Method for scheduling multiple paths with fixed time slots and bandwidth reservation based on high performance networks

Abstract

The application relates to a multi-path fixed time slot and bandwidth reservation scheduling method based on a high-performance network. According to the reservation bandwidth condition in the network resource, the scheduling request sequence is dynamically adjusted to specify a more reasonable scheduling sequence. On the basis of the single-path reservation scheme, the double-path reservation scheme is supplemented, the number of successful request reservations is increased, the requirement for the network bandwidth is reduced, and the reservation bandwidth resource rate is increased.

Description

Technical Field

[0001] This application relates to the field of high-performance network technology, and more specifically, to a multi-path fixed time slot and bandwidth reservation scheduling method based on high-performance networks. Background Technology

[0002] Currently, large-scale applications of high-performance networks (HPNs) are used in engineering, computer science, and business, generating massive amounts of data every moment. These vast amounts of data typically need to be transmitted to designated sites for information storage and data analysis. In recent years, to meet the ever-increasing demand for data transmission, HPNs have been rapidly developed and deployed, and their efficient and reliable transmission capabilities have been widely recognized.

[0003] High-performance networks built on SDN (Software Defined Network) architecture technology have the ability to reserve bandwidth in advance and provide dedicated data transmission channels. Bandwidth reservation, as a core function of the control plane in SDN, is responsible for calculating network paths and allocating bandwidth resources for data transmission. Therefore, the bandwidth scheduler in a high-performance network plays a crucial role in providing Quality of Service (QoS). In this respect, bandwidth scheduling is very important for the efficiency and reliability of data transmission, and many scheduling algorithms have been proposed to optimize network performance. Under the constraints of data transmission, it is necessary to meet the corresponding transmission requirements while considering the rational allocation of network resources. Based on this, researchers have proposed numerous network user models and bandwidth scheduling algorithms to optimize various aspects of network performance, thereby improving data transmission efficiency and network resource utilization.

[0004] In real-world applications, user requests are highly personalized and diverse. Bandwidth scheduling algorithms often focus on optimizing specific network performance parameters, such as scheduling success rate, earliest data transmission completion time, shortest data transmission duration, user satisfaction rate, and data transmission completion rate. For these specific performance parameters, bandwidth scheduling must ensure data transmission is completed within constraints while fully utilizing network resources, optimizing network performance, and improving bandwidth utilization. Therefore, existing bandwidth scheduling methods cannot meet the personalized and diverse needs of user requests. Summary of the Invention

[0005] To overcome at least one deficiency in the prior art, this application provides a multi-path fixed time slot and bandwidth reservation scheduling method based on high-performance networks.

[0006] Firstly, a multi-path fixed time slot and bandwidth reservation scheduling method based on high-performance networks is provided, including:

[0007] Step 1: Obtain m requests within a scheduling period; the parameter information for each request includes the source node. Target node Start time of reserved bandwidth End time of reserved bandwidth Reserved bandwidth size B r r is the request number, r = 1, 2, ..., m;

[0008] Step 2: Use a single-path scheme to reserve bandwidth for m requests. If bandwidth reservation is successful for all m requests, the process ends; if there are requests that fail to reserve bandwidth, proceed to step 3.

[0009] Step 3: Use a dual-path scheme to reserve bandwidth for requests that failed to be reserved.

[0010] In one embodiment, step 2 employs a single-path scheme to reserve bandwidth for m requests, including:

[0011] Step 21: All requests constitute an unscheduled request set Q1. Select one request from the unscheduled request set Q1 as the current request, and remove requests whose bandwidth in the initial network link is less than the reserved bandwidth B of the current request. r The updated network links are obtained from the original links;

[0012] Step 22: Determine if there is a path in the updated network links that satisfies the current request. If not, return to step 21 and select the next request from the unscheduled request set Q1 as the current request. If there is a path, determine the shortest path in the updated network links that satisfies the current request and calculate the hop count N of the shortest path. Return to step 21 and select the next request from the unscheduled request set Q1 as the current request. Obtain all requests that can find the shortest path in the updated network links, denoted as request set Q2, and obtain the hop count of the shortest path corresponding to each request in request set Q2.

