On-chip system network simulation method, device, electronic device and storage medium

Abstract

The present application provides a method, device, electronic device and storage medium for simulating a network of a system on a crystal. The method includes: obtaining an object input by a user for a customized prompt for simulating a network of a system on a crystal; the input object includes at least customized fault information of the fault location of the fault-tolerant network; the fault location includes the node location of the faulty router and / or the location of the link where the fault occurs between adjacent routers; the input object includes at least one of a fault-tolerant network, a routing algorithm and a traffic pattern; globally querying the status of routers and links in the fault-tolerant network, locating and recording the fault location; using a routing algorithm, planning a route of a routing table for planning a transmission path for data that avoids the fault location according to the node location and the link location; using traffic simulation data packets, transmitting data according to the traffic pattern and the transmission path, simulating the function and performance of the fault-tolerant network, and outputting the simulation results. In this way, the cost is low and the versatility is good.

Description

Technical Field

[0001] The present invention relates to the field of network simulation technology, and in particular to a method, device, electronic device and storage medium for on-chip system network simulation. Background Art

[0002] With the rapid development of integrated circuit technology, System on Wafer (SoW) has become an important platform for high-performance computing and large-scale data processing. SoW achieves extremely high computing density and data transmission efficiency by integrating a large number of computing units and communication links on a single wafer.

[0003] During the use of on-chip systems, actual network failures may occur. To detect faults before they occur, designers conduct fault testing on the actual network in advance. Currently, designers temporarily construct fault-tolerant networks and routing algorithms corresponding to the actual network in real applications and then perform fault testing on the fault-tolerant network. During this process, designers write their own simulation code for a one-time test. This requires designers to write simulation code for each test, resulting in a high workload and limited test versatility. Summary of the Invention

[0004] The present application provides an improved on-chip system network simulation method, device, electronic device and storage medium.

[0005] This application provides a network simulation method for a wafer-based system, including:

[0006] Obtaining an object input by a user for a custom prompt of an on-chip system network simulation; the input object includes at least custom fault information; the fault information includes a fault location of a fault-tolerant network; the fault location includes a node location of a faulty router and / or a location of a link between adjacent routers where a fault occurs; the input object includes at least one of a fault-tolerant network, a routing algorithm, and a traffic pattern;

[0007] Globally query the status of routers and links in the fault-tolerant network, locate and record the fault location;

[0008] Using the routing algorithm, planning a route in a routing table according to the node location and the link location; the route in the routing table is used to plan a data transmission path avoiding the fault location;

[0009] Traffic simulation data packets are used to perform data transmission according to the traffic pattern and the transmission path, the function and performance of the fault-tolerant network are simulated, and simulation results are output.

[0010] Furthermore, the fault information includes a fault rate; the fault rate includes a user-defined fault rate; accordingly, the method further includes: randomly generating a fault location in the network to be simulated according to the user-defined fault rate, and constructing a fault-tolerant network including the fault location;

[0011] or,

[0012] The fault information includes a user-defined fault location input; accordingly, the method further includes: constructing a fault-tolerant network including the user-defined fault location in the network to be simulated according to the user-defined fault location.

[0013] Furthermore, the customization prompt includes prompt information; the prompt information is used to prompt the user whether to customize the object, and when customization is required, the identifier of the input object is displayed;

[0014] The object for obtaining the user's customized prompt input for the on-chip system network simulation includes:

[0015] When receiving a user input indicating that an object needs to be customized in response to the prompt information, obtaining the input object;

[0016] When receiving the user input indicating that a customized object is not required in response to the prompt information, the user is prompted to input customized fault information and obtain a default object as the input object.

[0017] Furthermore, the custom prompt includes an identifier of the input object; the identifier of the input object is used to display the custom input identifier of each object;

[0018] The object for obtaining the user's customized prompt input for the on-chip system network simulation includes:

[0019] Upon receiving an object selected by a user based on the identifier of the input object, obtaining the input object;

[0020] When receiving the user identification of the input object but no object is selected, the user is prompted to input customized fault information and obtain a default object as the input object.

[0021] Furthermore, the input objects include a custom fault-tolerant network and / or a default fault-tolerant network;

[0022] The object for obtaining the user's customized prompt input for the on-chip system network simulation includes:

[0023] In case of receiving a default fault-tolerant network selected by the user in response to the custom prompt, using the default fault-tolerant network as the fault-tolerant network;

[0024] or,

[0025] In case of receiving a customized fault-tolerant network input by the user in response to the customized prompt, using the customized fault-tolerant network as the fault-tolerant network;

[0026] or,

[0027] When a modification instruction for modifying the default fault-tolerant network selected by the user for the customized prompt is received, the default fault-tolerant network is modified according to the modification instruction to obtain a modified fault-tolerant network as the fault-tolerant network.

