CN106789731B

CN106789731B - Queue scheduling method based on energy Internet service importance

Info

Publication number: CN106789731B
Application number: CN201611024465.6A
Authority: CN
Inventors: 金志刚; 徐珮轩; 罗咏梅
Original assignee: Tianjin University
Current assignee: Tianjin University
Priority date: 2016-11-17
Filing date: 2016-11-17
Publication date: 2020-03-06
Anticipated expiration: 2036-11-17
Also published as: CN106789731A

Abstract

The invention discloses a queue scheduling method based on energy internet service importance, which comprises the following steps: and (4) carrying out weight distribution on the data packet of the node to be forwarded by using an Analytic Hierarchy Process (AHP) and endowing the data packet with an importance attribute. And a queue scheduling stage: after the data packets enter the forwarding nodes or are generated by the forwarding nodes, the data packets and all the data packets at the same time are sorted according to the importance degrees, four queues are configured for each forwarding node according to the calculation capacity of the node and the traffic volume, the priority levels are from high to low, the data packets are sorted into the queues according to the importance degrees, the arrival sequence of the data packets in each queue is recorded, and 3 counters are used for maintaining the first three queues respectively. Firstly, the forwarding node carries out queue check, and the check contents are whether a data packet with the waiting time more than or equal to the maximum waiting time exists in the queue: if yes, the data packet is preferentially sent; and if the data packet does not exist, forwarding the data packet in the queue according to the order of first-in first-out.

Description

Queue scheduling method based on energy Internet service importance

Technical Field

The invention belongs to the field of computer networks, relates to a QOS (quality of service) technology, and particularly relates to a queue scheduling method based on intelligent power grid service importance.

Background

The energy internet is a novel information energy fusion wide area network constructed in the internet concept, a large power grid is used as a backbone network, a micro-grid is used as a local area network, and an equivalent information energy integration framework is opened, so that bidirectional on-demand transmission and dynamic balance use of energy are really realized, and the energy internet can be adapted to the access of new energy to the maximum extent. With the development of networks and the construction of smart power grids, power communication networks are continuously extended, so that the power communication network covering all 6 links of power generation, transmission, transformation, distribution, utilization and scheduling of the power grids is formed, and the characteristics of numerous systems and increasingly frequent communication among the systems are presented. Although the electric energy source is only one type of energy, the electric energy has incomparable advantages in the aspects of energy transmission efficiency and the like, and the transmission aspect of future energy infrastructure is also the power grid, so the future energy internet is basically an internet type power grid. The energy internet transforms a centralized, unidirectional grid into a grid that interacts with more consumers. And the power grid is closely combined with the power communication network, and the state of the power communication network directly relates to the safe and stable operation of the power grid. But due to the randomness and frequency of faults in the network, there is a potential risk of power communication traffic channels. How to quantitatively analyze the potential risks of the electric power communication service and adopt effective control measures becomes one of the important problems for guaranteeing the stable operation of the electric power system, and the classified transmission of the electric power communication service is a new idea aiming at the aspect.

Queue management is an important means for controlling network congestion, and plays a great role in network transmission, and scheduling is one of the core mechanisms of queue management. At present, a scheduler exists in each interface, a plurality of queues exist in the scheduler, and an output link is shared, so that the possibility of researching a complex queue scheduling algorithm is provided.

The communication network has a great variety of services, and each forwarding power saving not only needs to classify the data packets to be forwarded, but also has the capability of distinguishing and forwarding important service information in advance while ensuring that the communication network has good service capability, so that the communication network has better service quality. However, data packets that forward important traffic first and ignore general traffic will also cause the service quality of the communication network to degrade or even collapse, and therefore, such a routing method should be improved.

Disclosure of Invention

Aiming at some defects of the prior art, the invention provides a queue scheduling method based on the importance of energy Internet services.

In order to achieve the purpose, the invention adopts the following method:

a queue scheduling method based on energy Internet service importance degree comprises the following steps:

(1) and (3) carrying out weight distribution on the data packet of the node to be forwarded by using an Analytic Hierarchy Process (AHP), and giving an importance attribute to the data packet, wherein the importance attribute comprises attributes of category and priority.

(2) And a queue scheduling stage: after the data packets enter the forwarding nodes or are generated by the forwarding nodes, the data packets and all the data packets at the same time are sorted according to the importance degrees, four queues are configured for each forwarding node according to the calculation capacity of the node and the traffic volume, the priority levels are from high to low, the data packets are sorted into the queues according to the importance degrees, the arrival sequence of the data packets in each queue is recorded, and 3 counters are used for maintaining the first three queues respectively.

