CN114172849A - Deterministic traffic shaping method based on game theory - Google Patents
Deterministic traffic shaping method based on game theory Download PDFInfo
- Publication number
- CN114172849A CN114172849A CN202111425795.7A CN202111425795A CN114172849A CN 114172849 A CN114172849 A CN 114172849A CN 202111425795 A CN202111425795 A CN 202111425795A CN 114172849 A CN114172849 A CN 114172849A
- Authority
- CN
- China
- Prior art keywords
- flow
- controller
- deterministic
- traffic
- resource
- 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 18
- 238000007493 shaping process Methods 0.000 title claims abstract description 15
- 238000002347 injection Methods 0.000 claims abstract description 39
- 239000007924 injection Substances 0.000 claims abstract description 39
- 238000013507 mapping Methods 0.000 claims abstract description 29
- 230000000737 periodic effect Effects 0.000 claims abstract description 10
- 230000015556 catabolic process Effects 0.000 claims abstract description 4
- 238000006731 degradation reaction Methods 0.000 claims abstract description 4
- 238000004891 communication Methods 0.000 claims description 6
- 238000004401 flow injection analysis Methods 0.000 claims description 5
- 230000003068 static effect Effects 0.000 claims description 5
- 239000007943 implant Substances 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000002452 interceptive effect Effects 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 230000000977 initiatory effect Effects 0.000 claims description 2
- 238000012545 processing Methods 0.000 claims description 2
- 238000013468 resource allocation Methods 0.000 claims description 2
- 230000002159 abnormal effect Effects 0.000 abstract description 2
- 238000009499 grossing Methods 0.000 abstract description 2
- 239000000243 solution Substances 0.000 description 4
- 235000003642 hunger Nutrition 0.000 description 2
- 238000012552 review Methods 0.000 description 2
- 230000037351 starvation Effects 0.000 description 2
- 208000014596 Berardinelli-Seip congenital lipodystrophy Diseases 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000001356 surgical procedure Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/22—Traffic shaping
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention relates to the technical field of deterministic networks, and provides a deterministic traffic shaping method based on a game theory, which is used for sensing the global requirement of deterministic traffic, establishing a mapping relation between traffic requirements and abstract resources, calculating the optimal injection time of periodic deterministic traffic while reserving bandwidth and buffer resources, smoothing a resource requirement curve along with time in a DetNet network by reasonably planning the injection offset of deterministic traffic, reducing peak values and improving the resource utilization rate. In addition, the controller reserves a protection bandwidth to consider the influence of the burst flow and the best effort flow on the queue resources, and for the abnormal flow which cannot predict the resource demand, compared with directly discarding the packet overflowing the queue, the method is more friendly to the burst flow and the best effort flow by adjusting the queue forwarding threshold and performing flow degradation, and the utilization rate of the queue resources is higher.
Description
Technical Field
The invention relates to the field of deterministic networks, in particular to a deterministic traffic shaping method based on a game theory.
Background
The internet is a best effort service with no guarantees on end-to-end delay, jitter and packet loss rate. With the increasing applications of industry 4.0, remote driving, remote surgery and the like, the requirements on ultra-low time delay and micro jitter of a network are higher and higher. In order to meet the requirements of related applications, a Time Sensitive Network (TSN) and a deterministic network (DetNet) respectively serve a link layer and a network layer to achieve deterministic and reliable forwarding.
While the TSN and DetNet standards are capable of achieving limited delay, they lack a mechanism to ensure packet delivery reliability, DetNet achieves limited delay and low packet loss by reserving bandwidth and buffer resources, and limits the worst-case impact of each node on end-to-end delay through the standard queuing and scheduling algorithms in TSNs.
However, in a large-scale DetNet environment, the problem that gating lists (GCLs) are difficult to calculate exists in time-aware shaper scheduling, and the criteria such as a frame preemption algorithm, circular queuing and forwarding do not have the ability to determine a deterministic flow in which a certain time slot exceeds the upper limit of resource allocation, and starvation of non-deterministic traffic can be caused.