[0028] Furthermore, the routing algorithm includes a user-defined routing algorithm and / or a default routing algorithm;

[0029] The object for obtaining the user's customized prompt input for the on-chip system network simulation includes:

[0030] Upon receiving a default routing algorithm selected by the user in response to the custom prompt, using the default routing algorithm as the routing algorithm;

[0031] or,

[0032] When a customized routing algorithm input by the user in response to the customized prompt is received, the customized routing algorithm is used as the routing algorithm.

[0033] Furthermore, the traffic mode includes a user-defined traffic mode and / or a default traffic mode;

[0034] The object for obtaining the user's customized prompt input for the on-chip system network simulation includes:

[0035] In the case of receiving a default traffic mode selected by the user for the customized prompt, using the default traffic mode as the traffic mode;

[0036] or,

[0037] When a customized traffic pattern input by the user in response to the customized prompt is received, the customized traffic pattern is used as the traffic pattern.

[0038] The present invention provides a network simulation device for a system on a die, including:

[0039] A network simulation object acquisition module is configured to acquire an object input by a user for a custom prompt of an on-chip system network simulation; the input object includes at least custom fault information; the fault information includes a fault location of a fault-tolerant network; the fault location includes a node location of a faulty router and / or a location of a link between adjacent routers where a fault occurs; the input object includes at least one of a fault-tolerant network, a routing algorithm, and a traffic pattern;

[0040] A fault detection module, configured to globally query the status of routers and links in the fault-tolerant network, locate and record the fault location;

[0041] A routing algorithm using module, configured to use the routing algorithm to plan routes in a routing table according to the node locations and the link locations; the routes in the routing table are configured to avoid the fault location and plan a data transmission path;

[0042] The simulation module is used to use traffic simulation data packets, transmit data according to the traffic pattern and the transmission path, simulate the function and performance of the fault-tolerant network, and output simulation results.

[0043] The present application provides an electronic device, comprising one or more processors, for implementing any of the methods described above.

[0044] The present application provides a computer-readable storage medium having a program stored thereon. When the program is executed by a processor, the method described in any one of the above items is implemented.

[0045] The present application provides a computer program product, comprising a computer program / instruction, which implements any of the above methods when executed by a processor.

[0046] In some embodiments, the on-chip system network simulation method of the present application allows users to input objects according to their needs through the custom prompts of the on-chip system network simulation. In this way, users can input their own simulation requirements according to the custom prompts of the on-chip system network simulation, which can be applied to the fault simulation requirements of different users, and different simulation requirements can be set repeatedly, thereby improving the versatility and repeatability of this method. Moreover, based on the input object, the simulation is automatically completed and the simulation results are output. In this way, the framework of the network simulation method is built, which reduces the workload of designers in writing simulation code each time and improves the simulation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 FIG2 is a flow chart of a network simulation method for a system on a die according to an embodiment of the present application;

[0048] Figure 2 Shown Figure 1A schematic diagram of a network topology structure of a fault-tolerant network with link failures in the on-wafer system network simulation method shown;

[0049] Figure 3 Shown Figure 1 A schematic diagram of a network topology structure of a fault-tolerant network with node failures in the on-wafer system network simulation method shown;

[0050] Figure 4 Shown Figure 1 A schematic diagram of fault avoidance of a fault-tolerant network with link failure in the on-chip system network simulation method is shown;

[0051] Figure 5 Shown Figure 1 A schematic diagram of fault avoidance for a fault-tolerant network with node failures in the on-chip system network simulation method is shown;

[0052] Figure 6 FIG2 is a schematic diagram showing the structure of a network simulation device for a system on a die according to an embodiment of the present application;

[0053] Figure 7 Shown is a schematic structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with one or more embodiments of this specification. Rather, they are merely examples of apparatuses and methods consistent with certain aspects of one or more embodiments of this specification, as detailed in the appended claims.

[0055] It should be noted that in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in this specification. In some other embodiments, the method may include more or fewer steps than those described in this specification. In addition, a single step described in this specification may be broken down into multiple steps for description in other embodiments, and multiple steps described in this specification may be combined into a single step for description in other embodiments.

[0056] In order to solve the technical problems of heavy workload and poor versatility of each test for designers, an embodiment of the present application provides a network simulation method for a system on a chip, through which users can input objects according to their needs through the custom prompts of the network simulation for the system on a chip. In this way, users can input their own simulation requirements according to the custom prompts of the network simulation for the system on a chip, which can be applied to the fault simulation requirements of different users, and different simulation requirements can be set repeatedly, thereby improving the versatility and repeatability of this method. Moreover, based on the input object, the simulation is automatically completed and the simulation results are output. In this way, the framework of the network simulation method is built, which reduces the workload of designers in writing simulation code each time and improves the simulation efficiency.

[0057] Due to IC manufacturing yields, as well as issues such as electromigration and other wear-out effects, network reliability can be affected by transient or permanent faults. These faults not only affect network functionality but also degrade network performance. Therefore, to improve the reliability of on-chip system networks, it is crucial to design fault-tolerant networks (also known as fault-tolerant networks) that can tolerate defects in some chip regions while remaining functional.

