Scheduling strategy generation method and system based on same identification request

Abstract

The invention discloses a scheduling strategy generation method and system based on requests with the same identification. The method comprises the following steps: acquiring a basic identifier transmission request and a current identifier transmission demand corresponding to target master equipment through an identifier request processing device, performing scene matching on the current identifier transmission demand based on a preset scene division rule, and determining a target scene type; determining a target scheduling strategy and a basic processing unit set corresponding to the basic identifier transmission request based on the target scene type through a data decision module; and performing reinforcement processing on the basic identifier transmission request based on the target scheduling policy through the basic processing unit set to obtain a target identifier transmission request, and executing the target identifier transmission request to obtain a target response result corresponding to the target master device. Through the technical scheme of the invention, on the premise of maintaining the correctness of the request response sequence, the same-identifier advanced transmission requests can be efficiently and dynamically scheduled and managed, and the processing efficiency of the same-identifier advanced transmission requests is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data transmission, and in particular to a scheduling strategy generation method and system based on same-identifier requests. BACKGROUND

[0002] With the development of high-performance computing and complex on-chip systems, the Advanced extensible Interface Coherency Extension (ACE) protocol has become a key technology to ensure cache coherency in multi-core processors and multi-master device systems. This protocol and its extension technology are widely used in fields that have strict requirements for data throughput, consistency, and order, such as cloud computing, industrial control, and aerospace. In these scenarios, the system needs to handle high-concurrency data access and real-time control instructions simultaneously, ensuring efficient and reliable data collaboration under conditions of heterogeneous computing power of multiple master device nodes, dynamic changes in business size, and a high mix of read and write operations.

[0003] To achieve higher bus utilization and system throughput, the same-identifier advance transmission request technology is introduced. This technology allows subsequent requests to be initiated for requests with the same identifier without receiving the response of the previous request, thereby hiding access delays. However, existing technologies face significant challenges when handling such advance transmission requests. First, when multiple same-identifier requests are concurrent, it is easy to cause fierce competition for bandwidth and queue resources in the interconnection network, resulting in some requests being discarded or severely delayed. Second, due to dynamic differences in transmission paths and uneven loads of each computing node, the response return order is often inconsistent with the request sending order. This directly causes business logic errors for downstream applications that rely on strict data order.

[0004] To avoid the out-of-order problem, the current mainstream solution usually forces the number of in-transit requests for any identifier to be single, i.e., blocks the issuance of subsequent same-identifier requests until the response of the previous request is returned. This solution ensures order, but in low-latency, high-throughput demand scenarios, it severely restricts overall bandwidth and often becomes a system performance bottleneck. In addition, existing technologies have obvious shortcomings in architecture flexibility, energy efficiency, and reliability. Core control parameters are mostly statically hard-coded, and cannot be dynamically adapted and adjusted according to the characteristics of master device nodes or real-time business load, resulting in idle resources under light load and performance degradation under overload, making it difficult to adapt to diverse deployment scales. At the same time, existing technologies generally lack a differentiation mechanism for identifier capabilities, and all identifiers are assumed to support advance transmission, resulting in unnecessary logical resources and power waste for requests and responses of identifiers that do not require such transmission.

[0005] Therefore, how to efficiently schedule and manage the same-identity pre-transmission request under the premise of maintaining the correctness of the request response sequence, and realize the collaborative optimization of throughput, flexibility, energy efficiency and reliability, has become a key technical problem to be solved in the field. SUMMARY

[0006] The application provides a scheduling strategy generation method and system based on same-identity requests, which can solve the problem of low processing efficiency of same-identity pre-transmission requests.

