computer program being stored on the memory and being executable on the processor, the processor implementing the following steps when executing the computer program:
acquiring data flow switching equipment to be adjusted, which needs to adjust the load in the network topology framework, and adjusting the load by adjusting the hash seed value;
setting the hierarchy of the data stream switching equipment to be adjusted in the network topology architecture as an Nth hierarchy, and acquiring a first hash seed value of the data stream switching equipment to be adjusted at the current moment and first hash seed value sets of all the data stream switching equipment in the Nth hierarchy; wherein N is a natural number;
acquiring a second hash seed value set of all data flow switching equipment of an N-1 level and/or an N +1 level in the network topology architecture;
judging whether the first hash seed value and the second hash seed value set have the same value or not;
if the same value exists and the number of the data stream switching devices of the Nth level is greater than 1, randomly selecting an adjusting value from the first hash seed value set for updating the first hash seed value; wherein the adjustment value and the first hash seed value are different.
The present invention further provides a computer-readable storage medium storing a computer program which, when executed by a processor, performs the steps of:
acquiring data flow switching equipment to be adjusted, which needs to adjust the load in the network topology framework, and adjusting the load by adjusting the hash seed value;
setting the hierarchy of the data flow switching equipment to be adjusted in the network topology structure as an Nth hierarchy, and acquiring a first Hash seed value of the data flow switching equipment to be adjusted at the current moment and first Hash seed value sets of all the data flow switching equipment in the Nth hierarchy; wherein N is a natural number;
acquiring a second hash seed value set of all data flow switching equipment of an N-1 level and/or an N +1 level in the network topology architecture;
judging whether the first hash seed value and the second hash seed value set have the same value or not;
if the same value exists and the number of the data stream switching devices of the Nth level is greater than 1, randomly selecting an adjusting value from the first hash seed value set for updating the first hash seed value; wherein the adjustment value and the first hash seed value are different.
Compared with the prior art, the technical scheme of the invention has the following technical effects:
the data flow load balancing method comprises the steps of obtaining the data flow exchange equipment to be adjusted and adjusting the hash seed value of the data flow exchange equipment to be adjusted;
firstly, acquiring data stream switching equipment to be adjusted, which needs to adjust load in a network topology framework, and adjusting the load by adjusting a hash seed value;
then, obtaining an update value or an adjustment value of the data stream switching device to be adjusted, specifically including:
setting the hierarchy of the data stream switching equipment to be adjusted in a network topology framework as an Nth hierarchy, and acquiring a first Hash seed value of the data stream switching equipment to be adjusted at the current moment and first Hash seed value sets of all the data stream switching equipment in the Nth hierarchy; wherein N is a natural number;
acquiring second hash seed value sets of all data flow switching equipment of the (N-1) th level and/or the (N + 1) th level in the network topology architecture;
then judging whether the first hash seed value and the second hash seed value set have the same value or not;
if the same value exists and the number of the data stream switching equipment of the Nth level is more than 1, randomly selecting an adjusting value from the first Hash seed value set for updating the first Hash seed value; wherein, the adjusting value and the first hash seed value are different;
therefore, according to the load balance adjustment rule, finding out equipment needing to adjust the hash seed value, calculating the hash seed value needing to be adjusted, and updating the hash seed value of the data stream switching equipment to be adjusted, thereby realizing data stream load balance adjustment on the data stream switching equipment to be adjusted;
that is, for the situation that multiple paths usually exist between hosts in the structure adopted by the network topology of the data center at present, the balancing method can solve the phenomenon of uneven load sharing possibly caused by multiple Hash scenes (i.e. Hash operation path selection scenes);
meanwhile, the excessive labor cost consumption and the risk of wrong configuration of network management personnel can be reduced.
Therefore, the balancing method can avoid the occurrence of Haichi polarization in a network environment, distribute the data flow to different paths for data transmission, avoid congestion and improve the resource utilization rate in the data center.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic diagram of data flow load balancing in a prior art network topology;
fig. 2 is a schematic diagram illustrating the flow direction distribution of data flow through device a in a network topology in the prior art;
fig. 3 is a schematic flow diagram of a data flow after a hash seed value of a device C is changed in a network topology in the prior art;
FIG. 4 is a flowchart illustrating a data flow load balancing method according to an embodiment of the present invention;
FIG. 5 is a diagram illustrating that an in-band telemetry probe packet obtains all hash seed values in an actual embodiment of the present invention;
FIG. 6 is a schematic diagram of the current network environment architecture in the practice of the present invention;
FIG. 7 is a diagram illustrating the states of hash seed values of devices in a network in accordance with an embodiment of the present invention;
fig. 8 is a block diagram of a data flow load balancing apparatus according to a second embodiment of the present invention;
fig. 9 is an internal structural diagram of a computer device according to a second embodiment of the present invention.
