CN113938176B - Low-delay service space-based calculation method - Google Patents

Abstract

A low-delay service day-based calculation method comprises the steps of obtaining network state information, combining task attributes, utilizing matching of tasks and calculation nodes to obtain a task calculation arrangement result, then carrying out subtask distribution and calculation, and converging the result. The method can improve the service efficiency of the communication and calculation resources on the satellite, reduce the service transmission processing time delay and meet the service quality requirement of the low-time-delay service.

Description

Low-delay service space-based calculation method

Technical Field

The invention relates to a low-delay business space-based calculation method, and belongs to the technical field of space-based calculation.

Background

The low-orbit networking constellation develops rapidly, the requirement of completing military and civil service through the low-orbit constellation is continuously increased, the low-orbit constellation comprises satellite low-time delay processing transmission service, such as military applications including rapid target extraction, rapid environment perception and the like, and emergency rescue applications including search and rescue and the like, in the applications, the data volume generated by a sensor is larger and larger, the data processing calculation is more and more complex, the calculation dependence is more and more serious, and contradiction is formed between the low-time delay requirement of the service and the data processing calculation. Meanwhile, the satellite communication network is unbalanced in bearing distribution caused by regional and time zone differences, constellation operation and earth rotation, and the traffic load has time-varying characteristics. A single star may cover both areas of scarce traffic, such as polar areas and the ocean, and areas of dense population and data volume, such as the main cities of developed countries. The satellite calculation and the communication resources are used in an unbalanced way, part of the satellites run under high load, and part of the satellite calculation communication resources are idle.

With the development of satellite on-orbit processing technology, a certain degree of computing capability can be provided, however, due to the limitation of the size and the power consumption, the space-based computing platform has the problems of low computing capability, small storage space and other resources, and a single satellite node is difficult to meet the low-time-delay computing requirement, so how to realize the processing and transmission of low-time-delay services through the cooperation of a plurality of satellite nodes based on the limited space-based computing resources becomes a problem which needs to be solved urgently.

The existing space-based routing algorithm mainly uses data communication, does not consider data content in the data transmission process, does not process the data, does not fully utilize the calculation resources of the space-based calculation platform, and has large transmission data quantity and prolonged transmission time.

Disclosure of Invention

The technical solution of the invention is as follows: the method overcomes the defects of the prior art, provides a low-delay service space-based calculation method, solves the problem of long service calculation time of the traditional method, and meets the requirement of low-delay service.

The technical scheme of the invention is as follows:

A low-delay service space-based calculation method comprises the following steps:

1) Setting a to-be-transmitted processing task of a task source node satellite S _s as a task T (D, R), wherein D is input data to be processed by the task T, and R is output data processed by the task T; decomposing the T into a plurality of parallel subtasks, and completing the task T when all the subtasks of the T are completed;

2) Each satellite in the low-orbit network periodically reports the network state of the satellite to the high-orbit satellite, and the high-orbit satellite periodically broadcasts the network state of the low-orbit satellite;

Each satellite in the low-orbit network has 2 front and back in orbit and 2 total 4 inter-satellite links between the left and right in orbit, and the content reported by the ith satellite S _i comprises an available bandwidth set B _i Mbps of four inter-satellite links and the current available computing resource per se Mflops; b _i＝{B_i,1,B_i,2,B_i,3,B_i,4, wherein B _i,1 is the available bandwidth of the first inter-satellite link of the ith satellite S _i, B _i,2 is the available bandwidth of the second inter-satellite link of the ith satellite S _i, B _i,3 is the available bandwidth of the third inter-satellite link of the ith satellite S _i, and B _i,4 is the available bandwidth of the fourth inter-satellite link of the ith satellite S _i;

3) Determining a candidate transmission path node set according to the low orbit satellite network state, sorting the subtasks according to the subtasks and the subtask attributes decomposed in the step 1), arranging the subtasks on the candidate transmission path node set according to the sorting, and mapping the subtasks to corresponding satellite nodes;

4) According to the arrangement result, the input data to be processed of the subtasks is sent to the corresponding satellite nodes, the satellite nodes load processing programs according to the subtask calculation requirements, the received subtask data are processed, the processing result is sent to the target nodes, the processing result is set to be a high-priority service, and the high-priority service is sent through reserved bandwidth;

5) And after each subtask is correctly executed, feeding back to the task source node, and for the task with failed execution, reassigning the transmission calculation process by the task source node until the calculation result is correct, and converging all subtasks at the target node after the subtasks are correctly executed, so as to complete the transmission calculation of the whole task.