[0007] According to an aspect of the application, a scheduling strategy generation method based on same-identity requests is provided, comprising:

[0008] The target master device corresponding basic identity transmission request and current identity transmission demand are obtained by the identity request processing device, the current identity transmission demand is matched with a scene based on a preset scene division rule, the target scene type corresponding to the target master device is determined, and the target scene type is sent to a data decision module;

[0009] The target scheduling strategy and basic processing unit set corresponding to the basic identity transmission request are determined by the data decision module based on the target scene type, and the target scheduling strategy is sent to the basic processing unit set;

[0010] The basic identity transmission request is enhanced processed by the basic processing unit set based on the target scheduling strategy, the target identity transmission request corresponding to the basic identity transmission request is obtained, and the target identity transmission request is executed to obtain the target response result corresponding to the target master device.

[0011] According to another aspect of the application, a scheduling strategy generation system based on same-identity requests is provided, comprising:

[0012] An identity request processing device is configured to obtain a target master device corresponding basic identity transmission request and current identity transmission demand, match the current identity transmission demand with a scene based on a preset scene division rule, determine a target scene type corresponding to the target master device, and send the target scene type to a data decision module;

[0013] A data decision module is configured to determine a target scheduling strategy and a basic processing unit set corresponding to the basic identity transmission request based on the target scene type, and send the target scheduling strategy to the basic processing unit set;

[0014] A set of basic processing units is configured to perform enhanced processing on the basic identification transmission request based on the target scheduling strategy, to obtain a target identification transmission request corresponding to the basic identification transmission request, and to execute the target identification transmission request to obtain a target response result corresponding to the target master device.

[0015] The technical solution of the embodiment of the present application obtains the basic identification transmission request and the current identification transmission demand corresponding to the target master device through the identification request processing device, performs scene matching on the current identification transmission demand based on a preset scene division rule, determines the target scene type corresponding to the target master device, and sends the target scene type to the data decision module. Then, the data decision module determines the target scheduling strategy and the set of basic processing units corresponding to the basic identification transmission request based on the target scene type, and sends the target scheduling strategy to the set of basic processing units. Finally, the set of basic processing units performs enhanced processing on the basic identification transmission request based on the target scheduling strategy, to obtain a target identification transmission request corresponding to the basic identification transmission request, and executes the target identification transmission request to obtain a target response result corresponding to the target master device. Due to the use of the differential screening mechanism, the identification with advanced transmission capability and the identification without advanced transmission capability can be accurately identified. The request and response of the latter are directly transmitted without entering the processes such as substitution, reallocation, and sorting, which significantly reduces unnecessary resource occupation, improves system resource utilization efficiency, solves the problem of low processing efficiency of the same-identification advanced transmission request, and improves the processing efficiency of the same-identification advanced transmission request while maintaining the correctness of the request and response sequence.

[0016] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0018] Figure 1 is a flowchart of a same-identification request-based scheduling strategy generation method according to the first embodiment of the present application;

[0019] Figure 2 is a flowchart of a same-identification request-based scheduling strategy generation method according to the second embodiment of the present application;

[0020] Figure 3 is a flow chart of an optional same-identity request-based scheduling strategy generation method according to Embodiment Two of the present application;

[0021] Figure 4 is a structural schematic diagram of a same-identity request-based scheduling strategy generation system according to Embodiment Three of the present application. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should fall within the scope of protection of the present application.

[0023] It should be noted that the terms "first", "second", "target", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0024] Embodiment One

[0025] Figure 1 A flow chart of a same-identity request-based scheduling strategy generation method according to Embodiment One of the present application is provided, and the present embodiment can be applicable to the case of dynamically scheduling same-identity advance transmission requests. The method can be executed by a same-identity request-based scheduling strategy generation system, which can be realized in the form of hardware and / or software. As shown in the figure, the method comprises the following steps. Figure 1

[0026] S110, obtaining, by an identity request processing apparatus, a basic identity transmission request corresponding to a target master device and a current identity transmission demand, performing scene matching on the current identity transmission demand based on a preset scene division rule, determining a target scene type corresponding to the target master device, and sending the target scene type to a data decision module.

