Method, device, equipment and readable storage medium for constructing independent sets of interconnected networks

Abstract

The present application discloses a method, apparatus, device and readable storage medium for constructing an independent set of interconnected networks, and relates to the field of computer technology. The method for constructing an independent set of interconnected networks includes: obtaining an initial folding partitioning and switching cube network (FDSC n ) and create an initial folded split and exchange cube network (FDSC n ) is the node number of each processor node, where n = 2 d And d ≥ 1; According to the network dimension n and node label, the initial folding split and exchange cube network (FDSC n ) is divided into several target folding partitioning and exchange cube networks (FDSC2); a processor node is extracted from each target folding partitioning and exchange cube network (FDSC2) to form the initial folding partitioning and exchange cube network (FDSC n ) independent sets, where the processor nodes in the independent sets are not adjacent to each other. This solution reduces communication conflicts between processor nodes and improves the overall stability and reliability of the interconnection network.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, apparatus, device and readable storage medium for constructing an independent set of interconnected networks. Background Art

[0002] A high-performance parallel computer is a computer system whose processor counts are increasing as computer system performance continues to improve. A network formed by connecting several processors in a specific manner is called an interconnection network. An interconnection network can be represented by a simple graph G = (V, E), where V(G) represents the set of processor nodes in the graph and E(G) represents the set of edges. The processor nodes in graph G represent the processor nodes in the interconnection network, and the edges represent the links between the processor nodes.

[0003] Nodes in independent sets of an interconnected network do not interfere with each other. When allocating frequency resources, bandwidth, storage space, or computing power, resources are allocated to nodes that are not subject to competition and thus inefficient, achieving efficient resource utilization and optimal allocation. In an interconnected network, when multiple nodes simultaneously attempt to transmit data over the same communication path, communication conflicts can occur, leading to packet loss and increased transmission delays. By constructing independent sets, we ensure that no direct communication paths exist between certain nodes, preventing conflicts when these nodes simultaneously transmit data. Therefore, only nodes in independent sets are allowed to transmit data, avoiding conflicts between directly adjacent nodes. This keeps the communication conflict rate low even under heavy network loads, improving network stability and reliability.

[0004] Independent sets are a key concept in graph theory, playing a crucial role in the design, optimization, and security of interconnected networks. The study and application of independent sets can effectively improve network performance, ensure network security, and promote the efficient flow of information.

[0005] There is currently no information available on the folded divide-and-swap cube (FDSC) network. n ) is a method for constructing independent sets on . Summary of the Invention

[0006] The object of the present invention is to provide a method, apparatus, device and readable storage medium for constructing an independent set of interconnected networks by using a folded partitioning and switching cube network (FDSC) n) can ensure that resources are allocated to nodes that will not reduce efficiency due to mutual competition when allocating frequency resources, bandwidth, storage space or computing power. This can achieve effective utilization and optimal configuration of resources, which not only improves resource utilization but also improves the overall stability of the system.

[0007] In order to achieve the above object, the present invention provides the following technical solutions:

[0008] In a first aspect, the present invention provides a method for constructing an independent set of interconnected networks, wherein the interconnected network is a folded partitioning and switching cube network (FDSC). n ), an interconnection network is a network obtained by connecting several processors in a specific manner, and the method includes:

[0009] Get the initial folded split and exchange cube network (FDSC n ) and create an initial folded split and exchange cube network (FDSC n ) is the node number of each processor node, where n = 2 d and d≥1;

[0010] According to the network dimension n and node label, the initial folded split and exchange cube network (FDSC n ) is divided into several target folded partitioning and exchange cube networks (FDSC2);

[0011] A processor node is extracted from each target folding partitioning and switching cube network (FDSC2) to form the initial folding partitioning and switching cube network (FDSC n ), and the processor nodes in the independent set are not adjacent to each other.

[0012] In some embodiments, the initial folded partition and exchange cube network (FDSC) is converted into a n ) is divided into several target folded segmentation and exchange cube networks (FDSC2), including:

[0013] Determine whether n is equal to 2;

[0014] If n is equal to 2, the target folded partitioning and switching cube network (FDSC2) is obtained.

[0015] In some embodiments, the method further comprises:

[0016] If n is greater than 2, the initial folded split and exchange cube network (FDSC n ) is divided into several intermediate folding, splitting and switching cube networks

[0017] Let n=n / 2, and repeat the process of determining whether n is equal to 2 until n is equal to 2.