[0013] Step 23: Calculate the weight α of each request in request set Q2. i , Let i be the request number in request set Q2, i = 1, 2, ..., I, and let I be the number of requests in request set Q2. i Let α be the number of hops in the shortest path corresponding to request i in request set Q2. i Let i be the weight of request i in request set Q2. The start time for reserving bandwidth for request i in request set Q2. B i is the end time of the reserved bandwidth of the request i in the request set Q2;

[0014] Step 24, the weight of each request in the request set Q2 is sorted, and the request with the largest weight is selected for bandwidth reservation; the selected request with the largest weight is deleted from the unscheduled request set Q1, and step 21 is returned until all requests in the unscheduled request set Q1 are successfully reserved bandwidth, or there is no path in the network link that satisfies the request with unsuccessful bandwidth reservation.

[0015] In one embodiment, the request with unsuccessful bandwidth reservation in step 3 is reserved bandwidth by adopting a double-path scheme, including:

[0016] Step 31, the request with unsuccessful bandwidth reservation constitutes a request set Q3, for each request j, j is the number of requests in the request set Q3, j=1, 2, …J, J is the number of requests in the request set Q3, the request j is divided into two sub-requests, which are the first sub-request j1 and the second sub-request j2; if the size B j of the reserved bandwidth of the request j is odd, the size B j1 of the reserved bandwidth of the first sub-request j1 is (B j +1) / 2, and the size B j2 of the reserved bandwidth of the second sub-request j2 is B j -B j1 ; if the size B j of the reserved bandwidth of the request j is even, the size B j1 of the reserved bandwidth of the first sub-request j1 is B j / 2, and the size B j2 of the reserved bandwidth of the second sub-request j2 is B j -B j1 ; the other parameter information of the first sub-request j1 and the second request j2 is the same as that of the request j;

[0017] Step 32, a request in the request set Q3 is selected as the current request j; the links with a bandwidth size smaller than the size B j1 of the reserved bandwidth of the first sub-request j1 of the current request j are removed to obtain an updated network link;

[0018] Step 33, it is determined whether there is a path satisfying the first sub-request j1 in the updated network link, if not, the next request in the request set Q3 is selected as the current request in step 32; if yes, the shortest path satisfying the first sub-request j1 in the updated network link and the hop number N j1and the link bandwidth occupied by the shortest path in the time slot is reduced by the size B of the reserved bandwidth of the first sub-request j1 j1 to obtain the updated link bandwidth occupied by the shortest path in the time slot;

[0019] Step 34: remove the link in the initial network link whose bandwidth size is less than the size B of the reserved bandwidth of the second sub-request j2 of the current request j, to obtain the updated network link; determine whether there is a path satisfying the second sub-request j2 in the updated network link, if not, restore the link bandwidth occupied by the shortest path in the time slot in step 33, and return to step 32 to select the next request in the request set Q3 as the current request; j1

[0020] If yes, determine the shortest path satisfying the second sub-request j2 in the updated network link, and the hop number N of the shortest path j2 ; calculate the weight α of the current request j j , α j = α j1 + α j2 , where α j1 is the weight of the first sub-request, is the start time of the reserved bandwidth of the first sub-request, is the end time of the reserved bandwidth of the first sub-request, α j2 is the weight of the second sub-request, is the start time of the reserved bandwidth of the second sub-request, is the end time of the reserved bandwidth of the second sub-request; return to step 32 to select the next request in the request set Q3 as the current request, obtain all the requests in which the shortest path can be found in the updated network link, and record them as the request set Q4, and obtain the weight of each request in the request set Q4;

[0021] Step 35: sort the weight of each request in the request set Q4, select the request with the largest weight to reserve bandwidth, delete the request with the largest weight in the request set Q3, and return to step 32 until all the requests in the request set Q3 are successfully reserved bandwidth, or there is no path in the network link satisfying the request in the request set Q3 whose bandwidth reservation is unsuccessful.

[0022] In the second aspect, a computer readable storage medium is provided, the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the above-mentioned multi-path fixed time slot and bandwidth reservation scheduling method based on high-performance network.