In the step 3), the subtasks are ordered according to the following method:

Let the order factor of the j-th subtask be lambda _j,

Taking lambda _j＝size(D_j)/size(D_max)-c_{task_j}/c_max, wherein D _j is input data to be processed of the jth subtask, size (D _j) is the size of the input data to be processed of the jth subtask, D _max is the input data to be processed corresponding to the subtask with the largest size of the input data to be processed in the subtask set, c _{task_j} is the calculated amount required for processing the jth subtask, c _max is the calculated amount of the subtask with the largest calculated amount required in the subtask set, and lambda _j is a task ordering factor;

each subtask is ordered from big to small by an ordering factor.

In the step 3), the method for determining the candidate transmission path node set is as follows:

The low-orbit satellite network is abstracted into a 2-dimensional mesh network, all nodes contained in a path set with the minimum hop count from a task source node S _s to a target node S _d form a candidate transmission path node set, and the total number of nodes in the set is L.

Setting a candidate transmission path node set from a task source node S _s to a target node S _d as S= { S _i }, abstracting a low-orbit satellite network into a 2-dimensional mesh network, wherein in the 2-dimensional mesh network topology, an inter-orbit link is recorded as an x-axis direction, and an in-orbit link is recorded as a y-axis direction; for a certain communication, setting the coordinates of a task source node S _s as (x _s,y_s) and the coordinates of a target node S _d as (x _d,y_d);

If the constellation without reverse slots is adopted, the logical distance difference between the x and y direction communication between the task source node S _s and the target node S _d is as follows:

If the constellation is a constellation with reverse slots, the difference of the communication logical distances in the x and y directions between the task source node S _s and the target node S _d is as follows:

Wherein N is the number of track surfaces, and M is the number of satellites on one track surface;

Then at most common from S _s to S _d A minimum hop path with hop distance delta x + delta y; the set of nodes on the paths is a candidate transmission path node set, and the minimum hop count path set is a candidate transmission path set.

If the constellation without reverse slots is adopted, the path in the x-axis direction in the minimum hop path is as follows:

(a) If |x _d-x_s|≤N-|x_d-x_s |, the path in the x-axis direction is:

When x _s>x_d, x _s→x_s-1→x_s-2→…→x_d;

when x _s≤x_d, x _s→x_s+1→x_s+2→…→x_d;

(b) If |x _d-x_s|>N-|x_d-x_s |, the path in the x-axis direction is:

When x _s≤x_d, x _s→x_s-1→x_s-2→…→1→N→…→x_d;

When x _s>x_d, x _s→x_s+1→x_s+2→…→N→1→…→x_d.

Otherwise if there is a constellation with reverse slots:

When x _s>x_d, x _s→x_s-1→x_s-2→…→x_d;

When x _s≤x_d, x _s→x_s+1→x_s+2→…→x_d.

The constellation without reverse slot and the constellation with reverse slot have the same path in the Y-axis direction, and are

(A) If |y _d-y_s|≤M-|y_d-y_s |, the path in the y-axis direction is:

Y _s→y_s-1→y_s-2→…→y_d when y _s>y_d;

Y _s→y_s+1→y_s+2→…→y_d when y _s≤y_d;

(b) If |y _d-y_s|>M-|y_d-y_s |, the path in the y-axis direction is:

Y _s→y_s-1→y_s-2→…→1→M→…→y_d when y _s≤y_d;

Y _s→y_s+1→y_s+2→…→M→1→…→y_d when y _s>y_d.