[0058] The on-chip system network simulation method (referred to as network simulation) described in this application is applied to a simulator. Using the simulator, the input interface (also known as customized prompts) for the fault-tolerant network and routing algorithm is pre-configured. Users can obtain customized fault information through the simulator, simulating network operation under different fault scenarios. The pre-configured simulated network and routing algorithm also provide data support for simulated network design. Furthermore, the simulator allows engineers to pre-evaluate the network's fault tolerance, optimize network topology, and verify the effectiveness of routing algorithms.

[0059] The above-mentioned simulator is installed in an electronic device. Specifically, the electronic device can be: a desktop computer, a portable computer, an intelligent mobile terminal, a server, a PDA (Personal Digital Assistant) and a handheld terminal, etc. Among them, the PDA can include an industrial PDA and a consumer PDA. Any electronic device that can implement the embodiment of the present invention falls within the scope of protection of the present invention and is not limited here. In this way, the electronic device serves as a carrier of the simulator, and the user can use the simulator through the electronic device to execute the on-chip system network simulation method in this specification. Please see below for detailed description.

[0060] Figure 1 The figure shows a flow chart of the on-chip system network simulation method provided in an embodiment of the present application.

[0061] like Figure 1As shown, the on-chip system network simulation method may include but is not limited to the following steps 110 to 140:

[0062] Step 110: Obtain an object input by the user for a customized prompt for on-chip system network simulation; the input object includes at least customized fault information; the fault information includes the fault location in the fault-tolerant network; the fault location includes the node location of the faulty router and / or the location of the failed link between adjacent routers; the input object includes at least one of a fault-tolerant network, a routing algorithm, and a traffic pattern. In this way, a fault-tolerant network requiring network simulation can be constructed.

[0063] It should be noted that the customized prompt is used to prompt the user to input the actual simulation requirements of the network to be simulated, and the customized fault information is a required option for the user to simulate the simulation conditions of the network to be simulated under different fault scenarios according to the user's actual simulation requirements.

[0064] The network to be simulated is configured according to the input objects, resulting in a fault-tolerant network. This means that even if network elements (such as nodes or links) fail, the network system can still maintain basic functionality or quickly restore normal service.

[0065] Next, the input objects are used to represent the objects required for network simulation. The input objects include at least user-defined fault information. This fault information then represents the specific fault scenario of the network to be simulated. The input objects may include, but are not limited to, pre-configured default objects and / or user-defined objects. For example, if a user-defined object is used, for example, a user imports an input object as a fault-tolerant network, the user only needs to customize the input object and does not need to write the entire simulation code. This significantly reduces the workload and allows users to set objects according to their needs, improving the user experience.

[0066] The aforementioned default objects may include, but are not limited to, at least one of a pre-configured default fault-tolerant network, a pre-configured default routing algorithm, and a pre-configured default traffic pattern. The network to be simulated may be used as the default fault-tolerant network. Thus, if the user does not require customization or selects the default configuration, the user's simulation requirements can be met by using the default objects, such as at least one of the pre-configured network to be simulated, the default routing algorithm, and the pre-configured traffic pattern, along with any required customized fault information.

[0067] For example, the number of pre-configured simulated networks, pre-configured routing algorithms, and pre-configured traffic patterns can be one or more. These pre-configured default objects can display selection buttons for selecting the input objects, thereby receiving the objects input by the user through the selection buttons. This facilitates user input.

[0068] Step 120: globally query the status of routers and links in the fault-tolerant network, locate and record the fault location, and thus perform fault detection.

[0069] The above step 120 may include but is not limited to globally querying the status of all routers and links in the fault-tolerant network, wherein the status may include but is not limited to connection status, whether it is working properly, and whether there is a fault.

[0070] The above-mentioned fault location is used to represent the specific fault condition of the fault-tolerant network. The fault location may include, but is not limited to, the node location of the failed router (also known as a node failure) and / or the location of the link between adjacent routers (also known as a link failure).

[0071] When executing step 120, the method further includes graphically displaying the fault-tolerant network including the fault location. This allows the user to intuitively check and verify whether the graphical display of the fault-tolerant network including the fault location is consistent with the set actual simulation requirements, thereby improving the effectiveness of the simulation.

[0072] Step 130: Use a routing algorithm to plan routes in a routing table based on node and link locations. Routes in the routing table are used to plan data transmission paths that avoid fault locations. In this way, routes are planned using a routing algorithm for subsequent simulations.

[0073] In this step 130, the above-mentioned routing algorithm is used to plan the transmission path formed by the communication links and routers in the fault-tolerant network except for the fault location. In this way, based on at least one information such as the network topology structure, link status, node status, etc. of the fault-tolerant network, the optimal transmission path from the source node to the target node is calculated to ensure the efficient delivery of the traffic simulation data packet. Moreover, when the method of the present application is executed again, when a link failure, node failure change or network topology adjustment in the fault-tolerant network is received, the transmission path is recalculated to maintain communication continuity. In addition, the traffic of each communication link in the fault-tolerant network is balanced to avoid congestion and improve the overall network throughput and reliability.