[0023] ​In a third aspect, a computer program product comprises computer programs / instructions which, when executed by a processor, implement the above-mentioned high-performance network-based multi-path fixed time slot and bandwidth reservation scheduling method.

[0024] Compared with the prior art, the present application has the following beneficial effects:

[0025] 1. The present application dynamically adjusts the order of scheduling requests according to the status of reserved bandwidth in network resources, to specify a more reasonable scheduling order.

[0026] 2. The present application supplements the double-path reservation scheme on the basis of the single-path reservation scheme, which not only increases the number of successful reservation requests, but also reduces the requirement for network bandwidth, while increasing the reserved bandwidth resource rate. BRIEF DESCRIPTION OF DRAWINGS

[0027] The present application can be better understood by referring to the following description in conjunction with the accompanying drawings, which form a part of this specification and are included to further illustrate and explain the principles of the application. In the drawings:

[0028] Figure 1 A flow chart of a high-performance network-based multi-path fixed time slot and bandwidth reservation scheduling method according to an embodiment of the present application is shown;

[0029] Figure 2 A network topology according to an embodiment of the present application is shown;

[0030] Figure 3 A comparison chart of the reserved bandwidth resource rate RBR of the present application and the Greedy and Pro-BTH algorithms in a network with 10 nodes and 40 links scheduling different numbers of requests is shown;

[0031] Figure 4 A comparison chart of the reserved bandwidth resource rate RBR of the present application and the Greedy and Pro-BTH algorithms in a network with 20 nodes and 80 links scheduling different numbers of requests is shown;

[0032] Figure 5 A comparison chart of the reserved bandwidth resource rate RBR of the present application and the Greedy and Pro-BTH algorithms in a network with 50 nodes and 200 links scheduling different numbers of requests is shown;

[0033] Figure 6 A comparison chart of the reserved bandwidth resource rate RBR of the present application and the Greedy and Pro-BTH algorithms in a network with 100 nodes and 400 links scheduling different numbers of requests is shown;

[0034] Figure 7The application and the Greedy, Pro-BTH two algorithms in the 200 nodes 800 link scale network scheduling different number of requests for reserved bandwidth resource rate RBR comparison chart is shown;

[0035] Figure 8 The application and the Greedy, Pro-BTH two algorithms in the different scale network scheduling different number of requests for reserved bandwidth resource rate RBR comparison chart is shown. DETAILED DESCRIPTION

[0036] In the following, exemplary embodiments of the application will be described with reference to the accompanying drawings. In the description, not all features of a practical embodiment are described in order to be clear and concise. It should be appreciated, however, that many embodiment-specific decisions can be made in the development of any such practical embodiment in order to achieve the specific goals of the developer, and these decisions can vary from embodiment to embodiment.

[0037] It should also be noted here that, in order to avoid obscuring the application due to unnecessary details, only the device structure closely related to the scheme according to the application is shown in the drawings, and other details not closely related to the application are omitted.

[0038] It should be understood that the application is not limited to the described embodiments by virtue of the following description with reference to the drawings. In this context, embodiments can be combined with each other, features can be replaced or borrowed between different embodiments, and one or more features can be omitted in an embodiment.

[0039] The embodiment of the application provides a multi-path fixed time slot and bandwidth reservation scheduling method based on a high-performance network, Figure 1 A flow chart of the multi-path fixed time slot and bandwidth reservation scheduling method based on a high-performance network according to the embodiment of the application is shown, referring to Figure 1 The method comprises the following steps:

[0040] Step 1, obtaining m requests in a scheduling period; the parameter information of each request comprises a source node destination node start time of reserved bandwidth end time of reserved bandwidth size B of reserved bandwidth r r is the number of requests, r = 1, 2,..., m;

[0041] Step 2, performing bandwidth reservation on the m requests by using a single-path scheme, if the bandwidth reservation of the m requests is successful, the method ends; if there is a request whose bandwidth reservation is not successful, step 3 is performed;

[0042] Step 3, bandwidth reservation is performed on the request whose bandwidth reservation is unsuccessful by using the dual-path scheme.