Taking the circular queuing and forwarding mechanism as an example, the time division multiplexing problem of each switch does not need to be considered separately, but different queues are arranged for the flows with the same priority by differentiating the parity cycles, and the queuing operation and the forwarding operation are respectively carried out. The method theoretically ensures the upper and lower bounds of end-to-end delay, but for periodic time-sensitive flows, due to limited queue capacity, the lack of perception of global traffic easily causes the overflow of a queue in a certain period, and the strict gating setting of a circulating queuing and forwarding mechanism causes a plurality of time-sensitive data packets to be discarded.
Disclosure of Invention
The invention aims to solve the technical problem of providing a deterministic traffic shaping method based on a game theory. The method overcomes the defects of delay of packet loss and increase of jitter caused by flow overflow caused by global flow which are not considered by the existing flow shaping mechanism under the DetNet network.
In order to solve the technical problems, the invention is based on a circulating queuing and forwarding mechanism in the TSN standard, adjusts the injection time of periodic deterministic flows when allocating resources by sensing global flow, sets reasonable offset for each periodic deterministic flow, smoothes a resource demand curve in a network, provides a protection mechanism for burst flows and best effort flows, and greatly reduces jitter and packet loss when in high load under the condition of avoiding starvation.
In order to achieve the purpose, the invention provides a deterministic traffic shaping method based on a game theory, which comprises the following steps of:
step 1, measuring the queue resource state on the DetNet network, collecting the deterministic flow information in the network, and loading the network topology and the queue resource information deterministic flow characteristics into a controller.
And 2, planning the injection offset of each deterministic stream under the multi-constraint condition through an algorithm example of a layered complete static game, and generating corresponding update on the queue resource state when one stream is mapped.
And 3, updating the resource configuration condition in the network by taking the output result of the controller as the standard, and initiating a request to the controller and informing the flow characteristics when a new deterministic flow demand is generated on the host.
And 4, the controller issues feasible configuration to the host and the switch in the data plane, and the feasible configuration is not interacted with the host and the switch, so that Nash balance is achieved within limited game times.
And 5, for the best-effort flow, the switch performs non-deterministic flow forwarding window adjustment and flow degradation processing according to the current queue resource occupation condition and forwards the flow in a non-deterministic flow protection band.
And 6, when the periodic deterministic flow is finished, informing the controller to recover the mapped resources.
Preferably, step 1 comprises the sub-steps of:
step 1-1, in order to enable the controller to effectively map resources in the DetNet network and facilitate the calculation of deterministic traffic injection offsets, the controller constructs a topology capable of describing all nodes of the network and the connection thereof, and keeps communication with a switch and a host node in the network, thereby ensuring the real-time synchronization of the network state.
Step 1-2, maintaining, by a communication controller with a node, a state mapping of each queued resource block identified by space (switches and ports) and time (slots).
And 1-3, when the host computer needs to generate flow, actively communicating with the controller, informing the controller of flow characteristics including a source address, a destination address, a sending frequency and negotiating injection time.
Step 1-4: the controller loads abstracted information including network topology, resource state mapping and traffic demands.
Preferably, step 2 comprises the following sub-steps:
and 2-1, calculating a flow path and informing the host path and the corresponding class deterministic flow of resource occupation on the switch.
And 2-2, after the path is calculated, formalizing the injection scheduling problem of the periodic flow under the multi-constraint condition.
Step 2-3, the controller transmits the path and the switch resource state information which is used by the mapped corresponding category flow to the host Hl，HlAnd calculating the optimal flow injection time through a greedy algorithm.
And 2-4, based on the complete information static game, directly sending the flow after the host calculates the optimal flow offset, and informing the controller to modify the network resource mapping condition. And after receiving the traffic injection offset information, the controller updates the mapping of the characteristics of the traffic and the resource state of the corresponding priority queue.
Preferably, step 3 comprises the steps of:
and 3-1, the controller issues the received flow configuration information and the updated resource mapping condition to the host to update the resource configuration condition in the network.
And 3-2, when a new deterministic flow demand is generated on the host, no flow is directly injected, firstly, a request is initiated to the controller, the injection of the new flow is applied, the flow characteristics are informed to the controller, the controller approves the application, and the resource mapping is updated.
Preferably, step 4 comprises the steps of:
and 4-1, after the controller receives the flow configuration information and updates the resource mapping each time, broadcasting the new resource information to all hosts and switches of the data plane.