Detailed Description
As shown in fig. 2, in the prior art, the traffic passing through the device a is divided into two sides according to the result of the hash operation, so as to achieve a result of load balancing; however, the hash operation result in device C is the same as that in a, so that traffic passing through device C only goes to device E, thereby wasting the path from device C to device F.
As shown in fig. 3, a common solution is to solve the polarization problem by manually adjusting the Hash Seed value (Hash Seed) on the device, such as changing the Hash Seed value of device C.
However, this adjustment method not only excessively depends on the configuration capability of the network administrator, but also may cause a configuration error, which increases the risk of the error and causes a large amount of labor cost to be consumed.
Therefore, the invention provides a data flow load balancing method, a data flow load balancing device, a network topology and a data center, which are used for solving the problems.
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The first embodiment is as follows:
as shown in fig. 4, an embodiment of the present invention provides a data flow load balancing method, configured to perform data flow load balancing adjustment on multiple data flow paths in a network topology, where the network topology allocates a data flow to each data flow path through a hash operation path selection policy; the balancing method comprises the following steps:
s2, acquiring data stream switching equipment to be adjusted, which needs to adjust the load, in the network topology framework, and adjusting the load by adjusting the Hash seed value;
s31, setting the hierarchy of the data flow switching equipment to be adjusted in the network topology structure as an Nth hierarchy, and acquiring a first Hash seed value of the data flow switching equipment to be adjusted at the current moment and first Hash seed value sets of all the data flow switching equipment in the Nth hierarchy; wherein N is a natural number;
s32, acquiring second hash seed value sets of all data flow switching equipment of the (N-1) th level and/or the (N + 1) th level in the network topology architecture;
s33, judging whether the first hash seed value and the second hash seed value set have the same value or not;
s331, if the same value exists and the number of the data stream switching devices of the Nth level is greater than 1, randomly selecting an adjustment value from the first hash seed value set for updating the first hash seed value; wherein, the adjusting value and the first hash seed value are different.
In a specific embodiment, the data flow load balancing method includes a process of acquiring a data flow switching device to be adjusted, and a process of adjusting a hash seed value of the data flow switching device to be adjusted;
firstly, acquiring data stream switching equipment to be adjusted, which needs to adjust load in a network topology framework, and adjusting the load by adjusting a hash seed value;
then, obtaining an update value or an adjustment value of the data stream switching device to be adjusted, specifically including:
setting the hierarchy of the data stream switching equipment to be adjusted in a network topology framework as an Nth hierarchy, and acquiring a first Hash seed value of the data stream switching equipment to be adjusted at the current moment and first Hash seed value sets of all the data stream switching equipment in the Nth hierarchy; wherein N is a natural number;
acquiring a second hash seed value set of all data flow switching equipment of the (N-1) th level and/or the (N + 1) th level in the network topology architecture;
then judging whether the first hash seed value and the second hash seed value set have the same value or not;
if the same value exists and the number of the data stream switching equipment of the Nth level is more than 1, randomly selecting an adjusting value from the first hash seed value set for updating the first hash seed value; wherein, the adjusting value and the first hash seed value are different;
therefore, according to the load balance adjustment rule, finding out equipment needing adjusting the hash seed value, calculating the hash seed value needing adjusting, and updating the hash seed value of the data flow switching equipment to be adjusted, thereby realizing the load balance adjustment of the data flow switching equipment to be adjusted;
that is, for the situation that multiple paths usually exist between hosts in the structure adopted by the network topology of the existing data center, the balancing method can solve the phenomenon of uneven load sharing possibly caused by multiple Hash scenes (i.e. Hash operation path selection scenes);
meanwhile, the excessive labor cost consumption and the risk of configuration errors of network management personnel can be reduced.
Therefore, the balancing method can avoid the occurrence of Haichi polarization in a network environment, distribute the data flow to different paths for data transmission, avoid congestion and improve the resource utilization rate in the data center.
In practice, ECMP, equal-cost multi-path routing, represents an Equal-cost multi-path route; INT, in-band Network telemetric, stands for In-band Telemetry; SIP, source IP address, represents the Source address; DIP, namely Destination IP address, represents a target address; SPORT, source Port, represents a Source Port; DPORT, destination Port, represents the target Port.
In a preferred embodiment, S2 specifically includes:
and acquiring an Nth layer from a layer close to the Core network in the Spine network topology architecture of the Spine/Leaf network, and acquiring the data flow switching equipment to be adjusted in the Nth layer.