[0043] According to the reserved bandwidth in the network resource, the application dynamically adjusts the order of the scheduling request to specify a more reasonable scheduling order. On the basis of the single-path reservation scheme, the dual-path reservation scheme is supplemented, which increases the number of successful reservation requests and reduces the requirement for network bandwidth, and increases the reserved bandwidth resource rate.

[0044] In one embodiment, the bandwidth reservation is performed on the m requests by using the single-path scheme in step 2, which can include:

[0045] Step 21, all requests constitute an unscheduled request set Q1, a request in the unscheduled request set Q1 is selected as a current request, and the link whose bandwidth size is smaller than the reserved bandwidth size B of the current request in the initial network link is removed to obtain an updated network link. r

[0046] Step 22, it is determined whether there is a path satisfying the current request in the updated network link, if not, returning to step 21 to select the next request in the unscheduled request set Q1 as the current request; if yes, the shortest path satisfying the current request in the updated network link is determined, and the hop number N of the shortest path is calculated, returning to step 21 to select the next request in the unscheduled request set Q1 as the current request, obtaining all requests in the updated network link that can find the shortest path, denoted as a request set Q2, and obtaining the hop number of the shortest path corresponding to each request in the request set Q2.

[0047] Step 23, the weight α of each request in the request set Q2 is calculated i , i is the number of the request in the request set Q2, i = 1, 2, … I, I is the number of the request in the request set Q2, N i is the hop number of the shortest path corresponding to the request i in the request set Q2, α i is the weight of the request i in the request set Q2, T is the start time of the reserved bandwidth of the request i in the request set Q2, T is the end time of the reserved bandwidth of the request i in the request set Q2, and B i is the size of the reserved bandwidth of the request i in the request set Q2.

[0048] ​Step 24, sort the weight of each request in the request set Q2, select the request with the largest weight to perform bandwidth reservation; delete the selected request with the largest weight from the unscheduled request set Q1, return to step 21, until all requests in the unscheduled request set Q1 are successfully reserved bandwidth, or there is no path in the network link that meets the request with unsuccessful bandwidth reservation.

[0049] In this embodiment, the shortest path can be determined by using the improved Dijkstra algorithm. Here, the bandwidth reservation refers to updating the link bandwidth occupied by the shortest path in the time slot and updating the value of the reserved bandwidth resource, wherein the method for updating the link bandwidth occupied by the shortest path in the time slot is: subtracting the reserved bandwidth size B r of the request with the largest weight from the link bandwidth occupied by the shortest path in the time slot corresponding to the request with the largest weight, to obtain the updated link bandwidth occupied by the shortest path in the time slot; and the method for updating the value of the reserved bandwidth resource is: the value of the reserved bandwidth resource = the start time T e r of the request (the start time of the reserved bandwidth) * the reserved bandwidth size B r of the request.

[0050] In one embodiment, the bandwidth reservation of the request with unsuccessful bandwidth reservation in step 3 can include:

[0051] Step 31, the request with unsuccessful bandwidth reservation constitutes a request set Q3, for each request j, j is the number of the request in the request set Q3, j = 1, 2, … J, J is the number of requests in the request set Q3, the request j is divided into two sub-requests, which are a first sub-request j1 and a second sub-request j2; if the reserved bandwidth size B j of the request j is odd, the reserved bandwidth size B j1 of the first sub-request j1 is (B j +1) / 2, and the reserved bandwidth size B j2 of the second sub-request j2 is B j -B j1 ; if the reserved bandwidth size B j of the request j is even, the reserved bandwidth size B j1 of the first sub-request j1 is B j / 2, and the reserved bandwidth size B j2 of the second sub-request j2 is B j -B j1 ; the other parameter information of the first sub-request j1 and the second request j2 is the same as that of the request j;

[0052] Step 32, selecting one request in the request set Q3 as the current request j; eliminating the links in the initial network link whose bandwidth size is less than the reserved bandwidth size B of the first sub-request j1 of the current request j j1 , obtaining the updated network link;