[0018] In some embodiments, the initial Folded Split and Switch Cube Network (FDSC) is converted into a n ) is divided into several intermediate folding, splitting and switching cube networks include:

[0019] The initial folded split and exchange cube network (FDSC n ) The node number in Identical processor nodes are grouped into the same central folding, splitting, and switching cube network. Get several intermediate folding, splitting and swapping cube networks

[0020] In some embodiments, a processor node is extracted from each target folded partitioning and switching cube network (FDSC2) to form an initial folded partitioning and switching cube network (FDSC n ), including:

[0021] The first processor node is extracted from each target folded partitioning and switching cube network (FDSC2) to form a candidate set;

[0022] Let k = 2, and determine whether k is less than or equal to log2 n;

[0023] If k is greater than log2n, the candidate set is used as the initial folded split and exchange cube network (FDSC n ) is an independent set.

[0024] In some embodiments, the method further comprises:

[0025] If k is less than or equal to log2 n, find the number of processor nodes in the candidate set in e(2 k )The associated nodes on the edge;

[0026] Delete the associated nodes belonging to the candidate set from the candidate set;

[0027] Let k=k+1, and repeatedly determine whether k is less than or equal to log2n until k is greater than log2n.

[0028] In a second aspect, the present invention further provides a device for constructing an independent set of interconnected networks, the device comprising:

[0029] Label creation module to obtain the initial folded split and exchange cube network (FDSC n ) and create an initial folded split and exchange cube network (FDSCn ) is the node number of each processor node, where n = 2 d and d≥1;

[0030] The network partitioning module is used to divide the initial folded partition and exchange cube network (FDSC) into n ) is divided into several target folded partitioning and exchange cube networks (FDSC2);

[0031] The set building module is used to extract a processor node from each target folding partitioning and switching cube network (FDSC2) to form the initial folding partitioning and switching cube network (FDSC n ), and the processor nodes in the independent set are not adjacent to each other.

[0032] In a third aspect, the present invention further provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for constructing an independent set of interconnected networks provided in the first aspect is implemented.

[0033] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for constructing an independent set of interconnected networks provided in the first aspect.

[0034] In a fifth aspect, the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the method for constructing an independent set of interconnected networks provided in the first aspect.

[0035] The beneficial effects of the present invention are:

[0036] First obtain the initial folding split and exchange cube network (FDSC n ) and create an initial folded split and exchange cube network (FDSC n ) in each processor node, and then the initial folded partitioning and switching cube network (FDSC) is converted into n ) is divided into several target folding partitioning and switching cube networks (FDSC2); finally, a processor node is extracted from each target folding partitioning and switching cube network (FDSC2) to form the initial folding partitioning and switching cube network (FDSC n) is an independent set. The processor nodes in the independent set are not adjacent to each other. In this way, the independent set contains all the processor nodes in the target folded partitioning and switching cube network (FDSC2). These processor nodes are not interconnected and will not interfere with each other. Therefore, when allocating frequency resources, bandwidth, storage space, or computing power, it can be ensured that resources are allocated to nodes that will not reduce efficiency due to mutual competition, thus achieving effective resource utilization and optimal configuration. In an interconnected network, when multiple processor nodes simultaneously attempt to transmit data through the same communication path, communication conflicts may occur, resulting in problems such as packet loss and increased transmission delay. By constructing an independent set, it can be ensured that there is no direct communication path between these processor nodes. When these processor nodes send data simultaneously, communication conflicts will not occur, thereby improving the stability and reliability of the network.

[0037] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and implement it according to the contents of the specification, the following is a detailed description of the preferred embodiments of the present invention with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 Schematic diagram of the structure of a four-dimensional partitioning and switching cube network (DSC4) according to an embodiment of the present invention;

[0039] Figure 2 Schematic diagram of the structure of a four-dimensional folded partitioning and switching cube network (FDSC4) according to one embodiment of the present invention;

[0040] Figure 3 A simple structural diagram of an interconnection network according to an embodiment of the present invention;

[0041] Figure 4 A schematic flow chart of a method for constructing an independent set of interconnected networks according to an embodiment of the present invention;

[0042] Figure 5 Schematic diagram of the structure of the initial folding, dividing and switching cube network (FDSC2) according to one embodiment of the present invention;