[0027] ​The identification request processing apparatus can refer to an apparatus pre-constructed for scheduling processing of data access requests. Optionally, the identification request processing apparatus comprises a parameter configuration unit, a mode switching unit, a channel adaptation unit, an exception processing unit, and a basic processing unit set; the basic processing unit set comprises a request identification filtering unit, a request identification proxy unit, a request arbitration unit, a response filtering unit, a response reordering unit, and a response arbitration unit. The parameter configuration unit can refer to a logic module for dynamically setting and adjusting key behavior parameters of system runtime. The mode switching unit can refer to a logic module for dynamically switching between different pre-defined behavior modes. The channel adaptation unit can refer to a logic module for dynamically switching between different pre-defined working channels. The exception processing unit can refer to a logic module responsible for detecting, responding to, and recording unexpected or error events occurring in the system. The basic processing unit can refer to a logic module that is initially selected to actually participate in the scheduling processing of data access requests. The basic processing unit set can refer to a set composed of various basic processing units in the same identification request processing apparatus. The request identification filtering unit can refer to a logic module for identifying request identification with early transmission capability. The request identification proxy unit can refer to a logic module for converting request identification with early transmission capability into new identification. The request arbitration unit can refer to a logic unit for priority division of data access requests. The response filtering unit can refer to a logic module for identifying the identification type of response identification. The response reordering unit can refer to a logic module for reordering response requests according to the order of data access requests. The response arbitration unit can refer to a logic module for priority division of response replies.

[0028] The master device can refer to a component initiating a data access transaction. For example, a central processing unit (CPU) core, a direct memory access (DMA) controller, or a graphics processing unit (GPU). The slave device can refer to a component responding to the master device transaction. For example, a memory controller, a peripheral register, or a read-only memory. The target master device can refer to a master device selected for data access. Generally, the target master device can be determined according to actual application requirements. The identification transmission request can refer to a service request carrying an identification. For example, the identification transmission request can be a data access request carrying an identity document (ID). Generally, the identification transmission request can be a same-identification advance transmission request or a normal identification transmission request. The basic identification transmission request can refer to a preliminary generated identification transmission request. For example, the basic identification transmission request can be a data access request initiated by the target master device. The identification transmission requirement can refer to data information used to limit the data transmission scenario. The current identification transmission requirement can refer to the identification transmission requirement corresponding to the target master device at the current time. Generally, the current identification transmission requirement can be determined according to actual application requirements, which is not limited in the embodiments of the present application.

[0029] The preset scenario division rule can refer to a rule preset for dividing the transmission scenario of the data access request by type. The scenario type can refer to the transmission scenario corresponding to the identification transmission request. The target scenario type can refer to the scenario type matched by the basic identification transmission request corresponding to the target master device. Optionally, the target scenario type includes a target work type and a target channel type. The target work type includes a non-blocking mode or a blocking mode. The target channel type includes a read request type or a write request type. The target work type can refer to the work behavior type corresponding to the target master device. The non-blocking mode can refer to a work type in which the master device can continue to issue a subsequent request without waiting for a response to a previous request. The blocking mode can refer to a work type in which the master device cannot issue a request with the same identification before receiving a response to a previous request. The target channel type can refer to the running channel type corresponding to the target master device. The read request type can refer to a running channel type in which the master device requests data from the slave device. The write request type can refer to a running channel type in which the master device provides data to the slave device.

[0030] In S120, the target scheduling strategy corresponding to the basic identification transmission request and a basic processing unit set are determined based on the target scenario type by a data decision module, and the target scheduling strategy is sent to the basic processing unit set.

[0031] The data decision module can be a logic module for determining the processing logic corresponding to the identification transmission request. For example, the data decision module can be a processor. The scheduling strategy can be a strategy rule for defining the data processing flow of the identification transmission request. Generally, the scheduling strategy can include the data processing sequence and specific data processing content of each basic processing unit. The target scheduling strategy can be the scheduling strategy corresponding to the basic identification transmission request.