[0053] Step 33, determining whether there is a path satisfying the first sub-request j1 in the updated network link, if not, returning to step 32 to select the next request in the request set Q3 as the current request; if yes, determining the shortest path satisfying the first sub-request j1 in the updated network link and the hop number N of the shortest path j1 , and subtracting the reserved bandwidth size B of the first sub-request j1 from the link bandwidth occupied by the shortest path in the time slot j1 , obtaining the link bandwidth occupied by the updated shortest path in the time slot;

[0054] Step 34, eliminating the links in the initial network link whose bandwidth size is less than the reserved bandwidth size B of the second sub-request j2 of the current request j j1 , obtaining the updated network link; determining whether there is a path satisfying the second sub-request j2 in the updated network link, if not, restoring the link bandwidth occupied by the shortest path in the time slot in step 33 and returning to step 32 to select the next request in the request set Q3 as the current request;

[0055] if yes, determining the shortest path satisfying the second sub-request j2 in the updated network link and the hop number N of the shortest path j2 ; calculating the weight α of the current request j j , α j = α j1 + α j2 , wherein α j1 is the weight of the first sub-request, is the start time of the reserved bandwidth of the first sub-request, is the end time of the reserved bandwidth of the first sub-request, α j2 is the weight of the second sub-request, is the start time of the reserved bandwidth of the second sub-request, is the end time of the reserved bandwidth of the second sub-request; returning to step 32 to select the next request in the request set Q3 as the current request, obtaining all the requests in the request set Q4 that can find the shortest path in the updated network link, and obtaining the weight of each request in the request set Q4;

[0056] Step 35, the weight value of each request in the request set Q4 is sorted, the request with the largest weight value is selected to perform bandwidth reservation, the request with the largest weight value is deleted in the request set Q3, and step 32 is returned until all requests in the request set Q3 are successfully reserved bandwidth, or there is no path in the network link to meet the request in the request set Q3 which fails in bandwidth reservation.

[0057] In this embodiment, the request with the largest weight value is taken to perform bandwidth reservation, that is, the link bandwidth occupied by the shortest path of the first sub-request in the time slot is updated, the link bandwidth occupied by the shortest path of the second sub-request in the time slot is updated, and the value of the reserved bandwidth resource is updated. The specific updating method is the same as the bandwidth reservation method in step 2, and will not be described again.

[0058] In the above embodiment, for the user request of fixed time slot and bandwidth, the dynamic sorting idea and multiple fixed bandwidth fixed paths are used in the high-performance network to perform bandwidth reservation on the request, so as to maximize the reserved bandwidth resource rate.

[0059] In order to better explain the method of the present application, Figure 2 The network topology structure according to the embodiment of the present application is shown, in which Figure 2 The user input request in Table 1 is transmitted on the network topology structure, and Table 1 includes four requests, BRR1, BRR2, BRR3 and BRR4; Table 2 is Figure 1 There are three links in the network topology structure, and the link bandwidth occupied by the three links in four time slots (0-1, 1-2, 2-3 and 3-4) is shown in Table 2.

[0060] Table 1

[0061] V s ]]> V d ]]> [TECHNICAL FIELD] s ]] [TECHNICAL FIELD] e ]] B (Gb / s) BRR1 a b 0 4 5 BRR2 a c 0 2 3 BRR3 b c 1 3 5 BRR4 a b 2 4 4

[0062] Table 2

[0063]

[0064] The shortest path of the four requests in Table 1 in the existing network bandwidth resource is calculated, and the weight value of each request is calculated, and the calculation result is as follows:

[0065] BRR1: lasting for four time slots, the reserved bandwidth size is 5, the shortest path hop count is 1, and the weight value is calculated as 20; BRR2: lasting for two time slots, the reserved bandwidth size is 3, the shortest path hop count is 1, and the weight value is calculated as 6; BRR3: lasting for two time slots, the reserved bandwidth size is 5, the shortest path hop count is 1, and the weight value is calculated as 10; BRR4: lasting for two time slots, the reserved bandwidth size is 4, the shortest path hop count is 1, and the weight value is calculated as 8;

[0066] From the comparison above, BRR 1 has the largest weight value, so bandwidth reservation is performed for the request, i.e., the link bandwidth of each time slot of the shortest path a-b is updated, for example, the time slot 0-1: the updated link bandwidth = 7-5 (the reserved bandwidth size of BRR 1) = 2, and the value of the reserved bandwidth resource (i.e., the transmission amount of data) is updated, and the value of the reserved bandwidth resource is the product of time slot 4 of BRR 1 and the reserved bandwidth size 5 of BRR 1, which is equal to 20.