In an actual embodiment, the collector finds out a device that needs to adjust the hash seed value and calculates the hash seed value that needs to be adjusted according to the load balance adjustment rule: the specific process is as follows:
step 1) in a Spine/Leaf topological structure, sequentially comparing hash seed values from Core network Layer N (namely, nth level); n is a natural number;
step 2) comparing the hash seed value sets of the layers before and after, namely the hash seed value sets of Layer N-1 (namely the N-1 level) and Layer N +1 (namely the N +1 level);
step 3) according to the result of step 2), if Layer N has a repeated hash seed value, the following two possible situations are included:
a) If the Layer N has other equipment, randomly selecting a value according to the hash seed value set of the Layer N, and adjusting the hash seed value;
b) If no other equipment exists in the Layer N, randomly selecting a non-repeated value according to the hash seed value set of the Layer N-1 and the Layer N +1, and adjusting the hash seed value;
and 4) returning to the step 1) if the Layer N has no repeated hash seed value according to the result of the step 2), and sequentially judging the Layer N-1.
In a preferred embodiment, before S31, the balancing method further comprises:
s1, entering all data stream switching equipment in the Nth level through an INT in-band telemetering detection message, acquiring all hash seed values of all the data stream switching equipment, and sending all the hash seed values to a collector.
In a preferred embodiment, S1 specifically comprises:
and S11, sequentially entering all data stream switching equipment in the Nth level through the INT in-band telemetering detection message, acquiring all hash seed values of all the data stream switching equipment, and storing the hash seed values in INT in-band telemetering metadata.
In a specific embodiment, an INT probe message (i.e., an in-band telemetry probe message) enters the switch, and the hash seed value of the switch is placed in INT Metadata data (i.e., INT Metadata), so that the collector can obtain the hash seed values of all devices.
As shown in fig. 5, in a practical embodiment, an INT message (i.e., an in-band telemetry probe message) enters a certain switch, and its hash seed value is placed in INT Metadata data (i.e., INT Metadata); finally, the INT message passes through all the switch devices, so that the acquirer can acquire hash seed values of all the devices.
In a preferred embodiment, S2 specifically includes:
s21 obtains an nth layer from a layer close to the Core network in the network topology structure, and is configured to obtain a data stream switching device to be adjusted in the nth layer.
In an actual embodiment, in a Spine/Leaf topology (i.e., a Leaf-Spine network topology), the collector may preferentially change the hash seed value for a Core network device closer to the Core network device according to the load balance adjustment rule, so as to ensure that the current network impact is minimized when the configuration of the device is changed.
In a preferred embodiment, the balancing method further comprises:
and S34, if the first hash seed value and the second hash seed value set do not have the same value, sequentially carrying out data flow load balance adjustment on the high-load data flow switching devices in all the levels from the Nth level according to the descending order of the levels.
In the practical embodiment, if Layer N has no repeated hash seed value, layer N-1 is determined in sequence from the beginning, so as to perform load balance adjustment Layer by Layer.
In a preferred embodiment, the balancing method further comprises:
s332, if the same numerical value exists and the number of the data stream switching devices at the Nth level is 1, randomly selecting an adjusting value from the second hash seed value set for updating the first hash seed value; wherein, the adjusting value and the first hash seed value are different.
In an actual embodiment, if Layer N has no other device, a non-repeating value is randomly selected according to the hash seed value set of Layer N-1 and Layer N +1, and the hash seed value is adjusted.
In a preferred embodiment, the balancing method further comprises:
and judging whether the hash seed values of the data stream switching devices in the same level are the same value and/or whether the hash seed values of the data stream switching devices in different levels are different values, wherein the hash seed values are used as the basis for sequentially adjusting the load balance of the data streams.
In an actual embodiment, after the final adjustment, the hash seed values may be the same among devices in the same hierarchy, and the hash seed values in different hierarchies are not repeated, so that the occurrence of hash polarization is avoided as much as possible.
In a preferred embodiment, S21 specifically includes:
s211 obtains an nth layer from a layer close to the Core network in the network topology structure, and obtains a preset type device from the nth layer as the to-be-adjusted data stream switching device.
In addition, pick from the closest core network, where each pick picks only a specific device to adjust so that the impact on the overall network environment is minimized.
In a preferred embodiment, S11 specifically includes:
and acquiring all hash seed values and all device IDs of all data stream switching devices in the Nth level, and storing the hash seed values and all device IDs in INT in-band telemetry metadata.