[0032] S130, performing the target identification transmission request based on the target scheduling strategy by the basic processing unit set to obtain the target response result corresponding to the target master device.

[0033] The enhancement processing can be an operation of substituting the request identification in the basic identification transmission request. Generally, different request identifications correspond to different enhancement processing operations. For example, if the request identification is a same-identification look-ahead transmission request, the enhancement processing can be generating a dedicated code for each look-ahead transmission-capable identification and managing a ring buffer for identifying the request sending sequence. If the request identification is not a same-identification look-ahead transmission request, the enhancement processing can be simply adding a flag bit to the request identification. The target identification transmission request can be an identification transmission request obtained after the enhancement processing of the basic identification transmission request. The response result can be a feedback result obtained after the master device corresponding to the slave device responds to the identification transmission request. For example, if the identification transmission request is "please give me the data at address 0x100", the response result can be "the data at address 0x100 you want is here". Similarly, if the identification transmission request is "please write the data into address 0x200", the response result can be "the data has been successfully written into address 0x200". The target response result can be the response result corresponding to the target master device.

[0034] The technical scheme of the embodiment of the application obtains the basic identification transmission request corresponding to the target master device and the current identification transmission demand through the identification request processing device, performs scene matching on the current identification transmission demand based on a preset scene division rule, determines the target scene type corresponding to the target master device, and sends the target scene type to the data decision module. Then, the data decision module determines the target scheduling strategy corresponding to the basic identification transmission request and the basic processing unit set based on the target scene type, and sends the target scheduling strategy to the basic processing unit set. Finally, the basic processing unit set performs strengthened processing on the basic identification transmission request based on the target scheduling strategy, obtains the target identification transmission request corresponding to the basic identification transmission request, and executes the target identification transmission request to obtain the target response result corresponding to the target master device. Due to the use of the differentiated screening mechanism, the identification with the advanced transmission capability and the identification without the advanced transmission capability can be accurately identified. The request and response of the latter are directly transmitted without entering the processes such as substitution, reallocation, and sorting, which significantly reduces unnecessary resource occupation, improves system resource utilization efficiency, solves the problem of low processing efficiency of the same-identification advanced transmission request, and improves the processing efficiency of the same-identification advanced transmission request under the premise of maintaining the correctness of the request response sequence.

[0035] Embodiment two

[0036] Figure 2 A flowchart of a same-identification request-based scheduling strategy generation method provided by the second embodiment of the application is provided. The present embodiment is based on the above-described embodiment and is refined. In the present embodiment, the “strengthened processing of the basic identification transmission request by the basic processing unit set based on the target scheduling strategy, the target identification transmission request corresponding to the basic identification transmission request, the execution of the target identification transmission request, and the target response result corresponding to the target master device” is refined. Specifically, the method can include: identification matching of the basic identification transmission request by the request identification filtering unit based on the advanced transmission identification rule, determination of the identification matching result corresponding to the basic identification transmission request, determination of the target processing unit set corresponding to the basic identification transmission request based on the identification matching result, strengthened processing of the basic identification transmission request by the target processing unit set based on the target scheduling strategy, the target identification transmission request corresponding to the basic identification transmission request, the execution of the target identification transmission request, and the target response result corresponding to the target master device. As shown in FIG. 8, the method includes: Figure 2

[0037] S210, obtaining the basic identification transmission request corresponding to the target master device and the current identification transmission demand through the identification request processing device.