[0067] The shortest path and the weight value of the remaining requests are recalculated in the updated bandwidth resource:

[0068] BRR2: lasting for 2 time slots, the reserved bandwidth size is 3, the number of shortest path hops is 1, and the weight value is calculated as 6; BRR3: lasting for 2 time slots, the reserved bandwidth size is 5, the number of shortest path hops is 1, and the weight value is calculated as 10; BRR4: lasting for 2 time slots, the reserved bandwidth size is 4, there is no shortest path, and the weight value is set as 0.

[0069] From the comparison above, BRR 3 has the largest weight value, so bandwidth reservation is performed for the request, i.e., the link bandwidth of each time slot of the shortest path b-c is updated, and the value of the reserved bandwidth resource (i.e., the transmission amount of data) is updated, and the value of the reserved bandwidth resource is the product of time slot 2 of BRR3 and the reserved bandwidth size 5 of BRR3, which is equal to 10.

[0070] The shortest path and the weight value of the remaining requests are recalculated in the updated bandwidth resource:

[0071] BRR2: lasting for 2 time slots, the reserved bandwidth size is 3, the number of shortest path hops is 1, and the weight value is calculated as 6; BRR4: lasting for 2 time slots, the reserved bandwidth size is 4, there is no shortest path, and the weight value is set as 0;

[0072] From the comparison above, BRR 2 has the largest weight value, so bandwidth reservation is performed for the request, i.e., the link bandwidth of each time slot of the shortest path a-c is updated, and the value of the reserved bandwidth resource (i.e., the transmission amount of data) is updated, and the value of the reserved bandwidth resource is the product of time slot 2 of BRR2 and the reserved bandwidth size 3 of BRR2, which is equal to 6.

[0073] After the single-path bandwidth reservation is completed, the remaining request BRR4 that is not reserved is divided into two sub-requests, and a double-path reservation scheme is attempted to recalculate the path and the weight value of the remaining requests in the updated bandwidth resource, i.e., BRR4 is divided into a first sub-request BRR4.1 and a second sub-request BRR4.2:

[0074] The first sub-request BRR4.1: lasting 2 time slots, reserving bandwidth size of 2, shortest path hop number of 1, and weight value calculation of 4; the second sub-request BRR4.2: lasting 2 time slots, reserving bandwidth size of 2, shortest path hop number of 2, and weight value calculation of 2; the weight value of BRR4 is the sum of the weight values of the two sub-requests, which is 6;

[0075] According to the above comparison, the weight value of BRR4 is the largest, so the bandwidth reservation is performed for the request BRR4, that is, the link bandwidth of each time slot of the shortest path a-b of the first sub-request and the shortest path a-c-b of the second sub-request is updated, and the value of the reserved bandwidth resource (i.e. the transmission amount of data) is updated, which is the product of the time slot 2 of BRR4 and the reserved bandwidth size 4 of BRR4, equal to 8.

[0076] Therefore, the four requests are successfully scheduled, and the reservation bandwidth resource rate is 100% of the total amount of reserved bandwidth resources.

[0077] Experimental verification:

[0078] In order to verify the performance of the method (MaxRBR) of the present application, simulation experiments are performed on different numbers of requests (as shown in Table 4) in networks of different sizes (as shown in Table 3), and the method is compared with Greedy and Pro-BTH algorithms. In the simulation experiment, the total time slot of one scheduling period is set to 60s, and the bandwidth size of each link is initialized to 10 Gb / s.