In a practical embodiment, the INT Metadata contains at least the following:
1) Switch ID
2) Hash seed value
In a preferred embodiment, the balancing method further comprises:
and uniformly acquiring all hash seed values of all levels through the collector, and judging the network operation condition and checking a fault point.
In the prior art, the network topology scale of the existing data center is getting larger and larger, and network management personnel are difficult to find the network problem of the data center in a traditional mode.
In the practical embodiment of the invention, in the novel network state monitoring mode, the INT (i.e. in-band telemetry) collects the network information returned by each switch or router through a centralized collector, and accordingly judges the network operation condition and checks whether a fault point or congestion exists.
In a preferred embodiment, the balancing method further comprises:
acquiring system performance data of each data stream switching device, and performing packing operation according to a preset format for sending out from a designated port;
and receiving and decoding the packed data through the collector, and performing data analysis and data visual display.
In practical embodiments, telemetrology (i.e., telemetry) is a technique that collects data remotely from a physical or virtual device at high speed; the system performance data collected by the system performance management module is packed, coded and serialized according to a specific format, and then is sent out from a specific port; and the collector back-end system receives and decodes the data, and analyzes and visually displays the data.
In a preferred embodiment, the balancing method further comprises:
acquiring current real flow information of each data stream switching device in a network topology architecture;
and carrying out data flow load balance adjustment on the data flow switching equipment to be adjusted according to the current real flow information and the first Hash seed value.
In practice, the above balancing method is applied to a switch software platform SONiC, and the hash seed value of the packet-clamp switch is detected and packaged and then sent to a collector, and finally the collector selects specific equipment to adjust according to the calculation rule provided by the present case;
in the balancing method, real flow information and more detailed and helpful information can be transmitted to the collector, so that not only can the hash seed value be adjusted, but also the hash algorithm can be adjusted, thereby ensuring that the network environment load is more uniform and the utilization rate is maximized.
In a preferred embodiment, the balancing method further comprises:
and setting a disturbance factor for the first Hash seed value, generating a first Hash operation factor, and performing data flow load balance adjustment through Hash operation.
In an actual embodiment, in a plurality of data flow paths in a network topology architecture, the hash operation is a path selection policy of ECMP (equal cost multi-path routing technology), and a next hop routing can be calculated according to a hash factor; common hash operation factors can be classified into the following two types:
1) Five-element group: source address (sip), destination address (dip), source port (port), destination port (dport), and IP protocol number (protocol).
2) Hash seed value: in addition to the conventional quintuple described above, a perturbation factor may be added.
In an actual embodiment, a specific flow of the data flow load balancing method is as follows:
(1) As shown in fig. 5, firstly, an INT message (i.e., an in-band telemetry probe message) enters a certain switch, and a hash seed value thereof is placed in INT Metadata data (i.e., INT Metadata);
finally, the INT message passes through all the switch devices, so that the collector can obtain the hash seed values of all the devices.
Wherein INT Metadata contains the following:
1) Switch ID
2) Hash seed value
(2) The collector finds out the equipment needing to adjust the hash seed value (namely the hash seed value) and calculates the hash seed value needing to be adjusted according to the following load balance adjustment rules:
step 1) in a Spine/Leaf topological structure, sequentially comparing hash seed values from Core network Layer N (namely the Nth level);
step 2) comparing the hash seed value sets of the previous Layer and the next Layer, namely the hash seed value sets of the Layer N-1 (namely the N-1 level) and the Layer N +1 (namely the N +1 level);
step 3) according to the result of step 2), if Layer N has a repeated Hash seed value, the following two possible situations are included:
a) If the Layer N has other equipment, randomly selecting a value according to the hash seed value set of the Layer N, and adjusting the hash seed value;
b) If no other equipment exists in the Layer N, randomly selecting a non-repeated value according to the Hash seed value set of the Layer N-1 and the Layer N +1, and adjusting the Hash seed value;
and 4) returning to the step 1) if the Layer N has no repeated hash seed value according to the result of the step 2), and sequentially judging the Layer N-1.
In an actual application scenario, the following are exemplified:
fig. 6 is a schematic diagram of a current network environment architecture, and fig. 7 is a schematic diagram of a hash seed value state of a device in a network. The hash value set for each level is shown in the following table:
step 1:
starting from Layer 2, acquiring that the hash value sets of upper and lower layers adjacent to Layer 2 are {3,4,7}, wherein the hash seed value of the device G required to be adjusted by Layer 2 is 3; in order to avoid the hash polarization phenomenon, a hash seed value is selected from Layer 2 according to the load balance adjustment rule. At this point, the hash seed value of device G may be adjusted to 5, as detailed in the following table:
step 2:
the hash value sets of Layer 2 and Layer 0 obtained for Layer 1 are {3,5,2}, and the hash seed value of the device D to be adjusted for Layer 1 is 3; in order to avoid the hash polarization phenomenon, the hash seed value of the device D is adjusted to 4 according to the load balance adjustment rule.