(3) Firstly, the forwarding node carries out queue check, and the check contents are whether a data packet with the waiting time more than or equal to the maximum waiting time exists in the queue: if yes, the data packet is preferentially sent; and if the data packet does not exist, forwarding the data packet in the queue according to the order of first-in first-out.

(4) When the forwarding starts, the forwarding node checks the self forwarding capability, and whether the number of the data packets exceeds the self forwarding capability is judged: if the number of the data packets exceeds the forwarding capacity of the data packets, forwarding the data packets according to the principle of comparing importance levels firstly and then comparing maximum waiting time; if the quantity of the data packets does not exceed the self-forwarding capability, forwarding is only carried out according to the maximum waiting time of the data packets.

(5) After the forwarding is finished, adding 1 to the counter value of each data packet currently existing in the first 3 queues, and waiting for the next comparison and forwarding.

Drawings

FIG. 1 is a schematic diagram of a queue scheduling method based on service importance

Detailed Description

The following description will first describe an embodiment with reference to the drawings.

An Analytic Hierarchy Process (AHP) is used to perform weight distribution on the data packet of the node to be forwarded, and an importance attribute is given to the data packet, wherein the importance attribute includes attributes of category and priority. The advantages are that: different importance levels can be given to different services or the same service in different time periods according to factors such as the service requirement, the scale of the communication network, the service volume and the like. The process after the packet enters the forwarding node is shown in fig. 1.

In the figure, four large rectangles in a node represent queues, each small square represents a data packet, the upper part of the slash of the square represents the waiting time of the data packet, and the lower part represents the arrival sequence of the data packet, namely the sequence number. The queues with the importance degrees of "one", "two" and "three" all have the data packet with the latency time of 0, which indicates that the data packet which is just transmitted or generated at the node at the last time step has no latency time in the queue with the lowest importance degree of "four".

The queue scheduling process is as follows:

(1) after the time step starts, firstly, the forwarding node performs queue check, the check content is whether a data packet with the waiting time being more than or equal to the maximum waiting time exists in the queue, and the data packets with the sequence numbers of 1, 2, 3 and 6 meet the conditions through detection.

(2) At this time, a judgment needs to be made: (a) if the transmission capability C of the node is less than 4, assuming that C is 3, the data packets need to be sorted according to a method of comparing importance levels first and then comparing arrival times, and the data packets are sorted in the order of 6, 3, 1, and 2, and finally the data packets with the sequence numbers of 6, 3, and 1 are transmitted; (b) if the transmission capacity C of the node is greater than 4, assuming that C is 8, first,

data packets

1, 2, 3, and 6 satisfying the previous detection condition are transmitted, and then, data packets with

sequence numbers

4, 5, 7, and 8 are sequentially transmitted in the order of first-in first-out

(3) After the transmission is finished, the waiting time value of each data packet currently existing on the queue with the importance degree of 'one', 'two' and 'three' is added with 1.

Claims

1. A queue scheduling method based on energy Internet service importance degree comprises the following steps:

(1) using an Analytic Hierarchy Process (AHP) to carry out weight distribution on a data packet to be forwarded by a forwarding node, and giving an importance attribute to the data packet, wherein the importance attribute comprises attributes of categories and priorities;

(2) and a queue scheduling stage: after the data packets enter the forwarding nodes or are generated by the forwarding nodes, the data packets and all the data packets at the same time are sorted according to the importance degrees, four queues are configured for each forwarding node according to the calculation capacity of the node and the traffic volume, the priority levels are from high to low, the data packets are distributed into the queues according to the importance degrees, and meanwhile, the arrival sequence of the data packets in each queue is recorded;

(3) firstly, the forwarding node carries out queue check, and the check contents are whether a data packet with the waiting time more than or equal to the maximum waiting time exists in the queue: if yes, the data packet is preferentially sent; if the data packet does not exist, forwarding the data packet in the queue according to the first-in first-out sequence;

(4) when the forwarding starts, the forwarding node checks the self forwarding capability, and whether the number of the data packets exceeds the self forwarding capability is judged: if the number of the data packets exceeds the forwarding capacity of the data packets, forwarding the data packets according to the principle of comparing importance levels firstly and then comparing maximum waiting time; if the quantity of the data packets does not exceed the self-forwarding capacity, forwarding the data packets according to the maximum waiting time of the data packets;