And 4-2, continuously interacting the controller with all the hosts until all the hosts do not have a more optimal offset injection strategy to enable the queue resource occupation situation on the path to be smoother, and thus achieving a Nash equilibrium state.
Step 4-3. generating a new deterministic traffic demand f in a topology already in Nash equilibriumjThen, the controller still calculates path in the same manner as in step 2-1jAnd with H in steps 2-2 and 2-3lThe flow f is configured in an interactive modejThe implant offset of (1). The new queue resource state may cause the injection scheme of the original traffic in the topology to no longer be the optimal strategy. At each time slot, the host reviews the injection configuration for all class streams, calculates a new policy based on the optimization objective in step 2-2, and sends a policy adjustment application to the controller.
And 4-4, the controller measures the adjustment applications from all the hosts, traverses the injection offset applications proposed by a plurality of hosts in the same time slot, randomly approves one offset application, informs the whole network and updates the resource mapping state.
Preferably, step 5 comprises the steps of:
and 5-1, aiming at the deterministic traffic, performing flow identification on a link layer for the non-deterministic traffic, and reserving a guard band for the non-deterministic traffic when the controller allocates resources by the scheduling of the steps 2 to 4.
And step 5-2, when the output port of the switch receives the non-deterministic flow in the deterministic flow time slot in a period, combining the idle condition of the current output port resource to carry out queue scheduling.
And 5-3, restoring the adjustment of the time slot ratio after the non-deterministic flow is finished, and adjusting to the level set by the controller for each switch.
Preferably, step 6 comprises the steps of:
and 6-1, calculating the ending requirement of the deterministic flow.
And 6-2, when the flow is periodically determined to be finished, informing the controller, modifying the queue resource mapping by the controller, informing the whole network, and calculating a new offset strategy by each host to achieve new Nash equilibrium.
This problem can be solved by using the global queue more evenly, and specifically, the forwarding threshold of the global traffic adjustment queue and the adjustment period traffic offset can be sensed.
The patent provides a deterministic traffic shaping method based on game theory, which is used for perceiving the global requirement of deterministic traffic, establishing the mapping relation between traffic requirement and abstract resources, calculating the optimal injection time of periodic deterministic traffic while reserving bandwidth and buffer resources, smoothing a resource requirement curve along with time in a DetNet network by reasonably planning the injection offset of the deterministic traffic, reducing peak value and improving the resource utilization rate. In addition, the controller reserves a protection bandwidth to consider the influence of the burst flow and the best effort flow on the queue resources, and for the abnormal flow which cannot predict the resource demand, compared with directly discarding the packet overflowing the queue, the method is more friendly to the burst flow and the best effort flow by adjusting the queue forwarding threshold and performing flow degradation, and the utilization rate of the queue resources is higher.
Drawings
The technical solution of the present invention will be further specifically described with reference to the accompanying drawings and the detailed description.
FIG. 1 is a main flow chart of the present invention.
Detailed Description
Step 1-1, in order to enable the controller to effectively map resources in the DetNet network and facilitate the calculation of deterministic traffic injection offsets, the controller constructs a topology capable of describing all nodes of the network and the connection thereof, and keeps communication with a switch and a host node in the network, thereby ensuring the real-time synchronization of the network state.
Step 1-2, maintaining, by a communication controller with a node, a state mapping of each queued resource block identified by space (switches and ports) and time (slots).
And 1-3, when the host computer needs to generate flow, actively communicating with the controller, informing the controller of flow characteristics including a source address, a destination address, a sending frequency and negotiating injection time.
| fid | src | dst | cvcle | EndTime | priority |
| 0 | H0 | H1 | 20us | 10s | 7 |
| … | … | … | … | … | … |
| n | Hm | Ho | 100us | 5s | 6 |
Step 1-4: the controller loads abstracted information such as network topology, resource state mapping, traffic demand, and the like.
Step 2 comprises the following substeps:
step 2-1: and calculating a flow path, and informing the host path and the corresponding class deterministic flow of resource occupation on the switch.