[0079] For one request, the source node and the destination node are randomly selected in the network nodes, and the two nodes are ensured to be different. The start time of the request is also randomly generated, but the duration of data transmission satisfies the Poisson distribution. The start time and the duration jointly determine the end time of the request, which satisfies the formula and Here D max is set to 60s, and the end time is limited not to exceed the last time point of the set scheduling period, i.e. 60. The random generation of the bandwidth size also satisfies the Poisson distribution: B max is set to 5 Gb / s, and in the above formula, x is a random variable between (0, 1].

[0080] Table 3

[0081]

[0082] Table 4

[0083]

[0084] Through experiments, it is obtained thatFigures 3-8 The results are shown in the following tables, wherein, Figure 3 Fig. 2 shows a comparison of the reserved bandwidth resource ratio RBR of the application and the Greedy and Pro-BTH algorithms in a network with 10 nodes and 40 links for different numbers of requests. Figure 4 Fig. 3 shows a comparison of the reserved bandwidth resource ratio RBR of the application and the Greedy and Pro-BTH algorithms in a network with 20 nodes and 80 links for different numbers of requests. Figure 5 Fig. 4 shows a comparison of the reserved bandwidth resource ratio RBR of the application and the Greedy and Pro-BTH algorithms in a network with 50 nodes and 200 links for different numbers of requests. Figure 6 Fig. 5 shows a comparison of the reserved bandwidth resource ratio RBR of the application and the Greedy and Pro-BTH algorithms in a network with 100 nodes and 400 links for different numbers of requests. Figure 7 Fig. 6 shows a comparison of the reserved bandwidth resource ratio RBR of the application and the Greedy and Pro-BTH algorithms in a network with 200 nodes and 800 links for different numbers of requests. Figure 8 Fig. 7 shows a comparison of the reserved bandwidth resource ratio RBR of the application and the Greedy and Pro-BTH algorithms in networks with different sizes for different numbers of requests. According to Figures 3-8 It can be seen that the method of the application is superior to the prior art in terms of reserved bandwidth resource ratio.

[0085] An embodiment of the application provides a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the above-mentioned multi-path fixed time slot and bandwidth reservation scheduling method based on a high-performance network.

[0086] An embodiment of the application provides a computer program product, which includes computer programs / instructions. The computer programs / instructions are executed by a processor to implement the above-mentioned multi-path fixed time slot and bandwidth reservation scheduling method based on a high-performance network.

[0087] In summary, the application has the following technical effects:

[0088] 1. The application dynamically adjusts the order of scheduling requests according to the status of the reserved bandwidth in the network resources, to specify a more reasonable scheduling order.

[0089] 2. The application supplements the double-path reservation scheme on the basis of the single-path reservation scheme, increases the number of successful reservation requests, reduces the requirement for network bandwidth, and increases the reservation bandwidth resource rate.

[0090] The above merely describes various embodiments of the application, but the protection scope of the application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the application, which should be covered within the protection scope of the application. Therefore, the protection scope of the application should be subject to the protection scope of the claims.

