System and method for generating quality metrics for optimization tasks in topology synthesis of network
By analyzing NoC and generating quality metrics, the problems of NoC design complexity and frequent modifications are solved, the design process is simplified, the design efficiency is improved, and production delays are reduced.
Patent Information
- Application Number
- CN202411126220.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2020-12-22
- Filing Date
- 2021-12-03
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2041-12-03
AI Technical Summary
In network-on-chip (NoC) design, creating the best topology that meets performance requirements is a complex and time-consuming task, and the design needs to be frequently modified every time the requirements change, resulting in production delays.
Provides a system and method to analyze NoC and generate quality metrics to maximize the quality of the result by optimizing strategies, including analyzing network paths and edges, determining the actual and ideal distance differences between the source and destination, and performing edge clustering to minimize Make long sides simplify the design process and evaluate network performance.
Simplify the design process through quality measurement, improve designer work efficiency, effectively evaluate network optimization quality, reduce design time, and reduce production delays.
Smart Images

Figure CN119025470A_ABST
Abstract
Description
[0001] This application is a divisional application of the Chinese invention patent application with the application date of December 3, 2021, application number 202111464959.7, and invention name “System and method for generating quality metrics of optimization tasks in topological synthesis of networks”. Technical Field
[0002] The present invention is in the field of network design and, more particularly, relates to generating and using quality metrics in topology synthesis of a Network on Chip (NoC). Background Art
[0003] Multiprocessor systems have been implemented in System-on-Chip (SoC) that communicate via Network-on-Chip (NoC). The SoC includes instances of master (initiator) intellectual property (IP) and slave (target) intellectual property. Transactions are sent from a master to one or more slaves in the form of packets using an industry standard protocol. A master connected to the NoC uses an address to select a slave and sends a request transaction to the slave. The NoC decodes the address and transmits the request from the master to the slave. The slave processes the transaction and sends a response transaction, which is transmitted back to the master by the NoC.
[0004] For a given set of performance requirements, such as connectivity and latency between sources and destinations, frequencies of various components, maximum area available for NoC logic, minimum throughput between sources and destinations, locations of components connected to the NoC on the floorplan, creating an optimal NoC that meets all requirements with a minimum amount of logic and wires is a complex task. This is usually the job of a chip architect or chip designer to create an optimal NoC, which is a difficult and time-consuming task. In addition to being a difficult task, the design of the NoC is modified every time the requirements change, such as modifying the chip floorplan or modifying the expected performance. Therefore, this task needs to be redone frequently during the design time of the chip. This process is time-consuming and leads to production delays.
[0005] As mentioned earlier, creating an optimal NoC is a complex task, one that meets all requirements in terms of performance and minimizes the amount of logic and wires used. This is a difficult and time-consuming task. Moreover, the design of the NoC is modified every time a requirement changes. Therefore, the optimization of the NoC is often redone during design time. Various processes are considered to automatically create a NoC when a set of constraints is given. Regardless of the process used to automatically create a NoC, the results produced by the automatic process need to be evaluated to determine the quality of the results or output. Therefore, what is needed is a system and method for analyzing a NoC and generating quality metrics for a given generation solution for the NoC. Summary of the invention
[0006] According to various embodiments and aspects of the present invention, systems and methods are disclosed for analyzing a network on chip (NoC) and generating quality metrics for a given generated solution of the NoC. The disclosed systems and methods include processes for implementing optimization strategies to maximize the quality of results, which allows designers to understand whether the results generated are good, average, or bad solutions. Advantages of the present invention include simplifying the design process and the work of designers by using quality metrics generated for the network. Various quality metrics are determined. These quality indicators provide a method for quality assessment. Quality metrics can determine the optimization quality of the generated network. Quality metrics include analyzing paths and edges in the network. BRIEF DESCRIPTION OF THE DRAWINGS
[0007] In order to more fully understand the present invention, the accompanying drawings are provided for reference. The present invention is described according to aspects and embodiments in the following description with reference to the accompanying drawings or figures (Fig), wherein the same numbers represent the same or similar elements. It should be understood that these drawings should not be considered as limiting the scope of the present invention, and the currently described aspects and embodiments of the present invention and the best mode currently understood are described in additional detail through the use of the accompanying drawings.
[0008] Figure 1A A logical view of a network on a chip (NoC) is shown, including the various elements that create a NoC.
[0009] Figure 1B A NoC including elements placed on a floorplan of the network is shown.