The final adjustment result is detailed in the following table, the hash seed values of the devices in the same level can be the same, and the hash seed values in different levels are not repeated, so that the hash polarization is avoided as much as possible;
in addition, picking is started from the closest core network, wherein each picking only picks a specific device to adjust, so that the influence on the overall network environment is minimized.
To sum up, the load balancing method for data stream provided by the embodiment of the present invention, based on the load balancing hash seed algorithm of INT, implements the following functions by INT (i.e. in-band telemetry technology):
the collector collects the hash seed information returned by each switch or router without consuming excessive labor cost, and judges whether the hash seed value of a specific switch needs to be changed or not according to the hash seed information;
the specific functions are as follows:
1) The INT detection message (in-band telemetering detection message) enters the switch, and the hash seed value of the switch is placed in INT Metadata data (INT Metadata), so that the collector can acquire the hash seed values of all devices.
2) In the Spine/Leaf topology architecture, the collector preferentially changes the hash seed value for the Core network device closer to the Core according to the rules provided by the scheme, so that the current network influence is ensured to be minimized when the configuration of the device is changed.
3) Hash polarization in a network environment is avoided, data flow is distributed to different paths for data transmission, congestion is avoided, and the resource utilization rate in a data center is improved.
The main advantages of the above data flow load balancing method are as follows:
1) The problem of Hash polarization is solved, unnecessary labor cost expenses can be saved, and unnecessary configuration error risks can be avoided.
2) Choosing a particular device adjusts the hash seed value such that the current network impact is minimized.
3) The network basic equipment does not need to be upgraded additionally, and the influence of the equipment capacity is not limited.
It should be noted that, although the steps in the flowchart are shown in sequence as indicated by the arrows, the steps are not necessarily executed in sequence as indicated by the arrows. The steps are not limited to being performed in the exact order illustrated and, unless explicitly stated herein, may be performed in other orders. Moreover, at least a portion of the steps in the flowchart may include multiple sub-steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
The second embodiment:
as shown in fig. 8, an embodiment of the present invention further provides a data flow load balancing apparatus, configured to perform data flow load balancing adjustment on multiple data flow paths in a network topology, where the network topology allocates a data flow to each data flow path through a hash operation path selection policy; the balancing device comprises an acquisition unit of the data flow switching equipment to be adjusted and a first hash seed value calculation updating unit;
the data flow switching equipment to be adjusted acquiring unit is used for acquiring the data flow switching equipment to be adjusted needing load adjustment in the network topology framework and adjusting the load by adjusting the Hash seed value;
a first hash seed value calculation updating unit configured to:
setting the hierarchy of the data stream switching equipment to be adjusted in a network topology structure as an Nth hierarchy, and acquiring a first hash seed value of the data stream switching equipment to be adjusted at the current moment and first hash seed value sets of all the data stream switching equipment in the Nth hierarchy; wherein N is a natural number;
acquiring a second hash seed value set of all data flow switching equipment of an N-1 level and/or an N +1 level in a network topology architecture;
judging whether the first hash seed value and the second hash seed value set have the same value or not;
if the same value exists and the number of the data stream switching equipment of the Nth level is more than 1, randomly selecting an adjusting value from the first Hash seed value set for updating the first Hash seed value; wherein, the adjusting value and the first hash seed value are different.
In a preferred embodiment, the balancing apparatus further comprises an INT in-band telemetry management unit for:
and entering all data stream switching equipment in the Nth level through the INT in-band telemetering detection message, and acquiring all hash seed values of all the data stream switching equipment for sending all the hash seed values to the collector.
For the specific limitations of the above apparatus, reference may be made to the limitations of the method above, which are not described herein again.
The modules in the above devices may be wholly or partially implemented by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, or can be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
As shown in fig. 9, the computer device may be a terminal including a processor, a memory, a network interface, a display screen, and an input device, which are connected through a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
It will be appreciated that the arrangements shown in the above figures are merely block diagrams of some of the arrangements relevant to the inventive arrangements and do not constitute limitations on the computer apparatus to which the inventive arrangements may be applied, as a particular computer apparatus may comprise more or less components than those shown in the figures, or some of the components may be combined, or have a different arrangement of components.
All or part of the processes of the methods of the embodiments can be implemented by a computer program that can be stored in a non-volatile computer-readable storage medium and that, when executed, can include the processes of the embodiments of the methods.