In the step 3), the subtasks are arranged as follows:

For each of the sub-tasks after sequencing:

For each satellite node S _k, k=1,..l, S ₁ is the source node S _s when k=1, S _L is the target node S _d when k=l, the following process is performed:

The size of the computing resources allocated to the jth subtask for node S _k, The available computing resource size for node S _k;

For the maximum available bandwidth in the set of all candidate transmission paths from the source node S _s to the node S _k, the maximum available bandwidth is determined by the inter-satellite link available bandwidth in step 2), To allocate communication bandwidth for the j-th subtask on the candidate transmission path from the source node S _s to the node S _k,In order to place the jth subtask in the node S _k for execution, the switch is the total time delay of satellite node data exchange on the transmission path from the source node S _s to the node S _k, the size (D _j) is the size of input data to be processed by the jth subtask, task is a set of all subtasks, and c _{task_j} is the amount of computation required for processing the jth subtask;

If it is The j-th subtask is scheduled to be executed on node S _k and the system state is updated as follows:

Task = Task- { subtask j }

For all nodes S _i on the maximum bandwidth path from node S _s to node S _k:

otherwise, the j-th subtask is arranged to be executed by the source node, and the system state is updated as follows:

Task = Task- { subtask j }

Compared with the prior art, the invention has the beneficial effects that:

(1) The invention combines the data transmission delay, the task calculation delay and the result convergence delay by combining the routing and the service calculation distribution, globally considers the timeliness of task execution, shortens the information generation period and reduces the service overall delay.

(2) The invention abstracts the low orbit constellation into the self-organizing mesh network, and parallelizes and distributes the satellite calculation task by combining the distributed calculation thought so as to improve the timeliness and fault tolerance of task execution.

(3) The invention jointly considers the communication load and the calculation load, distributes according to the network state as required, can realize the load balance of the space-based nodes, and improves the calculation efficiency and the communication efficiency.

Drawings

FIG. 1 is a flow chart of the present invention;

fig. 2 is a scene diagram of an embodiment of the invention.

Detailed Description

The following describes in further detail the embodiments of the present invention with reference to the accompanying drawings.

The invention provides a low-delay service space-based calculation method, which provides routing and calculation strategies according to traffic load, calculation load and other conditions, selects paths in real time, can adapt to network traffic change and node calculation load change, has strong self-adaptability, and solves the problem of high-efficiency processing and transmission of low-delay service of a low-orbit satellite calculation platform by fully utilizing space-based calculation and communication resources to realize information from data acquisition and information extraction to information transmission. The method can improve the service efficiency of the communication and calculation resources on the satellite, reduce the service transmission processing time delay and meet the service quality requirement of the low-time-delay service.

As shown in fig. 1, a low-delay service space-based calculation method comprises the following steps:

11 Task decomposition

Let the task to be transmitted of satellite S _s be task T (D, R), D be input data to be processed of task T, and R output data after processing of task T. And decomposing the T into parallel subtasks to finish the task T together. When each subtask of T is completed, then task T is completed.

12 Network state acquisition for path acquisition and computing node selection is required as follows

Each satellite in the low-orbit network has 2 front and back in orbit and 2 left and right in orbit, 4 total inter-satellite links, and all satellites S _i in the low-orbit network periodically report own network load to the high-orbit satellites, wherein the reported content comprises available bandwidths B _i＝{B_i,1,B_i,2,B_i,3,B_i,4 Mbps of four inter-satellite links and available computing capacity C _i Mflops of the current satellites.

The high orbit satellite periodically broadcasts the low orbit satellite network status.

13 A) a calculation node and a transmission path selection method as follows

According to the network state obtained in the step 2) and the decomposed subtasks and subtask attributes in the step 1), sequencing the subtasks firstly, then arranging the tasks on candidate transmission paths in sequence, and mapping the subtasks to corresponding satellite nodes;

14 Transmission and calculation)

And 3) according to the arrangement result of the 3), transmitting the task data to be processed corresponding to the subtasks to the corresponding satellite nodes, loading a processing program according to the subtask calculation requirement by the satellite nodes, processing the received subtask data and transmitting the processing result to the target nodes. The processing result is set as high priority service and is sent through reserved bandwidth.

15 Subtask convergence

And after each subtask is correctly executed, feeding back to the task source node, and for the task which fails to be executed. And the task source node redistributes the transmission calculation process until the calculation result is correct, and all the subtasks are assembled at the target node after being correctly executed, so that the transmission calculation of the whole task is completed.