[0010] Figure 2 A process for determining actual relative to optimal distances between source (R) and sink (S) pair endpoints according to aspects and embodiments of the present invention is shown.
[0011] Figure 3A A block diagram of floorplanning with flight paths and shortest paths according to aspects and embodiments of the present invention is shown.
[0012] Figure 3B A block diagram of a layout plan with actual paths and flight paths according to aspects and embodiments of the present invention is shown.
[0013] Figure 4 A process for clustering or merging edges in a NoC according to various aspects and embodiments of the present invention is shown.
[0014] Figure 5A A block diagram illustrating pre-clustering of edges in a NoC according to various aspects and embodiments of the present invention is shown.
[0015] Figure 5B A block diagram showing edges of clusters in a NoC according to various aspects and embodiments of the present invention is shown.
[0016] Figure 6 A floorplan including a connection map according to various aspects and embodiments of the present invention is shown.
[0017] Figure 7 A diagram showing a circuit diagram including a logical edge and a switch according to various aspects and embodiments of the present invention. Figure 6 Layout planning.
[0018] Figure 8 A schematic diagram showing a segmented path including various aspects and embodiments of the present invention is shown. Figure 7 Layout planning.
[0019] Fig. 9 The diagram shows a flow chart including an optimal path and an actual path according to various aspects and embodiments of the present invention. Figure 8 Layout planning. DETAILED DESCRIPTION
[0020] Various examples of the present technology that illustrate various aspects and embodiments of the present invention are described below. In general, the examples may use the described aspects in any combination. All statements describing its principles, aspects, and embodiments and specific examples thereof herein are intended to include structural and functional equivalents thereof. In addition, it is intended that such equivalents include currently known equivalents and equivalents developed in the future, i.e., any element developed to perform the same function, regardless of its structure.
[0021] Note that, as used herein, the singular forms "a", "an", and "the" include plural referents unless the context clearly dictates otherwise. References in this specification to "an aspect", "an aspect", "some aspects", "various aspects", or similar language mean that a particular aspect, feature, structure, or characteristic described in connection with any embodiment is included in at least one embodiment of the present invention.
[0022] In this specification, the appearance of the phrases "in one embodiment", "in at least one embodiment", "in an embodiment", "in certain embodiments" and similar language may, but do not necessarily, refer to the same embodiment or similar embodiments. In addition, the aspects and embodiments of the invention described herein are merely exemplary and should not be construed as limitations on the scope or spirit of the invention as understood by those of ordinary skill in the art. The disclosed invention is effectively made or used in any embodiment including any novel aspect described herein. All statements describing the principles, aspects and embodiments of the invention herein are intended to include their structural and functional equivalents. It is intended that these equivalents include currently known equivalents and equivalents developed in the future.
[0023] As used herein, "source", "host" and "initiator" refer to intellectual property (IP) blocks, units or modules. The terms "source", "host" and "initiator" are used interchangeably in the scope and embodiments of the present invention. As used herein, "sink", "destination", "slave" and "target" refer to IP blocks; these terms are used interchangeably in the scope and embodiments of the present invention. As used herein, a "transaction" is a request transaction or a response transaction. Examples of request transactions include write requests and read requests.
[0024] Transactions are passed from a source to a destination via a routing or communication path that passes through an on-chip network (NoC) that includes various links and switches (a type of node). The terms "path" and "routing" are used interchangeably herein. A path includes and is composed of any combination of nodes and edges (also referred to herein as links), along which data is transmitted from a source to a destination. According to various aspects and embodiments of the present invention, edge clustering is performed to minimize long edges. Many long edges that pass through narrow corridors between two or more restricted areas (blocking areas) cause cable congestion. Minimizing the routing of long edges helps reduce congestion. According to various aspects and embodiments of the present invention, the length of an edge (link) is measured as the length of the routing cable between the endpoints of the edge.
[0025] As used herein, a node is defined as a distribution point or communication endpoint that is capable of creating, receiving and / or sending information on a communication path or channel. A node may refer to any of the following: a switch, a splitter, a combiner, a buffer, and an adapter. As used herein, splitters and combiners are switches; not all switches are splitters or combiners. As used herein and in accordance with various aspects and embodiments of the present invention, the term "splitter" describes a switch having a single inlet port and multiple outlet ports. As used herein and in accordance with various aspects and embodiments of the present invention, the term "combiner" describes a switch having a single outlet port and multiple inlet ports.