Any reference to memory, storage, databases, or other media used in embodiments provided herein may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), rambus (Rambus) direct RAM (RDRAM), direct Rambus Dynamic RAM (DRDRAM), and Rambus Dynamic RAM (RDRAM), among others.
It should be noted that the above-mentioned embodiments are only preferred embodiments of the present invention and the technical principles applied. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (20)
1. The data flow load balancing method is characterized by being used for carrying out data flow load balancing adjustment on a plurality of data flow paths in a network topology framework, wherein the network topology framework distributes data flows to each data flow path through a Hash operation path selection strategy; the balancing method comprises the following steps:
acquiring data flow switching equipment to be adjusted, which needs to adjust the load in the network topology framework, and adjusting the load by adjusting the hash seed value;
setting the hierarchy of the data stream switching equipment to be adjusted in the network topology architecture as an Nth hierarchy, and acquiring a first hash seed value of the data stream switching equipment to be adjusted at the current moment and first hash seed value sets of all the data stream switching equipment in the Nth hierarchy; wherein N is a natural number;
acquiring a second hash seed value set of all data flow switching equipment of an N-1 level and/or an N +1 level in the network topology architecture;
judging whether the first hash seed value and the second hash seed value set have the same value or not;
if the same value exists and the number of the data stream switching equipment of the Nth level is more than 1, randomly selecting an adjusting value from the first hash seed value set for updating the first hash seed value; wherein the adjustment value and the first hash seed value are different.
2. The data flow load balancing method according to claim 1, wherein before setting the hierarchy of the data flow switching device to be adjusted in the network topology as an nth hierarchy, the balancing method further comprises:
and entering all data stream switching equipment in the Nth level through an in-band telemetering detection message, and acquiring all hash seed values of all the data stream switching equipment for sending all the hash seed values to a collector.
3. The data flow load balancing method according to claim 2, wherein the step of entering all data flow switching devices in the nth layer through an in-band telemetry probe packet and acquiring all hash seed values of all data flow switching devices specifically includes:
and sequentially entering all data stream switching equipment in the Nth level through the in-band telemetry detection message, acquiring all hash seed values of all the data stream switching equipment, and storing the hash seed values in-band telemetry metadata.
4. The method for load balancing of data flows according to claim 1, wherein obtaining the data flow switching device to be adjusted, whose load needs to be adjusted, in the network topology framework specifically includes:
and acquiring the Nth layer from a layer close to a core network in the network topology structure, wherein the Nth layer is used for acquiring the data stream switching equipment to be adjusted in the Nth layer.
5. The data flow load balancing method of claim 4, wherein the balancing method further comprises:
and if the first hash seed value and the second hash seed value set do not have the same value, sequentially carrying out data flow load balance adjustment on the high-load data flow switching equipment in all the levels from the Nth level according to a descending order of the levels.
6. The data flow load balancing method of claim 5, wherein the balancing method further comprises:
if the same value exists and the number of the data stream switching devices of the Nth level is 1, randomly selecting an adjusting value from the second hash seed value set for updating the first hash seed value; wherein the adjustment value and the first hash seed value are different.
7. The data flow load balancing method of claim 6, wherein the balancing method further comprises:
and judging whether the hash seed values of the data stream switching devices in the same level are the same value and/or whether the hash seed values of the data stream switching devices in different levels are different values, wherein the hash seed values are used as the basis for sequentially adjusting the load balance of the data streams.
8. The data flow load balancing method according to claim 6, wherein the obtaining the nth tier from a tier close to a core network in the network topology architecture specifically includes:
and acquiring the Nth layer from a layer close to a core network in the network topology architecture, and acquiring preset type equipment from the Nth layer as the data stream switching equipment to be adjusted.
9. The data flow load balancing method according to any one of claims 4 to 8, wherein acquiring the data flow switching device to be adjusted, whose load needs to be adjusted, in the network topology framework specifically includes:
and acquiring the Nth layer from a layer close to the core network in a leaf ridge network topology structure, wherein the Nth layer is used for acquiring the data stream switching equipment to be adjusted in the Nth layer.
10. The data flow load balancing method according to claim 3, wherein obtaining all hash seed values of all data flow switching devices and storing the hash seed values in the in-band telemetry metadata specifically comprises:
and acquiring all hash seed values and all device IDs of all data stream switching devices in the Nth level and storing the hash seed values and all device IDs in the in-band telemetry metadata.
11. The data flow load balancing method of claim 2, wherein the balancing method further comprises:
and uniformly acquiring all hash seed values of all levels through the collector, and judging the network operation condition and checking fault points.