[0074] It should also be noted that the above routing algorithms may include but are not limited to at least one of a distance vector routing algorithm (DV), a link state routing algorithm (LS) and a path vector routing algorithm (Path-Vector) for user selection or input.

[0075] After step 130 , the method further includes displaying the transmission path in a graphical form.

[0076] Step 140 , using traffic simulation data packets, performing data transmission according to the traffic pattern and transmission path, simulating the function and performance of the fault-tolerant network, and outputting the simulation results.

[0077] Traffic simulation packets are primarily used to simulate nodes and links in fault-tolerant networks. They contain source and destination address identification data, source and destination ports, the amount of data carried by the packet, and a timestamp.

[0078] The traffic patterns in step 140 refer to the way traffic simulation packets are transmitted across the network for different communication tasks, such as application network traffic patterns, all-user traffic patterns, random traffic patterns, and hotspot traffic patterns. During the simulation, traffic simulation packets corresponding to different traffic patterns are generated and sent by terminals connected to the router. This understanding of the characteristics of different traffic patterns facilitates more precise network design, performance optimization, and troubleshooting.

[0079] For example, the above traffic pattern may include but is not limited to a hotspot pattern. The hotspot pattern includes a packet receiving node and link, and a packet sending node and link, and can implement packet sending to all points, for example, first to router A, through link B, then to router C, and so on.

[0080] In contrast, one way to output the simulation result in step 140 is to output the simulation result in the form of a document.

[0081] Another way to output the simulation results of the above step 140 is to output the simulation results in a graphical form. The simulation results include a real-time traffic heat map of each router used for transmission and / or a communication link used for transmission; the real-time traffic heat map is used to indicate the size of the transmission volume of each router used for transmission and / or a communication link used for transmission. The color depth of the real-time traffic heat map is positively correlated with the size of the transmission volume. For example, the darker the color in the real-time traffic heat map, the larger the transmission volume. The lighter the color in the real-time traffic heat map, the smaller the transmission volume. In this way, it is convenient for users to intuitively know that the specific routers used for transmission and / or the communication links used for transmission are used more frequently.

[0082] It should also be noted that the routers used for transmission are the remaining routers in the fault-tolerant network that have not experienced any failure, excluding the node location of the failed router. The communication links used for transmission are the communication links between adjacent routers that have not experienced any failure, excluding the link location of the failed router.

[0083] In the embodiments of this application, a fault-tolerant network and routing algorithm are constructed and configured to perform fault detection and complete the simulation. This effectively simulates the operation of the fault-tolerant network of the on-chip system under different fault scenarios, providing strong support for fault-tolerant network design and optimization. Furthermore, it reduces user input operations, enabling more efficient simulation.

[0084] Combine Figure 1 As shown, in one embodiment of an application, the network topology of the network to be simulated includes a regular topology. Accordingly, the method further includes: obtaining the fault location in the fault-tolerant network based on the fault information, and constructing a fault-tolerant network including the fault location. Furthermore, the network topology of the fault-tolerant network is a regular topology or an irregular topology. This step can be performed before the above-mentioned arrangement 120. In this way, the network to be simulated with a regular topology can be simulated with an irregular topology or a regular topology, thereby being applicable to various network topologies.

[0085] In another embodiment of the application, the network topology of the simulated network includes an irregular topology. Accordingly, the method further includes: obtaining the fault location in the fault-tolerant network based on the fault information, and constructing a fault-tolerant network including the fault location; and the network topology of the fault-tolerant network is a regular topology or an irregular topology. This step can be performed before the above-mentioned arrangement 120. In this way, the network to be simulated with an irregular topology can be simulated with either an irregular topology or a regular topology, thereby adapting to various network topologies.

[0086] The network to be simulated in the embodiment of the present application refers to a basic network for which input objects have not yet been set, and is used to represent the initial network during network simulation. The network topology structure of the network to be simulated can be of multiple types. One type of network topology structure of the network to be simulated is regular. The network topology structure of the network to be simulated is, for example, a 4x4 network topology structure. The network topology structure of the network to be simulated is, for example, a 3x3 network topology structure. In this way, the network topology structure has N rows and M columns. N and M can be positive integers respectively, and N and M can be the same or different. Another type of network topology structure of the network to be simulated is irregular.

[0087] Based on this, the network to be simulated may include multiple routers (R for short). Figure 2 、 Figure 3 、 Figure 4 and Figure 5The network topology of the simulated network shown is regular. Routers R of the simulated network exemplarily include router R1, router R2, router R3, router R4, router R5, router R6, router R7, router R8, router R9, router R10, router R11, router R12, router R13, router R14, router R15, router R16, and communication links between adjacent routers.