[0026] Reference now Figure 1A , a network on chip (NoC) 100 is shown according to various aspects and embodiments of the present invention. A NoC is a special type of network (or interconnect) used to handle communications between units of a system on a chip (SoC). As described below, switches route communication flows between sources and destinations and perform arbitration. Adapters handle various conversions between data widths, clocks, and power domains. Buffers are used to insert pipeline elements to span long distances, or to store packets to handle rate adaptation between fast senders and slow receivers, or vice versa. The number and precise function of each network element depends on the network on chip technology employed. NoC elements are implemented on the chip as logic functions using digital logic.
[0027] NoC 100 is an example of a network. According to various aspects and embodiments of the present invention, the network includes a set of nodes and a set of edges, each of which has a model and can be used at the core of the topology synthesis process to perform and implement transformations on the network and converge to the best solution suitable for the specified requirements. NoC 100 includes nodes and endpoints and uses assembled basic network functions, such as: network interface units (NIUs) 102, 104, 106, 108, 110, 112, 130, 132 and 134; nodes / switches 114, 116, 118, 120 and 122; buffers, such as buffer 124; and adapters, such as adapter 126. The basic functions of the NoC (represented by the components that implement these functions) communicate with each other using an internal NoC transport protocol based on packet transmission. The NIU converts the protocol used by the attached SoC unit (IP bloc) to the transport protocol used inside the NoC. The NoC basic network functions use an internal transport protocol specific to the NoC 100 to communicate with each other, usually based on packet transmission. The NIU converts the protocol used by an attached system-on-chip (SoC) unit (not shown) to the transport protocol used internally by the NoC 100. The switch routes the communication flow between sources and destinations. Buffers 124 are used to insert pipeline elements to span long distances, or to store packets to handle rate adaptation between fast senders and slow receivers, or vice versa. Adapters 126 handle various conversions between data width, clock, and power domains.
[0028] Reference now Figure 1B , a NoC 150 with various elements, such as NIUs, switches, and blocking areas, is shown in a floorplan. NoC 150 includes various connected elements through various switches. According to one aspect of the present invention, a set of constraints are used as input to the tool, which will be discussed in more detail below. According to some aspects of the present invention, the tool performs a set of sub-steps and produces a description of the resulting NoC, such as NoC 150, and its configured elements and the location of each element on the floorplan (topological synthesis). The generated description is used to actually implement the NoC hardware, using the generated physical information to provide guidance for the back-end implementation process.
[0029] Reference now Figure 2, discloses a process for determining a NoC quality metric (quantitative network quality metric) according to various aspects and embodiments of the present invention. In step 210, a network description and constraints for a network (e.g., a NoC) are loaded into a memory. The description of the NoC can be loaded from any source, such as a disk, or transmitted from a remote location. The format in which the data is encoded can be any format, and the scope of the present invention is not limited in this regard. In step 212, the process selects a source and sink pair (R, S). In step 214, routing information between (R, S) is retrieved for analysis. The routing information includes a sequence of switches through which the route passes. In step 216, the shortest path (or cable length) for the best path is determined or calculated. The length of the shortest path for the pair (R, S) is determined as L0. In step 218, the actual path or cable length for the pair (R, S) is calculated when passing through the NoC from the source to the destination or sink. This is determined by adding all consecutive paths along the route between consecutive nodes passed through by the route. The total actual cable length or path length is determined and identified as L1. With respect to Figure 3A and Figure 3B This is discussed in more detail.
[0030] Therefore, the minimum distance (shortest path) is calculated for each source (connected to the example network, NoC) that communicates with the destination. According to one aspect of the present invention, the minimum distance algorithm that determines the distance from the source to the destination takes into account the blocked areas that cannot have the layout plan of logic and cables. The minimum distance algorithm can use any well-known routing algorithm such as Djikstra, A*, etc., and the scope of the present invention is not limited in this regard. Once a minimum distance is determined, it is used as a reference.
[0031] According to some aspects and embodiments of the present invention, L0 and L1 are stored in a memory. At step 220, the difference between the actual path (L1) and the ideal path (L0) is determined to generate a deviation of the route of the pair (R, S). The deviation is stored in the memory and mapped for the pair (R, S). At step 230, the process determines whether all possible pairs (R, S) have been processed. If not, then at step 232, another pair (R1, S1) is selected and the process returns to step 214. This is repeated until all pairs (R, S) have been processed. n , S n ). In other words, the length of the path taken from source to destination (through the NoC) is calculated. In addition, the minimum distance for the same pair of (source, destination) points is calculated. The length of the actual path and the length of the minimum distance are compared to determine the difference for this pair of (source, destination) points. This difference represents the difference path metric. This is performed for all pairs of (source, destination) and a set of differences is generated. Once all pairs have been processed, as determined in step 230, the process then uses all pairs (Rn , S n ) to generate a histogram representation of the calculated differences and calculate the average difference.