12. The data flow load balancing method of claim 11, wherein the balancing method further comprises:
acquiring system performance data of each data stream switching device, and performing packing operation according to a preset format for sending out from a designated port;
and receiving and decoding the packed data through the collector, and performing data analysis and data visual display.
13. The data flow load balancing method of claim 2, wherein the balancing method further comprises:
acquiring current real flow information of each data stream switching device in the network topology architecture;
and carrying out data flow load balance adjustment on the data flow switching equipment to be adjusted according to the current real flow information and the first Hash seed value.
14. The data flow load balancing method of claim 13, wherein the balancing method further comprises:
and setting a disturbance factor for the first Hash seed value, generating a first Hash operation factor, and performing data flow load balance adjustment through Hash operation.
15. The data flow load balancing device is used for carrying out data flow load balancing adjustment on a plurality of data flow paths in a network topology structure, wherein the network topology structure distributes data flows for each data flow path through a Hash operation path selection strategy; the balancing device comprises an acquisition unit of the data stream switching equipment to be adjusted and a first hash seed value calculation updating unit;
the data flow switching equipment to be adjusted acquiring unit is used for acquiring the data flow switching equipment to be adjusted needing load adjustment in the network topology framework and adjusting the load by adjusting the hash seed value;
the first hash seed value calculation updating unit is configured to:
setting the hierarchy of the data flow switching equipment to be adjusted in the network topology structure as an Nth hierarchy, and acquiring a first Hash seed value of the data flow switching equipment to be adjusted at the current moment and first Hash seed value sets of all the data flow switching equipment in the Nth hierarchy; wherein N is a natural number;
acquiring a second hash seed value set of all data flow switching equipment of an N-1 level and/or an N +1 level in the network topology architecture;
judging whether the first hash seed value and the second hash seed value set have the same value or not;
if the same value exists and the number of the data stream switching equipment of the Nth level is more than 1, randomly selecting an adjusting value from the first hash seed value set for updating the first hash seed value; wherein the adjustment value and the first hash seed value are different.
16. The data flow load balancing apparatus of claim 15, wherein the balancing apparatus further comprises an in-band telemetry management unit configured to:
and entering all data stream switching equipment in the Nth level through an in-band telemetering detection message, and acquiring all hash seed values of all the data stream switching equipment for sending all the hash seed values to a collector.
17. A computer arrangement comprising a memory, a processor and a computer program, the computer program being stored on the memory and being executable on the processor, characterized in that the processor, when executing the computer program, is adapted to carry out the steps of the data flow load balancing method according to any of the claims 1-14.
18. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the data flow load balancing method according to any one of claims 1 to 14.
19. A network topology comprising a plurality of data flow paths, a plurality of data flow switching devices, and a data flow load balancing module for data flow load balancing adjustments to the data flow switching devices in the plurality of data flow paths by the data flow load balancing method of any one of claims 1-14.
20. A data center comprising a network topology, a plurality of data flow paths, a plurality of data flow switching devices, and a data flow load balancing module configured to perform data flow load balancing adjustments for data flow switching devices in the plurality of data flow paths by the data flow load balancing method of any one of claims 1-14.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310005574.7A CN115883469B (en) | 2023-01-04 | 2023-01-04 | Data stream load balancing method and device and data center |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310005574.7A CN115883469B (en) | 2023-01-04 | 2023-01-04 | Data stream load balancing method and device and data center |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115883469A true CN115883469A (en) | 2023-03-31 |
| CN115883469B CN115883469B (en) | 2023-06-02 |
Family
ID=85758010
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310005574.7A Active CN115883469B (en) | 2023-01-04 | 2023-01-04 | Data stream load balancing method and device and data center |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115883469B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116192636A (en) * | 2023-04-27 | 2023-05-30 | 苏州浪潮智能科技有限公司 | Network device hash group configuration method and device, electronic device and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102571558A (en) * | 2010-11-30 | 2012-07-11 | 马维尔以色列(M.I.S.L.)有限公司 | Load balancing hash computation for network switches |