[0043] Figure 6 FIG4 is a schematic diagram of the structure of another four-dimensional folded partitioning and switching cube network (FDSC4) according to an embodiment of the present invention;

[0044] Figure 7 Schematic diagram of the structure of an eight-dimensional folded partitioning and switching cube network (FDSC8) according to one embodiment of the present invention;

[0045] Figure 8A schematic flow chart of another method for constructing an independent set of interconnected networks according to an embodiment of the present invention;

[0046] Figure 9 A schematic structural diagram of a device for constructing an independent set of interconnected networks according to an embodiment of the present invention;

[0047] Figure 10 A schematic diagram of the structure of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0048] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0049] It should be noted that references to "one embodiment," "an embodiment," "an example embodiment," etc., in this specification indicate that the described embodiment may include specific features, structures, or characteristics. However, not every embodiment must include these specific features, structures, or characteristics. In addition, such references do not necessarily refer to the same embodiment. Furthermore, when specific features, structures, or characteristics are described in conjunction with an embodiment, whether or not explicitly described, it is indicated that it is within the knowledge of those skilled in the art to incorporate such features, structures, or characteristics into other embodiments.

[0050] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0051] In order to better understand the technical solution of this application, the following first introduces the folded split and exchange cube network (FDSC n ) is described in detail.

[0052] For integer n=2 d and d ≥ 1, a divide-and-swap cube (DSC n Each processor node μ is represented by an n-bit binary string label, μ=μ1μ2...μ n (μ i ∈{0,1},1≤i≤n). For each 1≤k≤log2 n=d, the binary string label of the processor node μ can be divided into 3 parts, denoted as: μ=t1t2t3. Note that when k=1, then t3 is an empty string.

[0053] If d = 1, then n = 2 1 ,1≤k≤log2 n=1. t1=μ1, t2=μ2, μ=μ1μ2.

[0054] If d = 2, then n = 2 2 =4, 1≤k≤log2 n=2.

[0055] When k=1, t1=μ1μ2, t2=μ3μ4, μ=μ1μ2μ3μ4.

[0056] When k=2, t1=μ1, t2=μ2, t3=μ3μ4, μ=μ1μ2μ3μ4.

[0057] A processor node w is adjacent to μ if and only if one of the following two conditions holds:

[0058] (1) in It is the complement of μ1 (0, 1 are complements of each other), and the edge between w and μ is called the e(1) edge.

[0059] (2) At this time, the edge between w and μ is called side.

[0060] like Figure 1 As shown, processor nodes 1110 and 0110 differ in the first digit of the processor node label, which meets condition 1. 1110 and 0110 are connected, and the edge between them is called e(1) edge.

[0061] When k=1, t1=μ1μ2, t2=μ3μ4, μ=μ1μ2μ3μ4. For processor node μ=1110, t1=11, t2=10. t1≠t2, so μ is connected to w=1011. The edge between 1110 and 1011 is called the e(4) edge.

[0062] When k = 2, t1 = μ1, t2 = μ2, t3 = μ3μ4. For processor node μ = 1110, t1 = 1, t2 = 1, t3 = 10. t1 = t2, so μ is connected to w = 0010. The edge between 1110 and 0010 is called the e(2) edge.

[0063] Split and swap cube DSC n With 2 n processor nodes and 2 n-1 (d+1) edges, each processor node has d+1 neighbors, connected by d+1 edges. The d+1 edges are e(1) and Edge (1≤k≤d). For example, if d=2, then n=4, DSC n Each processor node in will be connected to one e(1) edge, one e(4) edge, and one e(2) edge.

[0064] For integer n=2 d and d≥1, a folded partition and swap cube (FDSC n ) is in the Divide and Swap Cube (DSC n ) is built by adding an edge to each processor node. n ) has 2 n processor nodes, for each processor node μ=μ1μ2...μ n , add an edge connection At this time, the edge between w and μ is called the e(f) edge.

[0065] like Figure 2 As shown, in FDSC4, for processor node 1110, it has an additional e(f) edge connected to processor node 1010.

[0066] Let all e(i) edges be denoted as E i , FDSC n Has the following properties:

[0067] (1)FDSC n It is the d+2 rule.