2. The sorting method described in step 1) 13) is as follows:

Let the order factor of the j-th subtask be lambda _j,

Taking lambda _j＝size(D_j)/size(D_max)-c_{task_j}/c_max, wherein D _j is the data to be processed of the task j, size (D _j) is the data size, size (D _max) is the data size to be processed corresponding to the subtask with the largest input data size, c _{task_j} is the calculated amount required for processing the task j, and c _max is the calculated amount of the subtask with the largest calculated amount required.

Each subtask is ordered from big to small by an ordering factor.

3. The acquisition method of the candidate transmission path node set comprises the following steps:

The low-orbit satellite network is abstracted into a 2-dimensional mesh network, all nodes contained in a path set with the minimum hop count from a task source node S _s to a target node S _d form a candidate transmission path node set, and the total number of nodes in the set is L.

Let the set of all nodes in the minimum hop-count path range from the source node S _s to the target node S _d be s= { S _i }, the low-orbit satellite network can be abstracted into a 2-dimensional mesh network, in the 2-dimensional mesh network topology, the inter-orbit link is in the x-axis direction, and the in-orbit link is in the y-axis direction. For a certain communication, the data communication source node is S _s(x_s,y_s), the destination node is S _d(x_d,y_d), if the constellation is a constellation without reverse slots, the logical distance difference between the task source node S _s and the destination node S _d in the x and y directions is:

If the constellation is a constellation with reverse slots, the difference of the communication logical distances in the x and y directions between the task source node S _s and the target node S _d is as follows:

Wherein N is the number of track surfaces, and M is the number of satellites on one track surface;

Then at most common from S _s to S _d A minimum hop path with hop distance delta x + delta y; the set of nodes on the paths is a candidate transmission path node set, and the minimum hop count path set is a candidate transmission path set.

If the constellation without reverse slots is adopted, the path in the x-axis direction in the minimum hop path is as follows:

(a) If |x _d-x_s|≤N-|x_d-x_s |, the path in the x-axis direction is:

When x _s>x_d, x _s→x_s-1→x_s-2→…→x_d;

when x _s≤x_d, x _s→x_s+1→x_s+2→…→x_d;

(b) If |x _d-x_s|>N-|x_d-x_s |, the path in the x-axis direction is:

When x _s≤x_d, x _s→x_s-1→x_s-2→…→1→N→…→x_d;

When x _s>x_d, x _s→x_s+1→x_s+2→…→N→1→…→x_d.

Otherwise if there is a constellation with reverse slots:

When x _s>x_d, x _s→x_s-1→x_s-2→…→x_d;

When x _s≤x_d, x _s→x_s+1→x_s+2→…→x_d.

The constellation without reverse slot and the constellation with reverse slot have the same path in the Y-axis direction, and are

(A) If |y _d-y_s|≤M-|y_d-y_s |, the path in the y-axis direction is:

Y _s→y_s-1→y_s-2→…→y_d when y _s>y_d;

Y _s→y_s+1→y_s+2→…→y_d when y _s≤y_d;

(b) If |y _d-y_s|>M-|y_d-y_s |, the path in the y-axis direction is:

Y _s→y_s-1→y_s-2→…→1→M→…→y_d when y _s≤y_d;

Y _s→y_s+1→y_s+2→…→M→1→…→y_d when y _s>y_d.

4. For each of the sub-tasks after sequencing:

For each satellite node S _k (k=1..once., L), k=1..once., L, S ₁ is the source node S _s when k=1, S _L is the target node S _d when k=l, the following process is performed:

The size of the computing resources allocated to the jth subtask for node S _k, The available computing resource size for node S _k;

For the maximum available bandwidth in the set of all candidate transmission paths from the source node S _s to the node S _k, the maximum available bandwidth is determined by the inter-satellite link available bandwidth in step 12), To allocate communication bandwidth for the j-th subtask on the candidate transmission path from the source node S _s to the node S _k,In order to place the jth subtask in the node S _k for execution, the switch is the total time delay of satellite node data exchange on the transmission path from the source node S _s to the node S _k, the size (D _j) is the size of input data to be processed by the jth subtask, task is a set of all subtasks, and c _{task_j} is the amount of computation required for processing the jth subtask;

If it is The j-th subtask is scheduled to be executed on node S _k and the system state is updated as follows:

Task = Task- { subtask j }

For all nodes S _i on the maximum bandwidth path from node S _s to node S _k:

otherwise, the j-th subtask is arranged to be executed by the source node, and the system state is updated as follows:

Task = Task- { subtask j }

And repeating the subtask processing process until all the subtasks are arranged.