[0088] The regular network to be simulated can be a 2D mesh (two-dimensional mesh) or other common network topologies. Failures in these routers are called node failures. Failures in these communication links are called link failures. These faulty locations are marked as unavailable for subsequent simulations.

[0089] like Figure 2 、 Figure 3 、 Figure 4 and Figure 5 As shown, the embodiment of the present invention can locate, record and display the fault location. Figures 2 to 5 In the figure, circles represent routers and straight lines represent links between routers.

[0090] Continue as Figure 2 The fault-tolerant network with link failure is shown in FIG. The network topology of the network to be simulated is a 4x4 network topology. Figure 2 The cross shown represents a faulty link, which is set to 0. At this time, the network topology of the fault-tolerant network is irregular.

[0091] Continue as Figure 3 The fault-tolerant network with node failure is shown in FIG. The network topology of the network to be simulated is a 4x4 network topology. Figure 3 The cross shown represents a faulty node, which is set to 0. At this time, the network topology of the fault-tolerant network is irregular.

[0092] Continue as Figure 4 The fault-tolerant network with link failure and fault avoidance is shown. The network topology of the network to be simulated is a 4x4 network topology. Figure 4 The cross in the figure represents a faulty link. When the routers at both ends of the faulty link detect the link failure, they set their ports to 0 and route around the link. In this case, the network topology of the fault-tolerant network is irregular.

[0093] Continue as Figure 5 The figure shows a fault-tolerant network with node failure and fault avoidance. The network topology of the network to be simulated is a 4x4 network topology. Figure 5The cross in the figure represents a faulty node. When the router connected to the faulty node detects the fault, it sets the port to 0 and routes around the node. In this case, the network topology of the fault-tolerant network is irregular.

[0094] Combine Figures 1 to 5 As shown, in one aspect, embodiments of the present application can randomly select a link or faulty node in the simulated network as a fault location (also known as a fault point) based on a set failure rate. In another aspect, embodiments of the present application can set the fault location. Detailed description is as follows:

[0095] In a first optional approach, the fault information includes a failure rate; the failure rate includes a user-defined failure rate; and the failure rate represents the proportion of failures in the fault-tolerant network. Accordingly, the method further includes randomly generating fault locations in the simulated network according to the user-defined failure rate, and constructing a fault-tolerant network including the fault locations. This step may be performed before the arrangement 120 described above.

[0096] In some examples, the failure rate may include the failure rate of a single faulty node or the failure rate of a single faulty link. The failure rate of the faulty node is multiplied by the total number of router nodes in the simulated network, and the result is rounded up or down to obtain the number of faulty nodes. Similarly, the failure rate of the faulty link is multiplied by the total number of communication links in the simulated network, and the result is rounded up or down to obtain the number of communication links.

[0097] For example, a number less than 1 can be rounded up and the rounded-up fault location can be displayed to the user to remind the user. A number greater than 1 can be rounded down and the rounded-down fault location can be displayed to the user.

[0098] In another example, the failure rate may be a failure rate common to a failed node and a failed link.

[0099] For example, the total number of routing nodes is 4, and the total number of communication links is 8. If the failure rate is 0.1, there is 1 failed node and 1 failed link.

[0100] In the first optional method, links or faulty nodes in the network are randomly selected as failure points based on the set failure rate to simulate random failure scenarios. These failure points will be marked as unavailable for subsequent simulation processes.

[0101] In a second optional approach, the fault information includes a user-defined fault location. Accordingly, the method further includes: constructing a fault-tolerant network containing the customized fault location within the network to be simulated, based on the customized fault location. This step can be performed before the aforementioned arrangement 120. In this manner, a faulty link or node in the network to be simulated is manually specified. Based on specific testing requirements, the user can select a specific link or node as the fault point and mark it as unavailable to facilitate fault detection and routing algorithm simulation during subsequent simulations. Manually specifying a faulty link or node in this manner allows for the construction of a network with specific failure modes to meet simulation requirements in different scenarios.

[0102] In an embodiment of the present application, in the stage of building a fault-tolerant network, the above two construction methods can be used: one is to receive the failure rate set by the user in a regular network to simulate a random failure scenario; the other is to manually specify the fault link or fault node, and a network with a specific failure mode can be built to meet the simulation requirements in different scenarios.

[0103] Combine Figures 1 to 5 As shown, when the user enters a custom object in response to a customization prompt, prompting the user to enter custom fault information and obtaining a default object as the input object. When the user enters a custom object in response to a customization prompt, the user enters the object as input for the network to be simulated, and the input object is obtained. This allows the user to choose whether to customize the object, making it convenient for the user to set the object according to their needs.

[0104] The above step 110 may obtain the input object in at least one of the following ways:

[0105] In an optional manner, a customization prompt is displayed; the customization prompt includes prompt information, the prompt information is used to prompt the user whether to customize the object, and when customization is required, the identifier of the input object is displayed. The input object refers to the input object.

[0106] When receiving the user's input indicating that an object needs to be customized in response to the prompt information, the input object is obtained.