[0038] ​In an optional embodiment, before the target master device corresponding basic identification transmission request and current identification transmission demand are acquired by the identification request processing apparatus, the method further comprises: constructing a pre-transmission identification rule based on a preset identification set by a parameter configuration unit, and synchronously processing the pre-transmission identification rule and transmitting the pre-transmission identification rule to a basic processing unit set. The preset identification can refer to information used for representing a request identification with pre-transmission capability. For example, the preset identification can include an identification value, i.e., a specific ID value, an identification quantity, and a pre-transmission capability of the same identification. The preset identification set can refer to a set composed of various preset identifications. The pre-transmission identification rule can refer to an identification recognition rule constructed based on the preset identification set. Specifically, before a scheduling strategy generation process of the same identification request is performed, the pre-transmission identification rule can be constructed based on the preset identification set by the parameter configuration unit, and the pre-transmission identification rule can be transmitted to a request identification filtering unit, a request identification substitution unit, and a response reordering unit in the basic processing unit set, so as to determine which ID needs to enter the substitution and reordering process, and which ID can be directly transmitted, thereby providing an effective basis for subsequent operations.

[0039] S220, performing mode matching on the current identification transmission demand based on a preset mode division rule by a mode switching unit, determining a target working type corresponding to the target master device, and sending the target working type to a data decision module.

[0040] The preset mode division rule can refer to a rule used for specifically defining a mode matching process. For example, if the current identification transmission demand is low delay and high throughput, the target working type is determined as a non-blocking mode; if the current identification transmission demand is strong sequentiality and high reliability, the target working type is determined as a blocking mode.

[0041] S230, performing channel division on the current identification transmission demand based on a preset channel division rule by a channel adaptation unit, determining a target channel type corresponding to the target master device, and sending the target channel type to the data decision module.

[0042] The preset channel division rule can refer to a rule used for specifically defining a channel division process. For example, if the current identification transmission demand is read data, the target channel type is determined as a read request type; if the current identification transmission demand is write data, the target channel type is determined as a write request type.

[0043] Specifically, after the target master device corresponding current identification transmission requirement is acquired by the identification request processing apparatus, the current identification transmission requirement can be matched with a preset mode division rule by the mode switching unit to determine the target working type corresponding to the target master device, and the target working type is sent to the data decision module. At the same time, the current identification transmission requirement is divided into channels based on a preset channel division rule by the channel adaptation unit to determine the target channel type corresponding to the target master device, and the target channel type is sent to the data decision module. Therefore, through the "non-blocking first response then rearrangement" and "blocking waiting-response in order recycling" dual mode, real-time switching of the master device configuration signal can be supported, different scene requirements can be adapted without restarting the system, low delay and high reliability are considered, and multiple independent modules do not need to be deployed, which effectively reduces the hardware cost, and improves the maintenance efficiency.

[0044] S240, if the target working type is a non-blocking mode and the target channel type is a read request type, the first request processing strategy is used as the target scheduling strategy by the data decision module.

[0045] Among them, the request processing strategy can refer to the rule for limiting the execution order and specific execution operation of each processing unit. The first request processing strategy can refer to the request processing strategy under the condition that the target working type is a non-blocking mode and the target channel type is a read request type. For example, the first request processing strategy can be: first, the identification transmission request is transmitted to the request identification filtering unit, then the identification transmission request is transmitted to the request identification replacement unit or directly transmitted to the request arbitration unit through the request identification filtering unit, and then the identification transmission request is transmitted to the slave device corresponding to the target master device through the request arbitration unit, the identification transmission request is executed through the slave device and the response result is fed back, and the response result is transmitted to the response reordering unit or directly transmitted to the response arbitration unit through the response filtering unit.

[0046] It is worth noting that in the embodiment of the application, the request arbitration unit in the first request processing strategy is only responsible for the data transmission function.

[0047] S250, if the target working type is a non-blocking mode and the target channel type is a write request type, the second request processing strategy is used as the target scheduling strategy by the data decision module.

[0048] The second request processing strategy can refer to a request processing strategy under a target work type of a non-blocking mode and a target channel type of a write request type. For example, the second request processing strategy can be: first transmitting an identification transmission request to a request identification filtering unit, then transmitting the identification transmission request to a request identification substitution unit through the request identification filtering unit or directly transmitting the identification transmission request to a request arbitration unit, then transmitting the identification transmission request to a slave corresponding to a target master through the request arbitration unit, receiving write data from the target master through a channel adaptation unit and sending the write data to the slave, executing the identification transmission request through the slave and feeding back a response result, and transmitting the response result to a response reordering unit through a response filtering unit or directly transmitting the response result to a response arbitration unit.