An example scenario is shown in fig. 2, which is a part of a constellation, the entire constellation comprising 24 orbital planes of 24 satellites each. Wherein satellite 51611 generates data, which is processed for transmission to its subordinate nodes via satellite 51712.

The low-delay service space-based calculation method comprises the following steps:

11 Task decomposition

The task T to be transmitted of the satellite 51611 is target identification in the remote sensing image, and the input data D is 2000×10000×8bit remote sensing data. The T is decomposed into 2 subtasks { Ta ₁,Ta₂ }, which together complete task T. The data size of each subtask is 2000 x 5000 x 8 bit=80 Mbit, and the calculation amount required by each subtask is 100 gfps. When each subtask of T is completed, then task T is completed.

12 Network state acquisition for path acquisition and computing node selection is required as follows

All satellites S _i in the low-orbit network periodically report own network load to the high-orbit satellites, and the report content comprises available bandwidths B _i＝{B_i,1,B_i,2,B_i,3,B_i,4 Mbps of four inter-satellite links and available computing power C _i Mflops of the satellites.

The high orbit satellite periodically broadcasts the low orbit satellite network status.

The current computing power and communication power of each satellite are shown in tables 1 and 2.

Table 1 satellite node computing capability C

Satellite identification	Currently available computing power
		Satellite51611	6Gflops
Satellite51612	10Gflops
		Satellite51613	5Gflops
Satellite51710	4Gflops
		Satellite51711	11Gflops
Satellite51712	7Gflops

TABLE 2 available inter-satellite Bandwidth B

13 A) a calculation node and a transmission path selection method as follows

According to the network state obtained in the step 2) and the decomposed subtasks and subtask requirements in the step 1), sequencing the subtasks, arranging the tasks on candidate transmission paths according to the sequencing, and mapping the subtasks to corresponding satellite nodes;

14 Transmission and calculation)

And 3) according to the arrangement result of the 3), transmitting the task data to be processed corresponding to the subtasks to the corresponding satellite nodes, loading a processing program according to the subtask calculation requirement by the satellite nodes, processing the received subtask data and transmitting the processing result to the target nodes. The processing result is set as high priority service and is sent through reserved bandwidth.

15 Subtask convergence

And after each subtask is correctly executed, feeding back to the task source node, and for the task which fails to be executed. And the task source node redistributes the transmission calculation process until the calculation result is correct, and all the subtasks are assembled at the target node after being correctly executed, so that the transmission calculation of the whole task is completed.

2. The sequencing method described in step 1) results in the following steps:

Lambda ₁＝λ₂ in this example, i.e. ordered { Ta ₁,Ta₂ }.

3. In accordance with the method of the present invention, the set of candidate transmission path nodes is { Satellite51611, satellite51612, satellite51613, satellite51710, satellite51711, satellite51712},

The candidate transmission paths are:

Satellites 51611 to 51712 in this example The shortest path, hop count is 3:

Satellite51611->Satellite51612->Satellite51613->Satellite51712,

Satellite51611->Satellite51612->Satellite51711->Satellite51712

Satellite51611->Satellite51710->Satellite51711->Satellite51712

4. the arrangement method according to step 1 of the present invention results in the following:

Considering that the propagation delay is in the order of ms and the computation and transmission delays are in the order of seconds, the propagation delay is ignored in this embodiment.