[0107] When receiving the user input indicating that a customized object is not required in response to the prompt information, the user is prompted to input customized fault information and obtain a default object as the input object.

[0108] For example, if there is one default object, the default object is used as the input of the network to be simulated to obtain the input object.

[0109] If there are multiple default objects, the identifiers of the input objects are displayed, and the default object selected by the user according to the identifier of the input object is received; and the selected default object is used as the input of the network to be simulated to obtain the input object.

[0110] In the first optional method, the user can choose to determine whether to use the default object or the user-defined object, thereby improving the user's autonomy and the flexibility of the simulation input.

[0111] In a second optional method, a customized prompt is displayed; the customized prompt includes an input object identifier; the input object identifier is used to display the customized input identifier of each object. Upon receiving an object selection from the user based on the input object identifier, the input object is obtained. If the user does not select an object based on the input object identifier, the user is prompted to enter customized fault information and a default object is obtained as the input object. This approach is cost-effective and highly versatile.

[0112] For example, if there is one default object, the default object is used as the input of the network to be simulated to obtain the input object.

[0113] If there are multiple default objects, the identifiers of the input objects are displayed, and the default object selected by the user according to the identifier of the input object is received; and the selected default object is used as the input of the network to be simulated to obtain the input object.

[0114] In the second optional mode, the identifiers of the input objects are displayed by default for the user to select. The user can set the objects to be customized as needed, thereby reducing the steps of user operation and improving the user's autonomy and the flexibility of the simulation input.

[0115] Combine Figures 1 to 5 As shown, in the third optional method, the input object includes a custom fault-tolerant network and / or a default fault-tolerant network. Accordingly, when the user selects a default fault-tolerant network in response to the custom prompt, the default fault-tolerant network is used as the fault-tolerant network. In this way, when the user does not enter a custom fault-tolerant network in response to the custom prompt, subsequent simulation can be performed according to the default fault-tolerant network.

[0116] In a fourth optional manner, when a user-defined fault-tolerant network is received in response to a custom prompt, the user-defined fault-tolerant network is used as the fault-tolerant network. In this way, simulation is performed according to the user-defined fault-tolerant network to meet the user's simulation requirements.

[0117] In a fifth optional approach, upon receiving a modification instruction for a user-selected default fault-tolerant network in response to a custom prompt, the default fault-tolerant network is modified according to the modification instruction to obtain the modified fault-tolerant network as the fault-tolerant network. In this way, the fault-tolerant network can be adjusted under the conditions of the original default fault-tolerant network, reducing the user's need to enter simulation code from beginning to end.

[0118] In a sixth optional embodiment, the routing algorithm includes a user-defined routing algorithm and / or a default routing algorithm. Accordingly, upon receiving a user-selected default routing algorithm in response to a customized prompt, the default routing algorithm is used as the routing algorithm. In this manner, simulation is performed according to the user-defined routing algorithm, thereby satisfying the user's simulation requirements.

[0119] In a seventh optional manner, when a customized routing algorithm input by the user in response to a customized prompt is received, the customized routing algorithm is used as the routing algorithm.

[0120] In an eighth optional manner, the above-mentioned traffic mode includes a user-defined traffic mode and / or a default traffic mode; when a default traffic mode selected by the user in response to a customized prompt is received, the default traffic mode is used as the traffic mode.

[0121] In a ninth optional manner, when a user-defined traffic pattern is received in response to a customized prompt, the customized traffic pattern is used as the traffic pattern. In this way, simulation is performed according to the user-defined traffic pattern to meet the user's simulation requirements.

[0122] The detailed implementation process of the on-chip system network simulation method of the embodiment of the present application is as follows:

[0123] Step 1: Fault-tolerant network setup and construction: During the fault-tolerant network construction phase, the Jingshang system network simulation system provides two construction methods: one is to accept user-defined failure rates in a regular network to simulate random failure scenarios; the other is to manually specify faulty links or faulty nodes to build a network with specific failure modes. This can meet simulation requirements in different scenarios.

[0124] Specifically, in step 1.1: Random Fault Construction, a user-defined failure rate is received in a regular, fault-tolerant network topology. Based on the failure rate, a link or faulty node in the simulated network is randomly selected as the fault point. Furthermore, in step 1.2: Specific Fault Construction, a faulty link or faulty node in the simulated network is manually specified.

[0125] Step 2: Fault detection: During the initialization phase of the network topology, the network simulation system of the Jingshang system comprehensively queries the status of routers and links in the network, accurately locates the fault location in the network, and records the fault information.

[0126] Specifically, Step 2.1 queries the status of all routers and links in the network topology, including connectivity, normal operation, and faults. And, Step 2.2 uses this status information to precisely locate the fault in the network. The on-chip system network simulation system records the location of any detected faulty links or nodes in faulty routes, enabling subsequent routing algorithms to dynamically adjust based on this information.