[0032] According to one aspect of the invention, this set of differences is used to generate a report using an average difference. In addition, according to various aspects and embodiments of the invention, this set of differences can be reported as a list of differences shown as a sorted list or a histogram. The results of the differences are also used to generate an average of the differences. This produces a type of quality metric.
[0033] The quality metric is a representation of the average of the difference between the theoretical minimum distance on the floorplan and the actual distance that a signal travels from source to destination through the NoC or network. The quality metric is an indication of how good (or bad) the latency through the NoC is. The latency through the NoC measures the time it takes for a message to traverse or traverse the NoC from source to destination. Latency is affecting the performance of components such as CPUs connected to the NoC. Therefore, the smaller the difference between the actual distance and the minimum distance, the better the performance of the NoC; in other words, the lower the latency.
[0034] Reference now Figure 3A and Figure 3B , according to an example of one aspect and embodiment of the present invention, a floorplan 300 of a network is shown with a congestion area 310, a source or initiator 320 (R), which communicates with a destination or target 330 (S) via a route through a switch 322 in the network or NoC. The route is represented by a flight path 340 (dashed line). The flight path 340 is a visual representation of the connection between the initiator 320 and the target 300. The flight path 340 is represented by a minimum distance path 350. The minimum distance path 350 is compared to the actual path 352 and determined by a minimum distance finding algorithm (A* in this non-limiting example). According to some aspects and embodiments of the present invention, the algorithm is a combination of such a minimum distance algorithm and other heuristics, such as heuristics used to create a routing graph.
[0035] According to one aspect and embodiment of the present invention, the actual path 352 includes four segments of lengths l0, l1, l2, and l3 connected via switch 322. The actual total path length is L1, which is determined as follows:
[0036] L1=l0+l1+l2,+l3
[0037] The segments of length l0, l1, l2, and l3 are Manhattan paths that use both vertical and horizontal segments for routing. The difference between the length L1 (actual path 352) and the minimum distance path L0 (minimum or optimal path 350) is the difference between the aggregate length of the segments of the actual path and the aggregate length of the optimal network path. This process uses Figure 2 The lengths of these paths are calculated during the process.
[0038] Reference now Figure 4 , a process for determining another NoC quality metric or a quantitative network quality metric according to various aspects and embodiments of the present invention is shown. At step 410, segments of the NoC or network, also referred to as edges, are stored in a memory along with NoC constraints. The format for encoding the NoC description does not limit the scope of the present invention. As shown herein, at step 410, a set of segments (for a pair of source, sink (R, S)) is stored in a memory. The stored set of segments is examined to cluster links or edges.
[0039] According to one aspect and embodiment of the present invention, in step 420, edge (segment) merging or edge clustering is performed. The purpose of edge clustering is to minimize resources and improve performance goals through appropriate algorithms and techniques. According to some aspects of the present invention, edge clustering is combined with node clustering and applied in coordination. Edge clustering and node clustering can be used in combination by mixing, simultaneous application or sequential application. Its advantages and goals are to expand the spectrum of topological synthesis and span a larger solution space for the network.
[0040] The clustering process exploits the proximity and direction of the flow of transmitted packets: when two segments (edges) are close enough and carry packets in the same direction, one is removed and only the other is kept. If two segments S1 and S2 are close in the same direction and have very different lengths, the longest segment (e.g. S2) is split into two or three sub-segments (e.g. S2.1, S2.2, S2.3) so that one of the sub-segments (e.g. S2.2) has the same length as the other segment (e.g. S1) to be merged. Clustering is then attempted. The clustering is based on an assimilation scheme (merging) between sets of horizontal or vertical edges with packets transmitted in the same direction, such as from left to right, right to left, top to bottom, or bottom to top.
[0041] According to one embodiment and aspect of the present invention, the clustering algorithm considers other requirements, such as performance or avoiding deadlock, to decide when to assimilate two segments and cluster the segments. The clustering algorithm tries to pair up as many segments as possible. The clustering mechanism continues until no more segments are close to each other.