The subtask Ta1 is assigned to each node by Satellite51611, and the calculated Δt is shown in Table 3, respectively

TABLE 3 task 1 latency

Satellite	Δt
		Satellite51612	-9.52380952380953
Satellite51613	11.3725490196078
		Satellite51710	22.8205128205128
Satellite51711	-8.48484848484848
		Satellite51712	2.51082251082251

Because ofSubtask Ta1 is arranged to the node Satellite51612 for computation. The update information is:

Table 4 satellite node computing power

TABLE 5 available bandwidth and propagation delay between satellites

The Satellite51611 distributes the subtask Ta2 to each node, and the calculated Δt is respectively

TABLE 6 task 2 latency

Satellite	Δt
		Satellite51613	11.3725490196078
Satellite51710	22.8205128205128
		Satellite51711	-8.48484848484848
Satellite51712	2.51082251082251

Because ofSo this subtask Ta2 is assigned to Satellite 51711.

The total time delay of task execution obtained by the invention is max {100/5+80/21, 100/5.5+50/12} = 23.810 seconds.

The above embodiment illustrates that compared with the traditional independent processing mode of computing communication, the invention can effectively reduce service delay.

What is not described in detail in the present specification belongs to the known technology of those skilled in the art.

Claims (7)

1. A low-delay service space-based calculation method is characterized by comprising the following steps:

1) Setting a to-be-transmitted processing task of a task source node satellite S _s as a task T (D, R), wherein D is input data to be processed by the task T, and R is output data processed by the task T; decomposing the T into a plurality of parallel subtasks, and completing the task T when all the subtasks of the T are completed;

2) Each satellite in the low-orbit network periodically reports the network state of the satellite to the high-orbit satellite, and the high-orbit satellite periodically broadcasts the network state of the low-orbit satellite;

Each satellite in the low-orbit network has 2 front and back in orbit and 2 total 4 inter-satellite links between the left and right in orbit, and the content reported by the ith satellite S _i comprises an available bandwidth set B _i Mbps of four inter-satellite links and the current available computing resource per se Mflops; b _i＝{B_i,1,B_i,2,B_i,3,B_i,4, wherein B _i,1 is the available bandwidth of the first inter-satellite link of the ith satellite S _i, B _i,2 is the available bandwidth of the second inter-satellite link of the ith satellite S _i, B _i,3 is the available bandwidth of the third inter-satellite link of the ith satellite S _i, and B _i,4 is the available bandwidth of the fourth inter-satellite link of the ith satellite S _i;

3) Determining a candidate transmission path node set according to the low orbit satellite network state, sorting the subtasks according to the subtasks and the subtask attributes decomposed in the step 1), arranging the subtasks on the candidate transmission path node set according to the sorting, and mapping the subtasks to corresponding satellite nodes;

4) According to the arrangement result, the input data to be processed of the subtasks is sent to the corresponding satellite nodes, the satellite nodes load processing programs according to the subtask calculation requirements, the received subtask data are processed, the processing result is sent to the target nodes, the processing result is set to be a high-priority service, and the high-priority service is sent through reserved bandwidth;

5) And after each subtask is correctly executed, feeding back to the task source node, and for the task with failed execution, reassigning the transmission calculation process by the task source node until the calculation result is correct, and converging all subtasks at the target node after the subtasks are correctly executed, so as to complete the transmission calculation of the whole task.

2. The method for low-latency service space-based computation according to claim 1, wherein in the step 3), the subtasks are ordered according to the following method:

Let the order factor of the j-th subtask be lambda _j,

Taking lambda _j＝size(D_j)/size(D_max)-c_{task_j}/c_max, wherein D _j is input data to be processed of the jth subtask, size (D _j) is the size of the input data to be processed of the jth subtask, D _max is the input data to be processed corresponding to the subtask with the largest size of the input data to be processed in the subtask set, c _{task_j} is the calculated amount required for processing the jth subtask, c _max is the calculated amount of the subtask with the largest calculated amount required in the subtask set, and lambda _j is a task ordering factor;

each subtask is ordered from big to small by an ordering factor.

3. The method for calculating the low-latency traffic space base according to claim 1, wherein in the step 3), the method for determining the candidate transmission path node set is as follows:

The low-orbit satellite network is abstracted into a 2-dimensional mesh network, all nodes contained in a path set with the minimum hop count from a task source node S _s to a target node S _d form a candidate transmission path node set, and the total number of nodes in the set is L.