[0127] Step 3: Routing Algorithm Utilization: After fault detection is complete, all routers on the fault-tolerant network update their router and link status based on the detected fault location. The fault-tolerant network then activates the appropriate routing algorithm, supporting both the system's default fault-tolerant routing and user-defined routing in the routing table. This step supports system-provided, user-defined, and default global routing.

[0128] Specifically, 3.1: All routers on the fault-tolerant network update the router and link status based on the detected fault location. Each router dynamically adjusts the link status within the router based on the status information of its adjacent links to avoid faulty links or faulty nodes. And, 3.2: Enable the corresponding routing algorithm based on the fault location. The default network simulation of the Jingshang system is to modify the router's steering rules, based on logic-based distributed routing, and support users to customize routing algorithms. Among them, the router's steering rules can include 4 directions, such as east, south, west, and north. The router's steering rules can include 8 directions, such as southeast, southwest, northeast, northwest, east, south, west, and north.

[0129] Step 4: Network performance simulation: Based on the traffic pattern, simulation is performed in the above fault-tolerant network and the simulation results are output.

[0130] Specifically, 4.1 simulates the data packet transmission process in the fault-tolerant network according to the traffic model, and outputs the simulation results, including performance indicators such as throughput, delay and packet loss rate.

[0131] Based on the same inventive concept as the above method, the embodiment of the present application also provides a network simulation device for a system on a die, such as Figure 6 As shown, the on-chip system network simulation device includes the following modules:

[0132] The network simulation object acquisition module 31 is used to acquire an object input by a user for a customized prompt of an on-chip system network simulation; the input object includes at least customized fault information; the fault information includes a fault location of a fault-tolerant network; the fault location includes a node location of a faulty router and / or a location of a link between adjacent routers where a fault occurs; the input object includes at least one of a fault-tolerant network, a routing algorithm, and a traffic pattern;

[0133] A fault detection module 32 is used to globally query the status of routers and links in the fault-tolerant network, locate and record the fault location;

[0134] The routing algorithm using module 33 is used to use the routing algorithm to plan the routing of the routing table according to the node location and the link location; the routing of the routing table is used to plan the data transmission path to avoid the fault location;

[0135] The simulation module 34 is used to use traffic to simulate data packets, transmit data according to the traffic pattern and transmission path, simulate the function and performance of the fault-tolerant network, and output simulation results.

[0136] The modules of the above-mentioned device correspond to the steps of the above-mentioned method. The implementation process of the functions and effects of each module in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method. The same technical effects can be achieved and will not be repeated here.

[0137] An embodiment of the present application provides an electronic device, including the above-mentioned on-chip system network simulation device.

[0138] Figure 7 Shown is a schematic structural diagram of an electronic device 50 provided in an embodiment of the present application.

[0139] like Figure 7 As shown, the electronic device 50 includes one or more processors 51 for implementing the above-mentioned on-wafer system network simulation method.

[0140] In some embodiments, the electronic device 50 may include a storage medium 59. For example, the computer-readable storage medium may store a program that can be called by the processor 51, and may include a non-volatile storage medium. In some embodiments, the electronic device 50 may include a memory 58 and an interface 57. In some embodiments, the electronic device 50 may also include other hardware depending on the actual application.

[0141] The computer-readable storage medium of the embodiment of the present application stores a program thereon, and when the program is executed by the processor 51, it is used to implement the on-wafer system network simulation method described above.

[0142] The present application provides a computer program product, comprising a computer program / instruction, which implements any of the above methods when executed by a processor.

[0143] The embodiment of the present application further provides a computer program stored in a computer-readable storage medium, for example, Figure 7 The computer program is stored in a storage medium 59, and when the processor executes the computer program, the processor 51 is prompted to perform the method described above.

[0144] This application may take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to magnetic disk storage, CD-ROMs, optical storage, etc.) containing program code. Computer-readable storage media include both permanent and non-permanent, removable and non-removable media, and may implement information storage using any method or technology. The information may be computer-readable instructions, data structures, program modules, or other data. Examples of computer-readable storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device.

[0145] The above description is only a preferred embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of this specification should be included in the scope of protection of this specification.

[0146] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, the phrase "comprises a ..." defining an element does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

Claims

1. A network simulation method for a wafer-based system, characterized in that: include: Obtaining an object input by a user for a customized prompt of an on-chip system network simulation; the input object includes at least customized fault information; The fault information includes the fault location and fault rate of the fault-tolerant network; the fault rate includes a user-defined fault rate; the fault location includes the node location of the faulty router and / or the location of the link where the fault occurs between adjacent routers; the input objects include at least one of a fault-tolerant network, a routing algorithm, and a traffic pattern; the fault-tolerant network includes a user-defined fault-tolerant network and a default fault-tolerant network, for the user to select and input the fault-tolerant network; the routing algorithm includes a user-defined routing algorithm and a default routing algorithm, for the user to select and input the routing algorithm; the traffic pattern includes a user-defined traffic pattern, for the user to select and input the traffic pattern; According to the customized failure rate, randomly generating the failure location in the network to be simulated, and constructing a fault-tolerant network including the failure location; Globally query the status of routers and links in the fault-tolerant network, locate and record the fault location; Using the routing algorithm, planning a route in a routing table according to the node location and the link location; the route in the routing table is used to plan a data transmission path avoiding the fault location; Traffic simulation data packets are used to perform data transmission according to the traffic pattern and the transmission path, the function and performance of the fault-tolerant network are simulated, and simulation results are output.