[0049] It is worth noting that in the embodiment of the application, the request arbitration unit in the second request processing strategy is only responsible for a data transmission function.

[0050] S260, if the target work type is a blocking mode and the target channel type is a read request type, a third request processing strategy is taken as a target scheduling strategy through a data decision module.

[0051] The third request processing strategy can refer to a request processing strategy under a target work type of a blocking mode and a target channel type of a read request type. For example, the third request processing strategy can be: first transmitting an identification transmission request to a request identification filtering unit, then transmitting the identification transmission request to a request identification substitution unit through the request identification filtering unit or directly transmitting the identification transmission request to a request arbitration unit, then determining a priority of the identification transmission request based on a polling arbitration mechanism through the request arbitration unit and transmitting the identification transmission request to a slave corresponding to a target master based on the priority, executing the identification transmission request through the slave and feeding back a response result, and transmitting the response result to a response reordering unit through a response filtering unit or directly transmitting the response result to a response arbitration unit.

[0052] S270, if the target work type is a blocking mode and the target channel type is a write request type, a fourth request processing strategy is taken as a target scheduling strategy through a data decision module.

[0053] The fourth request processing strategy can refer to the request processing strategy when the target working type is blocking mode and the target channel type is write request type. For example, the fourth request processing strategy can be as follows: first, the identifier transmission request is sent to the request identifier filtering unit, then the identifier transmission request is sent to the request identifier replacement unit or directly to the request arbitration unit through the request identifier filtering unit. After that, the request arbitration unit determines the priority of the identifier transmission request based on the polling arbitration mechanism, and sends the identifier transmission request to the slave device corresponding to the target master device based on the priority. At the same time, the channel adaptation unit caches all write data from the master device and adjusts the order in which the write data is sent to the slave device according to the priority generated by the request arbitration unit. The slave device executes the identifier transmission request and sends back the response result. The response result is sent to the response reordering unit or directly to the response arbitration unit through the response filtering unit.

[0054] Specifically, after determining the target job type through the mode switching unit and the target channel type through the channel adaptation unit, different request processing strategies can be matched as target scheduling strategies based on different combinations of the target job type and the target channel type. Thus, through a dynamically switchable dual-mode architecture—non-blocking mode and blocking mode—it can adapt to multiple scenario requirements. In non-blocking mode, requests can be sent without waiting for preceding responses; after the responses are returned, they are reordered and returned to the master device in the order the requests were sent, adapting to low-latency, high-throughput scenarios. In blocking mode, requests and responses are strictly one-to-one, adapting to high-reliability scenarios, balancing low latency, high throughput, and high reliability with strong consistency requirements, and supporting real-time scenario switching without module reconstruction. Furthermore, core parameters can be dynamically configured to adapt to different master device node characteristics and business scales, improving the solution's versatility and scalability. Simultaneously, it supports integrated processing of read and write across all channels, reducing system deployment complexity and hardware costs, improving exception handling mechanisms, and enhancing system reliability and maintainability.

[0055] S280. The data decision module determines the set of basic processing units based on the target scenario type and sends the target scheduling strategy to the set of basic processing units.

[0056] S290. The request identifier filtering unit performs identifier matching on the basic identifier transmission request based on the advance transmission identifier rule, determines the identifier matching result corresponding to the basic identifier transmission request, and determines the target processing unit set corresponding to the basic identifier transmission request based on the identifier matching result.

[0057] The identity matching result can be a matching consistency or a matching inconsistency. If the identity matching result is the matching consistency, the base identity transmission request is the same as the pre-transmission request. If the identity matching result is the matching inconsistency, the base identity transmission request is not the same as the pre-transmission request. The target processing unit can be a processing unit in the base processing unit set that finally executes the target scheduling strategy. The target processing unit set can be a set of target processing units corresponding to the same identity matching result.