[0068] (2)FDSC n The number of processor nodes is 2 n , the number of sides is 2 n-1 (d+2)

[0069] (3)FDSC n Can be divided into indivual

[0070] (4)|E1|=|E2|=|E4|=...=|E n |=|E f |=2 n-1

[0071] like Figure 3 As shown, given an interconnection network represented by a simple graph G = (V, E), where V is a set of processor nodes, an independent set S is a subset of the processor node set V, satisfying that any two processor nodes in S are not adjacent in G. That is, for any two processor nodes u and v in S,

[0072] Figure 3 The independent set in can be S = {v2, v3}.

[0073] In some embodiments, as Figure 4 As shown in the figure, a method for constructing an independent set of interconnected networks is provided, wherein the interconnected network is a folded partitioned and switched cube network (FDSC n ), an interconnection network is a network obtained by connecting several processors in a specific manner, and the method includes:

[0074] S101, obtain the initial folded split and exchange cube network (FDSC n ) and create an initial folded split and exchange cube network (FDSC n ) is the node number of each processor node in.

[0075] Where n = 2 d And d ≥ 1, the initial folding partitioning and exchanging cube network is an n-dimensional folding partitioning and exchanging cube network, and the node label is the identifier of each processor node, which is used to represent each processor node.

[0076] Specifically, the initial folded split and exchange cube network (FDSC n ) as the initial folded split and exchange cube network (FDSC n ) dimension n, each processor node μ is represented by an n-bit binary string label, μ=μ1μ2...μ n (μ i ∈{0,1},1≤i≤n). For each 1≤k≤log2 n=d, the binary string label of the processor node μ can be divided into 3 parts, denoted as: μ=t1t2t3. Note that when k=1, then t3 is an empty string.

[0077] If d = 1, then n = 2 1 ,1≤k≤log2 n=1. t1=μ1, t2=μ2, μ=μ1μ2.

[0078] If d = 2, then n = 2 2 =4, 1≤k≤log2 n=2.

[0079] When k=1, t1=μ1μ2, t2=μ3μ4, μ=μ1μ2μ3μ4.

[0080] When k=2, t1=μ1, t2=μ2, t3=μ3μ4, μ=μ1μ2μ3μ4.

[0081] A processor node w is adjacent to μ if and only if one of the following two conditions holds:

[0082] (1) in It is the complement of μ1 (0, 1 are complements of each other).

[0083] (2)

[0084] Exemplarily, when n=2, the node identifiers of the processor nodes may be {00, 10, 11, 01}.

[0085] S102, according to the network dimension n and the node label, the initial folded split and exchange cube network (FDSC n ) is divided into several target folded segmentation and exchange cube networks (FDSC2).

[0086] Optionally, determine whether n is equal to 2; if n is equal to 2, stop the initial folding split and exchange cube network (FDSC n ) is divided to obtain the target folding partitioning and exchange cube network (FDSC2); if n is greater than 2, the initial folding partitioning and exchange cube network (FDSC n ) is divided into several intermediate folding, splitting and switching cube networks Let n=n / 2, and repeat the process of determining whether n is equal to 2 until n is equal to 2.

[0087] Specifically, all initial folded split and exchange cube networks (FDSC n ) are composed of several target folding segmentation and exchange cube networks (FDSC2). When n is equal to 2, it means that there is no need to continue the initial folding segmentation and exchange cube network (FDSC n ) to partition, initial folding partition and exchange cube network (FDSC n ) has been divided into the target folded partitioning and exchange cube network (FDSC2); when n is greater than 2, the initial folded partitioning and exchange cube network (FDSC n ) The node number in Identical processor nodes are grouped into the same central folding, splitting, and switching cube network. Get several intermediate folding, splitting and swapping cube networks Now let n = n / 2, and continue to judge whether n is equal to 2, until n is equal to 2, which proves that the initial folding partitioning and exchange cube network (FDSC n ) has been partitioned into multiple target folded segmentation and exchange cube networks (FDSC2).

[0088] For example, for an integer n=2 d and d≥2, a FDSC n The network can be divided into indivual For example, when n=8, the FDSC8 network is first divided into 16 FDSC4 networks, and the processor nodes are numbered according to the first The processor nodes 00000000, 01000000, 10000000, 110000000, 00010000, 01010000, 10010000, 11010000, 00100000, 01100000, 10100000, 11100000, 00110000, 01110000, 10110000, and 11110000, whose 5th to 8th bits are all 0000, will be assigned to the same FDSC4 network. Each FDSC4 network is then further divided into four FDSC2 networks, which are assigned to the corresponding FDSC2 networks based on bits 3 to 8 of the processor node number, for a total of 64 FDSC2 networks. For example, the processor nodes 00000000, 01000000, 10000000, 110000000 in the FDSC4 network above have the 3rd to 8th bits all as 000000, which will be divided into a FDSC2 network. And so on, the initial folding and switching cube network (FDSC n ) division.