| US20150078375A1 (en) * | 2013-09-13 | 2015-03-19 | Broadcom Corporation | Mutable Hash for Network Hash Polarization |
| CN109617986A (en) * | 2018-12-27 | 2019-04-12 | 华为技术有限公司 | A kind of load-balancing method and the network equipment |
| CN111010348A (en) * | 2019-12-30 | 2020-04-14 | 苏州浪潮智能科技有限公司 | Link resource load balancing method, device, equipment and storage medium |
| CN113132249A (en) * | 2019-12-31 | 2021-07-16 | 华为技术有限公司 | Load balancing method and equipment |
| CN113965518A (en) * | 2020-07-03 | 2022-01-21 | 华为技术有限公司 | Message processing method and device |
| CN114006863A (en) * | 2021-11-02 | 2022-02-01 | 北京科东电力控制系统有限责任公司 | Multi-core load balancing cooperative processing method and device and storage medium |
| CN114070797A (en) * | 2021-09-29 | 2022-02-18 | 新华三信息安全技术有限公司 | Method and system for controlling data transmission |
| CN114121124A (en) * | 2020-08-26 | 2022-03-01 | 美光科技公司 | Apparatus, system, and method for updating hash keys in memory |
-
2023
- 2023-01-04 CN CN202310005574.7A patent/CN115883469B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102571558A (en) * | 2010-11-30 | 2012-07-11 | 马维尔以色列(M.I.S.L.)有限公司 | Load balancing hash computation for network switches |
| US20150078375A1 (en) * | 2013-09-13 | 2015-03-19 | Broadcom Corporation | Mutable Hash for Network Hash Polarization |
| CN109617986A (en) * | 2018-12-27 | 2019-04-12 | 华为技术有限公司 | A kind of load-balancing method and the network equipment |
| CN111010348A (en) * | 2019-12-30 | 2020-04-14 | 苏州浪潮智能科技有限公司 | Link resource load balancing method, device, equipment and storage medium |
| CN113132249A (en) * | 2019-12-31 | 2021-07-16 | 华为技术有限公司 | Load balancing method and equipment |
| CN113965518A (en) * | 2020-07-03 | 2022-01-21 | 华为技术有限公司 | Message processing method and device |
| CN114121124A (en) * | 2020-08-26 | 2022-03-01 | 美光科技公司 | Apparatus, system, and method for updating hash keys in memory |
| CN114070797A (en) * | 2021-09-29 | 2022-02-18 | 新华三信息安全技术有限公司 | Method and system for controlling data transmission |
| CN114006863A (en) * | 2021-11-02 | 2022-02-01 | 北京科东电力控制系统有限责任公司 | Multi-core load balancing cooperative processing method and device and storage medium |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116192636A (en) * | 2023-04-27 | 2023-05-30 | 苏州浪潮智能科技有限公司 | Network device hash group configuration method and device, electronic device and storage medium |
| CN116192636B (en) * | 2023-04-27 | 2023-08-15 | 苏州浪潮智能科技有限公司 | Network device hash group configuration method and device, electronic device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115883469B (en) | 2023-06-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106464589B (en) | Method and apparatus for partial software defined network switch replacement in IP networks | |
| CN109495300B (en) | Reliable SDN virtual network mapping method | |
| CN112738820B (en) | Dynamic deployment method and device of service function chain and computer equipment | |
| US10374956B1 (en) | Managing a hierarchical network | |
| CN105379196B (en) | Method, system and computer storage medium for the routing of fault-tolerant and load balance | |
| Hu et al. | Dynamic slave controller assignment for enhancing control plane robustness in software-defined networks | |
| Zheng et al. | A heuristic survivable virtual network mapping algorithm | |
| CN104144135B (en) | Resource allocation methods for multicast virtual network and anti-ruin resource allocation methods | |
| CN112118176B (en) | Service reliability-oriented routing load optimization method for channel of integrated data network | |
| CN108156041B (en) | Differentiated virtual optical network mapping method based on security perception | |
| CN106549779A (en) | A kind of maximum non-intersect double route collocation method of multiple constraint energy communication service | |
| CN112653575B (en) | Network configuration method, controller and flow guiding system | |
| CN115883469A (en) | Data flow load balancing method and device, network topology and data center | |
| CN106130920A (en) | A kind of message forwarding method and device | |
| CN112350949A (en) | Rerouting congestion control method and system based on flow scheduling in software defined network | |
| CN109088822B (en) | Data flow forwarding method, device, system, computer equipment and storage medium | |
| CN116389365A (en) | Switch data processing method and system | |
| Jin et al. | Towards joint resource allocation and routing to optimize video distribution over future internet | |
| Dou et al. | ProgrammabilityMedic: Predictable path programmability recovery under multiple controller failures in SD-WANs | |
| CN114448899A (en) | Method for balancing network load of data center | |
| US10666580B1 (en) | Network switches using centralized improvement of routing based on link capacity | |
| CN111865789B (en) | SR path constraint method based on segment routing | |
| CN113490231A (en) | Network slice availability guarantee method and device | |
| CN106100722B (en) | A kind of virtual network survivability mapping method that can distinguish maximum sharing capability | |
| CN109361603B (en) | Method and system for dynamically adjusting equivalent path flow based on programmable switching chip |
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 |