4. The method for calculating the low-delay service space-based according to claim 3, wherein a set of candidate transmission path nodes from a task source node S _s to a target node S _d is set as S= { S _i }, the low-orbit satellite network is abstracted as a 2-dimensional mesh network, in a 2-dimensional mesh network topology, an inter-orbit link is recorded as an x-axis direction, and an in-orbit link is recorded as a y-axis direction; for a certain communication, setting the coordinates of a task source node S _s as (x _s,y_s) and the coordinates of a target node S _d as (x _d,y_d);

If the constellation without reverse slots is adopted, the logical distance difference between the x and y direction communication between the task source node S _s and the target node S _d is as follows:

If the constellation is a constellation with reverse slots, the difference of the communication logical distances in the x and y directions between the task source node S _s and the target node S _d is as follows:

Wherein N is the number of track surfaces, and M is the number of satellites on one track surface;

Then at most common from S _s to S _d A minimum hop path with hop distance delta x + delta y; the set of nodes on the paths is a candidate transmission path node set, and the minimum hop count path set is a candidate transmission path set.

5. The method for calculating the space-based on the low-latency service according to claim 4, wherein if the constellation is a constellation without reverse slots, the path in the x-axis direction in the minimum hop path is:

(a) If |x _d-x_s|≤N-|x_d-x_s |, the path in the x-axis direction is:

When x _s＞x_d, x _s→x_s-1→x_s-2→…→x_d;

when x _s≤x_d, x _s→x_s+1→x_s+2→…→x_d;

(b) If |x _d-x_s|＞N-|x_d-x_s |, the path in the x-axis direction is:

When x _s≤x_d, x _s→x_s-1→x_s-2→…→1→N→…→x_d;

when x _s＞x_d, x _s→x_s+1→x_s+2→…→N→1→…→x_d;

Otherwise if there is a constellation with reverse slots:

When x _s＞x_d, x _s→x_s-1→x_s-2→…→x_d;

When x _s≤x_d, x _s→x_s+1→x_s+2→…→x_d.

6. The method of claim 5, wherein the constellation without reverse slots and the constellation with reverse slots have the same paths in the Y-axis direction, and are all

(A) If |y _d-y_s|≤M-|y_d-y_s |, the path in the y-axis direction is:

Y _s→y_s-1→y_s-2→…→y_d when y _s＞y_d;

Y _s→y_s+1→y_s+2→…→y_d when y _s≤y_d;

(b) If |y _d-y_s|＞M-|y_d-y_s |, the path in the y-axis direction is:

Y _s→y_s-1→y_s-2→…→1→M→…→y_d when y _s≤y_d;

Y _s→y_s+1→y_s+2→…→M→1→…→y_d when y _s＞y_d.

7. The method for low-latency service day base calculation according to claim 6, wherein in the step 3), the subtasks are organized as follows:

For each of the sub-tasks after sequencing:

For each satellite node S _k, k=1,..l, S ₁ is the source node S _s when k=1, S _L is the target node S _d when k=l, the following process is performed:

The size of the computing resources allocated to the jth subtask for node S _k, The available computing resource size for node S _k;

For the maximum available bandwidth in the set of all candidate transmission paths from the source node S _s to the node S _k, the maximum available bandwidth is determined by the inter-satellite link available bandwidth in step 2), To allocate communication bandwidth for the j-th subtask on the candidate transmission path from the source node S _s to the node S _k,In order to place the jth subtask in the node S _k for execution, the switch is the total time delay of satellite node data exchange on the transmission path from the source node S _s to the node S _k, the size (D _j) is the size of input data to be processed by the jth subtask, task is a set of all subtasks, and c _{task_j} is the amount of computation required for processing the jth subtask;

If it is The j-th subtask is scheduled to be executed on node S _k and the system state is updated as follows:

Task = Task- { subtask j }

For all nodes S _i on the maximum bandwidth path from node S _s to node S _k:

otherwise, the j-th subtask is arranged to be executed by the source node, and the system state is updated as follows:

Task = Task- { subtask j }