[0089] S103, extract a processor node from each target folding partitioning and switching cube network (FDSC2) to form an initial folding partitioning and switching cube network (FDSC n ) is an independent set.

[0090] The processor nodes in the independent set are not adjacent to each other.

[0091] Optionally, the first processor node is extracted from each target folding partitioning and switching cube network (FDSC2) to form a candidate set; let k = 2, and determine whether k is less than or equal to log2 n; if k is greater than log2 n, the candidate set is used as the initial folding partitioning and switching cube network (FDSC n ) independent set; if k is less than or equal to log2 n, find the number of processor nodes in the candidate set in e(2 k ) edge; delete the associated nodes belonging to the candidate set from the candidate set; let k = k + 1, and repeatedly determine whether k is less than or equal to log2 n until k is greater than log2 n.

[0092] For example, when n=2, Figure 5 As shown, any processor node in FDSC2 is connected to every other processor node, and each processor node has d + 2 = 3 neighbors, i.e., 3 edges: e(1), e(2), and e(f). Therefore, for FDSC2, any set of processor nodes is an independent set of FDSC2, and the resulting independent set can be S = {00}.

[0093] When n=4, Figure 6 As shown, the FDSC4 network is first divided into four FDSC2 networks. These four FDSC2 networks are: {0000, 0100, 1000, 1100}, {0001, 0101, 1001, 1101}, {0010, 0110, 1010, 1110}, {0011, 0111, 1011, 1111}; then the first processor node is taken from each FDSC2 to form a candidate set. Each processor node in the FDSC2 network will have an e(4) edge connected to other FDSC2 networks. Therefore, if the e(4) edge of any processor node in the candidate set is connected to another processor node that also appears in the candidate set, then one of the processor nodes needs to be deleted, and the final candidate set is the required independent set of the FDSC4 network; the final independent set can be S = {0000, 0001, 0010, 0011}.

[0094] When n=8, Figure 7As shown, the FDSC8 network is divided into a network consisting of 64 FDSC2s. The 64 FDSC2s are: {00000000,01000000,10000000,11000000}, {00010000,01010000,10010000,11010000}, {00100000,01100000,10100000,11100000}, {00110000,01110000,10110000,11110000}, {000000001,01000001,10000001,11000000}, {00010001,01010001,1 0010001,11010001},{00100001,01100001,10100001,11100001},{00110001,01110001,10110001,11110001},{00000010,01000010,10000010,110 00010},{00010010,01010010,10010010,11010010},{00100010,01100010,10100010,11100010},{00110010,01110010,10110010,11110010},{000 00011,01000011,10000011,11000011},{00010011,01010011,10010011,11010011},{00100011,01100011,10100011,11100011},{00110011,01110 011,10110011,11110011},{00000100,01000100,10000100,11000100},{00010100,01010100,10010100,11010100},{00100100,01100100,1010010 0,11100100},{00110100,01110100,10110100,11110100},{00000101,01000101,10000101,11000101},{00010101,01010101,10010101,11010101} ,{00100101,01100101,10100101,11100101},{00110101,01110101,10110101,11110101},{00000110,01000110,10000110,11000110},{00010110,01010110,10010110,11010110},{00100110,01100110,10100110,11100110},{00110110,01110110,10110110,11110110},{00000111,01000111,10000111,11000111},{00010111,01010111,10010111,11010111},{00100111,01100111,10100111,11100111},{00110111,01110111,10110111,11110111},{00001000,01001000,10001000,11001000},{00011000,01011000,10011000,11011000},{00101000,01101000,10101000,11101000},{00111000,01111000,10111000,11111000},{00001001,01001001,10001001,11001001},{00011001,01011001,10011001,11011001},{00101001,01101001,10101001,11101001},{00111001,01111001,10111001,11111001},{00001010,01001010,10001010,11001010},{00011010,01011010,10011010,11011010},{00101010,01101010,10101010,11101010},{00111010,01111010,10111010,11111010},{00001011,01001011,10001011,11001011},{00011011,01011011,10011011,11011011},{00101011,01101011,10101011,11101011},{00111011,01111011,10111011,11111011},{00001100,01001100,10001100,11001100},{00011100,01011100,10011100,11011100},{00101100,01101100,10101100,11101100},{00111100,01111100,10111100,11111100},{00001101,01001101,10001101,11001101},{00011101,01011101,10011101,11011101},{00101101,01101101,10101101,1 1101101},{00111101,01111101,10111101,11111101},{00001110,01001110,10001110,11001110},{00011110,01011110,10011110,11011110},{ 00101110,01101110,10101110,11101110},{00111110,011111110,101111110,11111110},{00001111,01001111,10001111,11001111},{00101111,01101111,10101111,11101111},{00111111,0111111,10111111,1111111}, each of the FDSCs in the 282 network contains The processor node will have e(4) edges and e(8) edges connected to other FDSC2 networks. In each FDSC2, the first processor node is taken to form a candidate set. If the processor node connected by the e(4) edges and e(8) edges of any processor node in the candidate set is also included in the candidate set, then one of the processor nodes connected by the e(4) edges and e(8) edges needs to be deleted. The final candidate set is the required independent set of the FDSC8 network. Finally, the independent set of the FDSC8 network is S={00000011,00110101,00001100,00001101,00101110,00111111,00110100,001 01111,00010011,00011101,00010010,00011100,00101101,00111000,00001110,00000001,00000000,00101000,00110011,00111100,00100110,0 0111101,00101001,00011011,00010001,00100100,00111110,00100101,00000010,00000111,00000110,00001001,00101010,00101011,00011010,00010110,00011000,00010111,00011001,00100111,00110010,00001010,0000101,00000101,00000100,00001111,00100010,00110110,00111001,00110111,00101100,00011110,00010101,00010100,00011111,00111010,00111011,00001000}. ,