Step 2-1 specifically comprises the following steps that during initialization, no flow exists in the DetNet network, flow scheduling and queue resource occupation are in a balanced state, and the flow scheduling and queue resource occupation come from a host HlNew flow demand fiIs sent to the controller to calculate pathiReturning the occupation condition of the queue on the path to HlDescribe flow f in conjunction with src and dst in network topology G ═ { V, E, B } and traffic characteristicsiOf (2) a
pathi＝(S(1，0)，...，S(j，k))
Wherein S is(j，k)Indicating port k on the jth switch.
Step 2-2: after the path is calculated, the injection scheduling problem of the periodic flow under the multi-constraint condition is formalized.
Step 2-2 specifically includes the step of the injection scheduling problem being formulated as the smoothest consumption of queue resources on the flow paths, with the goal of minimizing the maximum queue resource overhead on all mapping flow paths when the number of mapping flows is fixed.
Step 2-3: the controller transmits the path and the switch resource state information which is used by the mapped corresponding category flow to the host Hl，HlAnd calculating the optimal flow injection time through a greedy algorithm.
Step 2-3 specifically includes the step of passing the current network queue resource status, HlAdopting a greedy strategy, traversing all selectable injection moments in a period, and counting the pathiThe resource occupation of all switches, ports and queues is increased, and the flow f is selected to be enablediThe path resource occupies the smoothest injection offset, the offset constraint needs to be considered at the moment, and the offset of the injection time slot needs to be less than fiThe period of (c).
Step 2-4: based on the complete information static game, the host calculates the optimal flow offset and then directly sends the flow, and informs the controller to modify the network resource mapping condition. And after receiving the traffic injection offset information, the controller updates the mapping of the characteristics of the traffic and the resource state of the corresponding priority queue.
The step 3 comprises the following steps:
step 3-1: the controller issues the received flow configuration information and the updated resource mapping condition to the host, and updates the resource configuration condition in the network.
Step 3-2: when a new deterministic flow demand is generated on the host, the flow is not directly injected any more, a request is firstly initiated to the controller, the injection of the new flow is applied, and the flow characteristic of the controller is informed.
Step 3-1 comprises in particular the step of, for a new flow fi+1And f is andiperforming parallel computations, corresponding host Hl′Simultaneously communicating with the controller to obtain the appropriate fi+1The implant offset of (1). However, the new traffic injection may make the original traffic injection configuration no longer the optimal choice, and the host H, which has been injecting traffic periodically, is informed by the controllerlA greedy strategy is used to compute a more optimal offset and apply it to the controller. The maximum error of the original offset from the new offset needs to satisfy the constraint of deterministic traffic error.
Step 4 comprises the following steps:
step 4-1: the controller broadcasts new resource information to all hosts and switches of the data plane after updating the resource map each time the controller receives traffic configuration information.
Step 4-2: the controller continuously interacts with all the hosts until all the hosts do not have a better offset injection strategy to enable the queue resource occupation situation on the path to be smoother, and then the Nash equilibrium state is achieved.
Step 4-3: generating a new deterministic traffic demand f in a topology already in Nash equilibriumjThen, the controller still calculates path in the same manner as in step 2-1jAnd with H in steps 2-2 and 2-3lThe flow f is configured in an interactive modejThe implant offset of (1). The new queue resource state may cause the injection scheme of the original traffic in the topology to no longer be the optimal strategy. At each time slot, the host reviews the injection configuration for all class streams, calculates a new policy based on the optimization objective in step 2-2, and sends a policy adjustment application to the controller.
Step 4-4: the controller measures adjustment applications from all the hosts, traverses injection offset applications proposed by a plurality of hosts in the same time slot, randomly approves one offset application, notifies the whole network, and updates the resource mapping state.
Step 5 comprises the following steps:
and 5-1, aiming at the deterministic traffic, performing flow identification on a link layer for the non-deterministic traffic, and reserving a guard band for the non-deterministic traffic when the controller allocates resources by the scheduling of the steps 2 to 4.
Step 5-1 specifically comprises the steps that the guard band is a fixed finer-grained time slot per time slot segment, and the controller allocates corresponding resources on the time slot based on the deterministic traffic and non-deterministic traffic share ratio in the network.