2. The on-chip system network simulation method according to claim 1, wherein: The fault information includes a user-defined fault location input; accordingly, the method further includes: constructing a fault-tolerant network including the user-defined fault location in the network to be simulated according to the user-defined fault location.

3. The on-chip system network simulation method according to claim 1, wherein: The customization prompt includes prompt information; the prompt information is used to prompt the user whether to customize the object, and when customization is required, the identifier of the input object is displayed; The object for obtaining the user's customized prompt input for the on-chip system network simulation includes: When receiving a user input indicating that an object needs to be customized in response to the prompt information, obtaining the input object; When receiving the user input indicating that a customized object is not required in response to the prompt information, the user is prompted to input customized fault information and obtain a default object as the input object.

4. The on-chip system network simulation method according to claim 1, wherein: The custom prompt includes an identifier of the input object; the identifier of the input object is used to display the custom input identifier of each object; The object for obtaining the user's customized prompt input for the on-chip system network simulation includes: Upon receiving an object selected by a user based on the identifier of the input object, obtaining the input object; When receiving the user identification of the input object but no object is selected, the user is prompted to input customized fault information and obtain a default object as the input object.

5. The on-chip system network simulation method according to claim 1 or 4, characterized in that: The input objects include a custom fault-tolerant network and / or a default fault-tolerant network; The object for obtaining the user's customized prompt input for the on-chip system network simulation includes: In case of receiving a default fault-tolerant network selected by the user in response to the custom prompt, using the default fault-tolerant network as the fault-tolerant network; or, In case of receiving a customized fault-tolerant network input by the user in response to the customized prompt, using the customized fault-tolerant network as the fault-tolerant network; or, When a modification instruction for modifying the default fault-tolerant network selected by the user for the customized prompt is received, the default fault-tolerant network is modified according to the modification instruction to obtain a modified fault-tolerant network as the fault-tolerant network.

6. The on-chip system network simulation method according to claim 1 or 4, characterized in that: The routing algorithm includes a user-defined routing algorithm and / or a default routing algorithm; The object for obtaining the user's customized prompt input for the on-chip system network simulation includes: Upon receiving a default routing algorithm selected by the user in response to the custom prompt, using the default routing algorithm as the routing algorithm; or, When a customized routing algorithm input by the user in response to the customized prompt is received, the customized routing algorithm is used as the routing algorithm.

7. The on-chip system network simulation method according to claim 4, characterized in that: The traffic mode includes a user-defined traffic mode and / or a default traffic mode; The object for obtaining the user's customized prompt input for the on-chip system network simulation includes: In the case of receiving a default traffic mode selected by the user for the customized prompt, using the default traffic mode as the traffic mode; or, When a customized traffic pattern input by the user in response to the customized prompt is received, the customized traffic pattern is used as the traffic pattern.

8. A network simulation device for a system on a chip, characterized in that: include: The network simulation object acquisition module is used to acquire the object input by the user for the customized prompt of the on-chip system network simulation; the input object at least includes customized fault information; The fault information includes a fault location and a fault rate of the fault-tolerant network; the fault rate includes a user-defined fault rate; the fault location includes a node location of a faulty router and / or a link location of a faulty router between adjacent routers; the input objects include at least one of a fault-tolerant network, a routing algorithm, and a traffic pattern; the fault-tolerant network includes a user-defined fault-tolerant network and a default fault-tolerant network, for the user to select and input the fault-tolerant network; the routing algorithm includes a user-defined routing algorithm and a default routing algorithm, for the user to select and input the routing algorithm; the traffic pattern includes a user-defined traffic pattern, for the user to select and input the traffic pattern; according to the user-defined fault rate, the fault location is randomly generated in the network to be simulated, and a fault-tolerant network including the fault location is constructed; A fault detection module, configured to globally query the status of routers and links in the fault-tolerant network, locate and record the fault location; A routing algorithm using module, configured to use the routing algorithm to plan routes in a routing table according to the node locations and the link locations; the routes in the routing table are configured to avoid the fault location and plan a data transmission path; The simulation module is used to use traffic simulation data packets, transmit data according to the traffic pattern and the transmission path, simulate the function and performance of the fault-tolerant network, and output simulation results.

9. An electronic device, characterized in that: The device comprises one or more processors, and is configured to implement the on-chip system network simulation method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that A program is stored thereon, and when the program is executed by a processor, the on-chip system network simulation method according to any one of claims 1 to 7 is implemented.

Images