[0095] From FDSC n Take the first processor node from each FDSC2 to form a candidate set. If any two processor nodes in the candidate set have e(2 k ) edges (2≤k≤log2 n), then delete one of the processor nodes in the candidate set, and the final candidate set is the required FDSC n Independent set of networks.

[0096] The above-mentioned process of S102-S103 can be implemented based on a program algorithm. The program algorithms corresponding to some embodiments are as follows:

[0097]

[0098]

[0099]

[0100]

[0101] In the above embodiment, the initial folded partitioning and switching cube network (FDSC) is first obtained. n ) and create an initial folded split and exchange cube network (FDSC n ) in each processor node, and then the initial folded partitioning and switching cube network (FDSC) is converted into n ) is divided into several target folding partitioning and switching cube networks (FDSC2); finally, a processor node is extracted from each target folding partitioning and switching cube network (FDSC2) to form the initial folding partitioning and switching cube network (FDSC n) independent sets of processor nodes. These processor nodes are not interconnected and are not subject to mutual interference. Therefore, when allocating frequency resources, bandwidth, storage space, or computing power, resources can be allocated to nodes that will not be degraded by competition, thus achieving efficient resource utilization and optimal allocation. In an interconnected network, when multiple processor nodes simultaneously attempt to transmit data through the same communication path, communication conflicts may occur, leading to packet loss, increased transmission delays, and other problems. By constructing independent sets, it is possible to ensure that there are no direct communication paths between these processor nodes. When these processor nodes send data simultaneously, communication conflicts will not occur, thus improving network stability and reliability.

[0102] In order to more comprehensively demonstrate this solution, this embodiment provides an optional method for constructing an independent set of interconnected networks, such as Figure 8 As shown:

[0103] S201, obtain the initial folding division and exchange cube network (FDSC n ) and create an initial folded split and exchange cube network (FDSC n ) is the node number of each processor node in.

[0104] Where n = 2 d And d≥1.

[0105] S202, determine whether n is equal to 2; if so, execute S203, if not, execute S204.

[0106] S203, obtaining a target folded segmentation and exchange cube network (FDSC2).

[0107] S204, the initial folded split and exchange cube network (FDSC n ) The node number in Identical processor nodes are grouped into the same central folding, splitting, and switching cube network. Get several intermediate folding, splitting and swapping cube networks

[0108] S205, set n = n / 2, and return to execute S202.

[0109] S206 , extracting the first processor node from each target folded partitioning and switching cube network (FDSC2) to form a candidate set.

[0110] S207, set k=2.

[0111] S208, determine whether k is less than or equal to log2 n, if so, execute S209, if not, execute S212.