[0042] At step 430, the difference between the actual path length and the merged path length is calculated. The aggregate length of the set of segments left after clustering represents the "minimal" network represented by cables; therefore, it gives an idea of the number of cables used by the NoC compared to the theoretical minimum. More cables result in more logic gates required and more area. Therefore, the closer the cables used by the minimal network are to the theoretical minimum, the better the NoC.
[0043] At step 440, according to one aspect and embodiment of the present invention, the average difference (the difference between the actual network aggregate length and the theoretical minimum network aggregate length) is calculated and presented. This information is used to generate a report based on the average difference. In addition, according to various aspects and embodiments of the present invention, this set of differences can be reported as a list of differences shown as a sorted list or a histogram. The result is an average value of the differences that represents a quality metric based on the cable lengths derived from the merged edges.
[0044] Reference now Figure 5A and Figure 5B , a block diagram of a floorplan 500 for implementing a process for determining a quality metric according to various aspects and embodiments of the present invention is shown. Floorplan 500 includes a blocking region 510. Two sources 520 communicate with two targets 530 using edges 550 and 540. The process merges edge 540 to produce edge 542. The process merges edge 550 to produce edge 552.
[0045] Reference now Figures 6 to 9 , according to various aspects and embodiments of the present invention, a process for merging or collapsing edges is shown in various stages. An initiator communicates with multiple targets, as shown. The flight paths are shown as dashed lines and are not labeled for clarity. Specifically, an example of a routing or communication path between an initiator (INIT1) and a target (TARG3) is shown, with the connection mapping as Figure 6 As shown, only the flight path or routing is used. Figure 7 The logical edges of the network (shown as flight paths with dashed lines) and the switches connecting the edges are shown. As shown, INIT1 communicates with TARG3 using a direct flight path as an option. Also shown, INIT1 communicates with TARG3 using a logical route that passes through three segments connected by two switches. Any number of paths and switches are contemplated for connecting INIT1 and TARG3, all of which are within the scope of various aspects and embodiments of the present invention. Figure 8 A detailed segment representing a possible edge connecting INIT1 and TARG3 is shown. Fig. 9The best path P13 connecting INIT1 and TARG3 is shown. P13 represents the cable length of the best conflict-free path directly connecting INIT1 and TARG3. In addition to the best conflict-free path, there is a route p1-p2-p3 from INIT1 to TARG3 that passes through switches sw1 and SW2. This route is the actual route through the NoC. The sum of the cable lengths of the route p1-p2-p3 is the sum of the cable lengths of all conflict-free paths along the route from INIT1 to sw1 to sw2 to TARG3. Routing implements connectivity. According to aspects of the present invention, a route is an ordered list of network elements, one for each pair (initiator, target) and one for each pair (target, initiator). A route indicates how the traffic between two pairs will flow and through which elements.
[0046] Therefore, the total cable length from INIT1 to TARG3 is expressed as follows:
[0047] WL(INIT1,TARG3)=WL(p1)+WL(p2)+WL(p3)
[0048] The total optimal WL from INIT1 to TARG3 is expressed as follows:
[0049] Optimal WL=WL(P13)
[0050] According to various aspects and embodiments of the present invention, the same process is applied to all initiator-target pairs represented by connections in the connection map to determine the quality metric.
[0051] Certain methods according to various aspects of the present invention may be performed by instructions stored on a non-transitory computer-readable medium. The non-transitory computer-readable medium stores code including instructions that, if executed by one or more processors, will cause a system or computer to perform the steps of the methods described herein. Non-transitory computer-readable media include: rotating magnetic disks, rotating optical disks, flash random access memory (RAM) chips, and other mechanically moving or solid-state storage media. Any type of computer-readable medium is suitable for storing code including instructions according to various examples.
[0052] Certain examples have been described herein, and it will be noted that different combinations of different components from different examples are possible. In order to better explain the examples, significant features are set forth; however, it is apparent that certain features may be added, modified, and / or omitted without modifying the functional aspects of the examples described.
[0053] The various examples are methods of behavior using any machine or combination of machines. The method examples are complete anywhere in the world where most of the constituent steps appear. For example, according to various aspects and embodiments of the present invention, the IP elements or units include: a processor (e.g., a CPU or GPU), a random access memory (RAM-e.g., off-chip dynamic RAM or DRAM), a network interface for wired or wireless connection, such as Ethernet, WiFi, 3G, 4G Long Term Evolution (LTE), 5G and other wireless interface standard radios. The IP may also include various I / O interface devices, such as those required by different peripheral devices, such as touch screen sensors, geolocation receivers, microphones, speakers, Bluetooth peripherals, and USB devices such as keyboards and mice. By executing instructions stored in RAM devices, the processor performs the steps of the method described herein.