And step 5-2, when the output port of the switch receives the non-deterministic flow in the deterministic flow time slot in a period, combining the idle condition of the current output port resource to carry out queue scheduling.
Step 5-2 specifically comprises the step of setting a deterministic time slot window in one period to be [ t ]1，cycleend]When burst traffic occurs in deterministic slots, the logic is as follows
Wherein,
represents the flow rate fiThe priority of (2).
After the flow is degraded, adding the tail part of a new queue and determining the starting time t of a window1Decrease, t1＝t1-cycle/20. When the deterministic traffic occurs in the non-deterministic time slot, the traffic is directly added to the tail of the deterministic traffic.
And 5-3, restoring the adjustment of the time slot ratio after the non-deterministic flow is finished, and adjusting to the level set by the controller for each switch.
Step 6 comprises the following steps:
and 6-1, calculating the ending requirement of the deterministic flow.
Step 6-1 specifically comprises the step of determining a flow fiWith a constraint on the end time
The slot offset in each cycle can significantly affect the time to eventually arrive at dst, this constraint is typically strict for each flow, so the expected end time is calculated and a reasonable final flow injection time is set before the deterministic flow ends.
And 6-2, when the flow is periodically determined to be finished, informing the controller, modifying the queue resource mapping by the controller, informing the whole network, and calculating a new offset strategy by each host to achieve new Nash equilibrium.
Finally, it should be noted that the above-mentioned embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, which should be covered by the claims of the present invention.
Claims (7)
1. A deterministic traffic shaping method based on game theory is characterized by comprising the following steps:
step 1, measuring the queue resource state on a DetNet network, acquiring deterministic flow information in the network, and loading network topology and queue resource information deterministic flow characteristics into a controller;
planning injection offset of each deterministic stream under a multi-constraint condition through an algorithm example of a layered complete static game, and generating corresponding update on the queue resource state when one stream is mapped;
step 3, updating the resource allocation condition in the network by taking the output result of the controller as the standard, initiating a request to the controller when a new deterministic traffic demand is generated on the host, and informing the traffic characteristics;
step 4, the controller issues feasible configuration to a host and a switch in a data plane, and the feasible configuration is not interacted with the host and the switch, so that Nash balance is achieved within limited game times;
step 5, for best-effort traffic, the switch performs non-deterministic traffic forwarding window adjustment and traffic degradation processing according to the current queue resource occupation condition, and forwards the traffic in a non-deterministic traffic protection band;
and 6, when the periodic deterministic flow is finished, informing the controller to recover the mapped resources.
2. A method for game theory based deterministic traffic shaping according to claim 1, characterized in that said step 1 comprises the following sub-steps:
step 1-1, in order to enable a controller to effectively map resources in a DetNet network and facilitate calculation of deterministic traffic injection offset, the controller constructs a topology capable of describing all nodes of the network and connection thereof, and keeps communication with a switch and a host node in the network so as to ensure real-time synchronization of network states;
step 1-2, maintaining state mapping of each queue resource block identified by space and time through a communication controller of a node;
step 1-3, when the host computer needs to generate flow, the host computer actively communicates with the controller, informs the controller of flow characteristics including a source address, a destination address and a sending frequency, and negotiates injection time;
step 1-4: the controller loads abstracted information including network topology, resource state mapping and traffic demands.
3. A method for game theory based deterministic traffic shaping according to claim 2, characterized in that said step 2 comprises the following sub-steps:
step 2-1, calculating a flow path and informing a host path and a corresponding category deterministic flow of resource occupation on a switch;
step 2-2, after the path is calculated, formalizing the injection scheduling problem of the periodic flow under the multi-constraint condition;
step 2-3, the controller transmits the path and the switch resource state information which is used by the mapped corresponding category flow to the host Hl，HlCalculating the optimal flow injection time through a greedy algorithm;
step 2-4, based on the complete information static game, directly sending the flow after the host calculates the optimal flow offset, and informing the controller to modify the network resource mapping condition; and after receiving the traffic injection offset information, the controller updates the mapping of the characteristics of the traffic and the resource state of the corresponding priority queue.