[0058] In an optional embodiment, the request identity filtering unit determines the target processing unit set corresponding to the base identity transmission request based on the identity matching result, including: if the identity matching result is the matching consistency, the request identity filtering unit determines the request identity replacement unit, the request arbitration unit, the response filtering unit, the response reordering unit, and the response arbitration unit as the target processing unit set corresponding to the base identity transmission request; and if the identity matching result is the matching inconsistency, the request identity filtering unit determines the request arbitration unit, the response filtering unit, and the response arbitration unit as the target processing unit set corresponding to the base identity transmission request. Specifically, after the request identity filtering unit performs the identity matching on the base identity transmission request based on the pre-transmission identity rule, different target processing unit sets can be reconstituted according to different identity matching results. Thus, the problem that all identities in the prior art have pre-transmission capability and all requests and responses corresponding to the identities are processed according to a complex process can be avoided. Even the identities without pre-transmission capability also need to pass through the cache and sorting links, causing a large amount of unnecessary resource consumption.

[0059] S2100, the target processing unit set performs the enhanced processing on the base identity transmission request based on the target scheduling strategy, obtains a target identity transmission request corresponding to the base identity transmission request, and executes the target identity transmission request to obtain a target response result corresponding to the target master device.

[0060] Specifically, after determining the target working type by the mode switching unit and determining the target channel type by the channel adaptation unit, the base identity transmission request can be identified and matched by the request identification filtering unit based on the ahead transmission identity rule to determine an identity matching result corresponding to the base identity transmission request, and a target processing unit set corresponding to the base identity transmission request can be determined based on the identity matching result. The target processing unit set can then perform strengthened processing on the base identity transmission request based on the target scheduling strategy to obtain a target identity transmission request corresponding to the base identity transmission request, and execute the target identity transmission request to obtain a target response result corresponding to the target master device.

[0061] It is worth noting that in the embodiments of the present application, the determination process of the target scheduling strategy and the determination process of the target processing unit set can be executed in parallel or in series, and the above-mentioned embodiments are only used as an explanation and do not specifically limit it.

[0062] S2110, the request sending time corresponding to the base identity transmission request and the response receiving time corresponding to the target response result are obtained by the exception handling unit, and the request sending time and the response receiving time are judged based on a preset time threshold to obtain a state evaluation result corresponding to the target response result.

[0063] The request sending time can refer to the time point at which the target master device sends the base identity transmission request. The response receiving time can refer to the time point at which the target master device receives the target response result. The preset time threshold can refer to a periodic numerical value preset for evaluating the time difference between the response receiving time and the request sending time. The state evaluation result can refer to the judgment result obtained by judging the request sending time and the response receiving time using the preset time threshold. For example, the state evaluation result can be response timeout or response not timeout. Generally, if the time difference between the response receiving time and the request sending time is less than the preset time threshold, the state evaluation result is response not timeout. If the time difference between the response receiving time and the request sending time is greater than or equal to the preset time threshold, the state evaluation result is response timeout.

[0064] Specifically, after the target master device receives the corresponding target response result, the response receiving time corresponding to the target response result is recorded by the exception handling unit, and the time difference between the response receiving time and the request sending time is calculated, and the state evaluation result corresponding to the target response result is obtained by judging the time difference using the preset time threshold. Therefore, the exception handling unit can support the bottom-up strategy of timeout response, which significantly enhances the system stability and maintainability.

[0065] It is worth noting that in the embodiments of the present application, if the state evaluation result is response timeout, the request retransmission or abnormal response reply mechanism can be started for abnormal situations.