[0112] S209, find the number of processor nodes in the candidate set in e(2 k ) edge of the associated node.

[0113] S210: Deleting associated nodes belonging to the candidate set from the candidate set.

[0114] S211, set k=k+1, and return to execute S208.

[0115] S212, the candidate set is used as the initial folding split and exchange cube network (FDSC n ) is an independent set.

[0116] The specific process of the above S201-S212 can be found in the description of the above method embodiment. The implementation principle and technical effects are similar and will not be repeated here.

[0117] Based on the same inventive concept, embodiments of the present application also provide an apparatus for constructing an independent set of interconnected networks, which is used to implement the aforementioned method for constructing an independent set of interconnected networks. The solution provided by this apparatus is similar to the solution described in the aforementioned method. Therefore, the specific limitations in the embodiments of one or more apparatuses for constructing an independent set of interconnected networks provided below can be found in the aforementioned definition of the method for constructing an independent set of interconnected networks, and will not be further elaborated here.

[0118] In one embodiment, Figure 9 As shown, a device for constructing an independent set of interconnected networks is provided, the device comprising:

[0119] The label creation module 30 is used to obtain the initial folded partitioning and switching cube network (FDSC n ) and create an initial folded split and exchange cube network (FDSC n ) is the node number of each processor node, where n = 2 d and d≥1;

[0120] The network partitioning module 31 is used to divide the initial folded partitioned and exchanged cube network (FDSC) into n ) is divided into several target folded partitioning and exchange cube networks (FDSC2);

[0121] The set building module 32 is used to extract a processor node from each target folding partitioning and switching cube network (FDSC2) to form an initial folding partitioning and switching cube network (FDSC n ), and the processor nodes in the independent set are not adjacent to each other.

[0122] In another embodiment, the above Figure 9 The network partitioning module 31 is specifically used to:

[0123] Determine whether n is equal to 2;

[0124] If n is equal to 2, the target folded partitioning and switching cube network (FDSC2) is obtained.

[0125] If n is greater than 2, the initial folded split and exchange cube network (FDSC n ) is divided into several intermediate folding, splitting and switching cube networks

[0126] Specifically, the initial folded split and exchange cube network (FDSC n ) The node number in Identical processor nodes are grouped into the same central folding, splitting, and switching cube network. Get several intermediate folding, splitting and swapping cube networks

[0127] Let n=n / 2, and repeat the process of determining whether n is equal to 2 until n is equal to 2.

[0128] In another embodiment, the above Figure 9 The set establishment module 32 is specifically used for:

[0129] The first processor node is extracted from each target folded partitioning and switching cube network (FDSC2) to form a candidate set;

[0130] Let k = 2, and determine whether k is less than or equal to log2 n;

[0131] If k is greater than log2 n, the candidate set is used as the initial folded split and exchange cube network (FDSC n ) is an independent set.

[0132] If k is less than or equal to log2 n, find the number of processor nodes in the candidate set in e(2 k )The associated nodes on the edge;

[0133] Delete the associated nodes belonging to the candidate set from the candidate set;

[0134] Let k = k + 1, and repeat the process of determining whether k is less than or equal to log2 n until k is greater than log2 n.

[0135] The present application also provides an electronic device, in some embodiments, referring to Figure 10As shown, electronic device 700 includes an input unit 710, a memory 720, a processor 730, and an output unit 740. Memory 720 stores program instructions that can be executed by processor 730. Processor 730 invokes the program instructions to execute the method and / or technical solution for constructing an independent set of interconnected networks based on the aforementioned embodiments. The electronic device 700 can be a mobile terminal device such as a mobile phone or a computer.

[0136] In addition, an embodiment of the present application further provides a computer-readable storage medium for storing a computer program for executing a method for constructing an independent set of interconnected networks. For example, computer program instructions, when executed by a computer, can call or provide the method and / or technical solution according to the present application through the operation of the computer. The program instructions for calling the method of the present application may be stored in a fixed or removable storage medium, and / or transmitted via a data stream in a broadcast or other signal-carrying medium and / or stored in a storage medium that operates according to the program instructions.

[0137] Obviously, those skilled in the art should understand that the modules or steps of the present application described above can be implemented using a general-purpose computing device. They can be concentrated on a single computing device or distributed across a network consisting of multiple computing devices. Alternatively, they can be implemented using program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be fabricated into separate integrated circuit modules, or multiple modules or steps can be fabricated into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0138] The various technical features of the above embodiments can be arbitrarily integrated. To make the description concise, not all possible integrations of the various technical features in the above embodiments are described. However, as long as there is no contradiction in the integration of these technical features, they should be considered to be within the scope of this specification.