4. A method for game theory based deterministic traffic shaping according to claim 3, characterized in that said step 3 comprises the following sub-steps:
step 3-1, the controller issues the received flow configuration information and the updated resource mapping condition to the host computer to update the resource configuration condition in the network;
and 3-2, when a new deterministic flow demand is generated on the host, no flow is directly injected, firstly, a request is initiated to the controller, the injection of the new flow is applied, the flow characteristics are informed to the controller, the controller approves the application, and the resource mapping is updated.
5. A method for game theory based deterministic traffic shaping according to claim 4, characterized in that said step 4 comprises the following sub-steps:
step 4-1, after the controller receives the flow configuration information and updates the resource mapping each time, the controller broadcasts new resource information to all hosts and switches of the data plane;
step 4-2, the controller continuously interacts with all the hosts until all the hosts do not have a more optimal offset injection strategy to enable the queue resource occupation situation on the path to be smoother, and then a Nash equilibrium state is achieved;
step 4-3. generating a new deterministic traffic demand f in a topology already in Nash equilibriumjThen, the controller still calculates path in the same manner as in step 2-1jAnd with H in steps 2-2 and 2-3lThe flow f is configured in an interactive modejThe implant offset of (1); the new queue resource state can cause the original flow injection scheme in the topology to be no longer the optimal strategy; in each time slot, the host machine inspects the injection configuration of all the category flows, calculates a new strategy according to the optimization target in the step 2-2 and sends a strategy adjustment application to the controller;
and 4-4, the controller measures the adjustment applications from all the hosts, traverses the injection offset applications proposed by a plurality of hosts in the same time slot, randomly approves one offset application, informs the whole network and updates the resource mapping state.
6. A method for game theory based deterministic traffic shaping according to claim 5, characterized in that said step 5 comprises the following sub-steps:
step 5-1, the dispatching from step 2 to step 4 aims at the deterministic traffic, and for the non-deterministic traffic, the flow identification is carried out at the link layer, and the controller reserves a guard band for the non-deterministic traffic when allocating resources;
step 5-2, when the output port of the switch receives the non-deterministic flow in the deterministic flow time slot in a period, combining the resource idle condition of the current output port to carry out queue scheduling;
and 5-3, restoring the adjustment of the time slot ratio after the non-deterministic flow is finished, and adjusting to the level set by the controller for each switch.
7. A method for game theory based deterministic traffic shaping according to claim 6 characterized in that said step 1 comprises the following sub-steps:
step 6-1, calculating the ending requirement of the deterministic flow;
and 6-2, when the flow is periodically determined to be finished, informing the controller, modifying the queue resource mapping by the controller, informing the whole network, and calculating a new offset strategy by each host to achieve new Nash equilibrium.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111425795.7A CN114172849B (en) | 2021-11-26 | 2021-11-26 | Deterministic traffic shaping method based on game theory |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111425795.7A CN114172849B (en) | 2021-11-26 | 2021-11-26 | Deterministic traffic shaping method based on game theory |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114172849A true CN114172849A (en) | 2022-03-11 |
| CN114172849B CN114172849B (en) | 2024-07-12 |
Family
ID=80481190
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111425795.7A Active CN114172849B (en) | 2021-11-26 | 2021-11-26 | Deterministic traffic shaping method based on game theory |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114172849B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115277561A (en) * | 2022-05-16 | 2022-11-01 | 国网福建省电力有限公司超高压分公司 | Transmission queue control method and terminal based on time sensitivity |
| CN115333860A (en) * | 2022-10-12 | 2022-11-11 | 北京合众方达科技有限公司 | TSN network control method based on zero trust |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150332145A1 (en) * | 2014-05-13 | 2015-11-19 | Cisco Technology, Inc. | Traffic shaping based on predicted network resources |