[0054] Some examples are one or more non-transitory computer-readable media arranged to store such instructions for the methods described herein. An example can be implemented by any machine storing a non-transitory computer-readable medium including any necessary code. Some examples can be implemented as: a physical device, such as a semiconductor chip; a hardware description language representation of the logical or functional behavior of such a device; and one or more non-transitory computer-readable media, which are arranged to store such hardware description language representations. The description of the principles, aspects, and embodiments described herein includes their structural and functional equivalents. Elements described as coupled herein have an effective relationship that can be achieved by direct or indirect connection with one or more other intervening elements.
[0055] Those skilled in the art will recognize many modifications and variations. The modifications and variations include any relevant combination of the disclosed features. The description of the principles, aspects and embodiments described herein includes their structural and functional equivalents. The elements described herein as "coupled" or "communicatively coupled" have an effective relationship that can be realized by direct connection or indirect connection, and the direct connection or indirect connection uses one or more other intervening elements. The embodiments described herein as "communicating with" or "communicating with" another device, module or element include any form of communication or link, and include an effective relationship. For example, a wired connection, a wireless protocol, a near field protocol or an RFID can be used to establish a communication link.
[0056] Whenever the terms "includes," "comprising," "having," "having," or variations thereof are used in the detailed description and claims, these terms are intended to be inclusive in a manner similar to the term "comprising."
[0057] Therefore, the scope of the present invention is not intended to be limited to the exemplary embodiments shown and described herein. Rather, the scope and spirit of the present invention are embodied by the appended claims.
Claims
1. A method for calculating a quantitative network quality metric using at least one total cable length quality metric, the method include: receiving, at a tool, network information for the network, the network connecting a plurality of initiators and a plurality of targets using a plurality of routes, wherein the network includes a plurality of edges such that each initiator-target pair is connected by one or more of the plurality of edges along the routes; determining an actual path length for each initiator-target pair based on the edges in a route connecting the initiator of each of the initiator-target pairs to the target; Merging the plurality of edges, each edge in the plurality of edges being part of a different route for a different initiator-target pair, the plurality of edges being clustered based on proximity and direction to generate a plurality of merged edges, the plurality of merged edges constituting a plurality of merged routes; determining a difference between each route in the plurality of routes and a corresponding merged route in the plurality of merged routes; as well as A visual representation of the difference is generated, wherein the visual representation is at least one overall cable length quality metric of the design of the network. The method of claim 1 , wherein the network information comprises segment information. The method according to claim 1 , wherein the network is a network on chip (NoC).
4. The method of claim 1, wherein the step of merging comprises conditionally clustering any edges based on performance requirements.
5. The method according to claim 1, wherein the step of combining include: If the resulting merged routes produce deadlock, clustering is avoided. The method according to claim 1 , wherein the total cable length of all remaining edges in the plurality of merged edges is the minimum network total cable length.
7. A system comprising a non-transitory computer readable medium storing code such that when the code is executed by one or more processors to calculate quantitative network quality metrics including at least one total cable length quality metric, the system: receiving network information for a network, the network connecting a plurality of initiators and a plurality of targets using a plurality of routes, wherein the network comprises a plurality of edges such that each initiator-target pair is connected by one or more edges of the plurality of edges along the routes; determining an actual path length for each initiator-target pair based on the edges in a route connecting the initiator of each of the initiator-target pairs to the target; Merging the plurality of edges, each edge in the plurality of edges being part of a different route for a different initiator-target pair, the plurality of edges being capable of being merged based on proximity and direction to generate a plurality of merged edges, the plurality of merged edges constituting a plurality of merged routes; determining a difference between each route in the plurality of routes and a corresponding merged route in the plurality of merged routes; as well as A visual representation of the difference is generated, wherein the visual representation is at least one overall cable length quality metric of the design of the network.
Citation Information
Patent Citations
Application-oriented network on chip generation method based on regular topology database
CN101847168A
NoC routing method guaranteeing service quality
CN103618673A
Network-on-chip topological structure and implementation method thereof
CN111104775A
EDITING A NoC TOPOLOGY ON TOP OF A FLOORPLAN
US20170177778A1