[0139] The above embodiments merely illustrate several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent. It should be noted that a person skilled in the art would be able to make various modifications and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the patent for this invention shall be determined by the appended claims.

Claims

1. A method for constructing an independent set of interconnected networks, characterized in that: The interconnection network is a folded split and switched cube network ( ), wherein the interconnection network is a network obtained by connecting a plurality of processors in a specific manner, and the method comprises: Get the initial fold split and swap cube network ( ) and create an initial folded split and swapped cube network ( ) is the node number of each processor node in, where and ; The node label is an n-bit binary string; According to the network dimension n and the node labels, the initial folded split and exchange cube network ( ) is divided into several target folding segmentation and exchange cube networks ( ); According to the network dimension n and the node labels, the initial folded split and exchange cube network ( ) is divided into several target folding segmentation and exchange cube networks ( ),include: Determine whether n is equal to 2; If n is equal to 2, the target folding segmentation and exchange cube network is obtained ( ); If n is greater than 2, the initial folding split and exchange cube network ( ) is divided into several intermediate folding partitioning and exchange cube networks ; Let n = n / 2, and repeat the process of checking whether n is equal to 2 until n is equal to 2; According to the node labels, the initial folded split and exchange cube network ( ) is divided into several intermediate folding partitioning and exchange cube networks ,include: The initial folded split and exchanged cube network ( ) The node number in +1~n identical processor nodes are divided into the same intermediate folding, splitting and switching cube network , we get several intermediate folding, splitting and swapping cube networks ; Folding split and exchange cube networks from each objective ( ) are extracted to form the initial folding, splitting and switching cube network ( ), where the processor nodes in the independent set are not adjacent to each other.

2. The method for constructing an independent set of interconnected networks according to claim 1, wherein: Folding split and exchange cube networks from each objective ( ) are extracted to form the initial folding, splitting and switching cube network ( ), including: Folding split and exchange cube networks from each objective ( ) extract the first processor node to form a candidate set; Let k=2, and determine whether k is less than or equal to ; If k is greater than , then the candidate set is used as the initial folding split and exchange cube network ( ) of the independent set.

3. The method for constructing an independent set of interconnected networks according to claim 2, characterized in that: The method further comprises: If k is less than or equal to , then find out the processor nodes in the candidate set in e( )The associated nodes on the edge; Deleting associated nodes belonging to the candidate set from the candidate set; Let k=k+1, and repeat the process to see if k is less than or equal to , until k is greater than .

4. A device for constructing an independent set of interconnected networks, characterized in that: The device comprises: Label creation module to obtain the initial folding split and exchange cube network ( ) and create an initial folded split and swapped cube network ( ) is the node number of each processor node in, where and ; The node label is an n-bit binary string; A network partitioning module is used to split the initial folded and exchanged cube network ( ) is divided into several target folding segmentation and exchange cube networks ( ); According to the network dimension n and the node labels, the initial folded split and exchange cube network ( ) is divided into several target folding segmentation and exchange cube networks ( ),include: Determine whether n is equal to 2; If n is equal to 2, the target folding segmentation and exchange cube network is obtained ( ); If n is greater than 2, the initial folding split and exchange cube network ( ) is divided into several intermediate folding partitioning and exchange cube networks ; Let n = n / 2, and repeat the process of checking whether n is equal to 2 until n is equal to 2; According to the node labels, the initial folded split and exchange cube network ( ) is divided into several intermediate folding partitioning and exchange cube networks ,include: The initial folded split and exchanged cube network ( ) The node number in +1~n identical processor nodes are divided into the same intermediate folding, splitting and switching cube network , we get several intermediate folding, splitting and swapping cube networks ; Ensemble building module for folding splits and swapping cube networks from various targets ( ) are extracted to form the initial folding, splitting and switching cube network ( ), where the processor nodes in the independent set are not adjacent to each other.

5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for constructing an independent set of interconnected networks according to any one of claims 1 to 3 is implemented.

6. 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 method for constructing an independent set of interconnected networks according to any one of claims 1 to 3.

7. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for constructing an independent set of interconnected networks according to any one of claims 1 to 3 is implemented.

Images