| CN111740924A (en) * | 2020-07-29 | 2020-10-02 | 上海交通大学 | Traffic shaping and routing planning scheduling method of time-sensitive network gating mechanism |
| CN112132202A (en) * | 2020-09-18 | 2020-12-25 | 嘉兴学院 | Edge computing collaborative member discovery method based on comprehensive trust evaluation |
| CN112511462A (en) * | 2020-12-17 | 2021-03-16 | 上海交通大学 | Software-defined industrial heterogeneous time-sensitive network system and resource scheduling method |
| CN112616189A (en) * | 2020-12-10 | 2021-04-06 | 北京邮电大学 | Static and dynamic combined millimeter wave beam resource allocation and optimization method |
-
2021
- 2021-11-26 CN CN202111425795.7A patent/CN114172849B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150332145A1 (en) * | 2014-05-13 | 2015-11-19 | Cisco Technology, Inc. | Traffic shaping based on predicted network resources |
| CN111740924A (en) * | 2020-07-29 | 2020-10-02 | 上海交通大学 | Traffic shaping and routing planning scheduling method of time-sensitive network gating mechanism |
| CN112132202A (en) * | 2020-09-18 | 2020-12-25 | 嘉兴学院 | Edge computing collaborative member discovery method based on comprehensive trust evaluation |
| CN112616189A (en) * | 2020-12-10 | 2021-04-06 | 北京邮电大学 | Static and dynamic combined millimeter wave beam resource allocation and optimization method |
| CN112511462A (en) * | 2020-12-17 | 2021-03-16 | 上海交通大学 | Software-defined industrial heterogeneous time-sensitive network system and resource scheduling method |
Non-Patent Citations (1)
| Title |
|---|
| 高平;张帆;张东;翟飞龙;: "基于SDN的云架构网络高确定性流量控制方法", 计算机工程, no. 12, 15 December 2018 (2018-12-15) * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115277561A (en) * | 2022-05-16 | 2022-11-01 | 国网福建省电力有限公司超高压分公司 | Transmission queue control method and terminal based on time sensitivity |
| CN115277561B (en) * | 2022-05-16 | 2023-12-15 | 国网福建省电力有限公司超高压分公司 | Transmission queue control method and terminal based on time sensitivity |
| CN115333860A (en) * | 2022-10-12 | 2022-11-11 | 北京合众方达科技有限公司 | TSN network control method based on zero trust |
| CN115333860B (en) * | 2022-10-12 | 2023-02-03 | 北京合众方达科技有限公司 | TSN network control method based on zero trust |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114172849B (en) | 2024-07-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11916782B2 (en) | System and method for facilitating global fairness in a network | |
| US11316774B2 (en) | Path selection method and apparatus | |
| US9338099B2 (en) | Dynamic queuing and pinning to improve quality of service on uplinks in a virtualized environment | |
| US10355981B1 (en) | Sliding windows | |
| US9571402B2 (en) | Congestion control and QoS in NoC by regulating the injection traffic | |
| US8537846B2 (en) | Dynamic priority queue level assignment for a network flow | |
| US20090116381A1 (en) | Method and system for congestion management in a fibre channel network | |
| US11552857B2 (en) | Methods, systems and appratuses for optimizing the bin selection of a network scheduling and configuration tool (NST) by bin allocation, demand prediction and machine learning | |
| CN106302227B (en) | hybrid network flow scheduling method and switch | |
| CN114172849A (en) | Deterministic traffic shaping method based on game theory | |
| US20220407808A1 (en) | Service Level Adjustment Method, Apparatus, Device, and System, and Storage Medium | |
| US10917355B1 (en) | Methods, systems and apparatuses for optimizing time-triggered ethernet (TTE) network scheduling by using a directional search for bin selection | |
| CN108476175B (en) | Transfer SDN traffic engineering method and system using dual variables | |
| CN105391651B (en) | Virtual optical network multi-layer resource convergence method and system | |
| US20230096063A1 (en) | Communication method and apparatus | |
| KR20130137539A (en) | System for performing data cut-through | |
| JP2016010138A (en) | Communication device, communication system, and communication method | |
| US20210203620A1 (en) | Managing virtual output queues | |
| US20230336486A1 (en) | Service flow scheduling method and apparatus, and system | |
| CN114531399B (en) | Memory blocking balancing method and device, electronic equipment and storage medium | |
| US12028265B2 (en) | Software-defined guaranteed-latency networking | |
| CN114095434B (en) | Method for controlling network congestion and related device | |
| WO2023236832A1 (en) | Data scheduling processing method, device, and apparatus, and storage medium | |
| WO2022246710A1 (en) | Method for controlling data stream transmission and communication device |
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 |