Claims

1. A multi-path fixed time slot and bandwidth reservation scheduling method based on high-performance networks, characterized in that, include: Step 1: Obtain m requests within a scheduling period; the parameter information of each request includes the source node. , destination node Start time of reserved bandwidth End time of reserved bandwidth The size of the reserved bandwidth , where r is the request number. ; Step 2: Use a single-path scheme to reserve bandwidth for the m requests. If bandwidth reservation is successful for all m requests, the process ends; if there are any requests that fail to reserve bandwidth, proceed to Step 3. Step 3: Use a dual-path scheme to reserve bandwidth for requests that failed to reserve bandwidth. In step 2, a single-path scheme is used to reserve bandwidth for the m requests, including: Step 21: All requests constitute an unscheduled request set Q1. Select one request from the unscheduled request set Q1 as the current request, and remove requests whose bandwidth in the initial network link is less than the reserved bandwidth of the current request. The updated network links are obtained from the original links; Step 22: Determine whether there is a path in the updated network link that satisfies the current request. If not, return to step 21 and select the next request from the unscheduled request set Q1 as the current request. If there is a path, determine the shortest path in the updated network link that satisfies the current request and calculate the hop count N of the shortest path. Return to step 21 and select the next request from the unscheduled request set Q1 as the current request. Obtain all requests in the updated network link that can find the shortest path, denoted as request set Q2, and obtain the hop count of the shortest path corresponding to each request in request set Q2. Step 23: Calculate the weight of each request in the request set Q2. , Let i be the request number in request set Q2, i = 1, 2, ..., I, and I be the number of requests in request set Q2. Let i be the number of hops in the shortest path corresponding to request i in request set Q2. Let i be the weight of request i in request set Q2. The start time for reserving bandwidth for request i in request set Q2. The end time for reserving bandwidth for request i in request set Q2. The amount of bandwidth reserved for request i in request set Q2; Step 24: Sort the weights of each request in the request set Q2, and select the request with the largest weight for bandwidth reservation; delete the selected request with the largest weight from the unscheduled request set Q1, and return to step 21 until all requests in the unscheduled request set Q1 have successfully reserved bandwidth, or there is no path in the network link that satisfies the request that failed to reserve bandwidth. Step 3, which employs a dual-path scheme to reserve bandwidth for requests that failed to be reserved, includes: Step 31: The requests that failed to reserve bandwidth constitute a request set Q3. For each request j, where j is the request number in request set Q3, j=1,2,…J, and J is the number of requests in request set Q3, request j is divided into two sub-requests, namely the first sub-request j1 and the second sub-request j2; if the reserved bandwidth size of request j... If the number is odd, the size of the reserved bandwidth for the first sub-request j1. The size of the reserved bandwidth for the second sub-request j2 If the reserved bandwidth size of request j If the number is even, the size of the reserved bandwidth for the first sub-request j1 The size of the reserved bandwidth for the second sub-request j2 The other parameters of the first sub-request j1 and the second sub-request j2 are the same as those of request j. Step 32: Select a request from the request set Q3 as the current request j; remove the reserved bandwidth of the first sub-request j1 whose bandwidth in the initial network link is less than that of the current request j. The updated network links are obtained from the original links; Step 33: Determine whether there exists a path in the updated network link that satisfies the first sub-request j1. If not, return to step 32 and select the next request from the request set Q3 as the current request. If so, determine the shortest path in the updated network link that satisfies the first sub-request j1 and the number of hops of the shortest path. The link bandwidth occupied by the shortest path in the time slot is subtracted from the reserved bandwidth of the first sub-request j1. The updated link bandwidth occupied by the shortest path in the time slot is obtained. Step 34: Eliminate the reserved bandwidth of the second sub-request j2 in the initial network link whose bandwidth is less than that of the current request j. The link is updated to obtain the updated network link; it is determined whether there is a path in the updated network link that satisfies the second sub-request j2. If not, the link bandwidth occupied by the shortest path in the time slot in step 33 is restored, and the process returns to step 32, where the next request is selected from the request set Q3 as the current request. If it exists, determine the shortest path in the updated network link that satisfies the second sub-request j2, and the hop count of the shortest path. ; Calculate the weight of the current request j , ,in, The weight of the first sub-request. , The start time for reserving bandwidth for the first sub-request. The end time for reserving bandwidth for the first sub-request. The weight of the second sub-request. , The start time for reserving bandwidth for the second sub-request. The end time for reserving bandwidth for the second sub-request; return to step 32, select the next request as the current request from the request set Q3, obtain all requests that can find the shortest path in the updated network link, denoted as request set Q4, and obtain the weight of each request in request set Q4; Step 35: Sort the weights of each request in the request set Q4, select the request with the largest weight for bandwidth reservation, delete the request with the largest weight in the request set Q3, and return to step 32 until all requests in the request set Q3 have successfully reserved bandwidth, or there is no path in the network link that satisfies the requests in the request set Q3 that have failed to reserve bandwidth.

2. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the multi-path fixed time slot and bandwidth reservation scheduling method based on a high-performance network as described in claim 1.

3. A computer program product, characterized in that, It includes a computer program / instruction, which, when executed by a processor, implements the multi-path fixed time slot and bandwidth reservation scheduling method based on high-performance networks as described in claim 1.

Images