[0066] The technical scheme of the embodiment of the present application obtains the basic identification transmission request and the current identification transmission demand corresponding to the target master device through the identification request processing device. The mode switching unit matches the current identification transmission demand based on the preset mode division rule to determine the target working type corresponding to the target master device, and sends the target working type to the data decision module. The channel adaptation unit divides the current identification transmission demand based on the preset channel division rule to determine the target channel type corresponding to the target master device, and sends the target channel type to the data decision module. Then, if the target working type is a non-blocking mode and the target channel type is a read request type, the first request processing strategy is taken as the target scheduling strategy through the data decision module. If the target working type is a non-blocking mode and the target channel type is a write request type, the second request processing strategy is taken as the target scheduling strategy through the data decision module. If the target working type is a blocking mode and the target channel type is a read request type, the third request processing strategy is taken as the target scheduling strategy through the data decision module. If the target working type is a blocking mode and the target channel type is a write request type, the fourth request processing strategy is taken as the target scheduling strategy through the data decision module. Further, the data decision module determines the basic processing unit set based on the target scene type, and sends the target scheduling strategy to the basic processing unit set. Further, the request identification filtering unit matches the basic identification transmission request based on the advanced transmission identification rule to determine the identification matching result corresponding to the basic identification transmission request, and determines the target processing unit set corresponding to the basic identification transmission request based on the identification matching result. The target processing unit set processes the basic identification transmission request based on the target scheduling strategy to obtain the target identification transmission request corresponding to the basic identification transmission request, and executes the target identification transmission request to obtain the target response result corresponding to the target master device. Finally, the abnormal processing unit obtains the request sending time corresponding to the basic identification transmission request and the response receiving time corresponding to the target response result, and judges the state of the request sending time and the response receiving time based on the preset time threshold to obtain the state evaluation result corresponding to the target response result. Due to the use of the differential screening mechanism, the identification with advanced transmission capability and the identification without advanced transmission capability can be accurately identified. The requests and responses of the latter do not need to go through processes such as substitution, reallocation, and sorting, significantly reducing unnecessary resource occupation, improving system resource utilization efficiency, solving the problem of low processing efficiency of the same identification advanced transmission request, and improving the processing efficiency of the same identification advanced transmission request under the premise of maintaining the correctness of the request response sequence.

[0067] Figure 3 An optional flow chart of a scheduling strategy generation method based on the same identification request provided by an embodiment of the application is shown. Specifically, the flow is divided into a configuration phase and a data interaction phase. Among them, the configuration phase needs to provide rules and environment for subsequent data interaction before the host device initiates data access. In the configuration phase, the mode switching unit determines the target working type corresponding to the target host device based on the preset mode division rule for mode matching of the current identification transmission demand, and sends the target working type to the data decision module. The channel adaptation unit determines the target channel type corresponding to the target host device based on the preset channel division rule for channel division of the current identification transmission demand, and sends the target channel type to the data decision module. And, the parameter configuration unit constructs the advance transmission identification rule based on the preset identification set, and synchronously processes the advance transmission identification rule, and transmits the advance transmission identification rule to the basic processing unit set.

[0068] In the data interaction stage: first, the data decision module determines the target scheduling strategy and the basic processing unit set corresponding to the basic identification transmission request based on the target scene type, and sends the target scheduling strategy to the basic processing unit set. If the target work type in the target scene type is a non-blocking mode and the target channel type is a read request type, first compare the request identification in the basic identification transmission request based on the advance transmission identification rule through the request identification filtering unit. If the request identification does not have advance transmission capability, do not process the request, directly transmit the request to the request arbitration unit and then send it to the slave device; if the request identification has advance transmission capability, the request is transmitted to the request identification substitution unit, which will generate a dedicated code for each identification with advance transmission capability, for example, multiple request identifications are 1, which can be re-identified as 11, 12 and 13 according to the order, and manage a ring buffer to identify the request sending order. The ring buffer does not need to record any information, only to manage the read and write pointers. When a new request is replaced, the write pointer is incremented by one, and when the corresponding response is returned, the read pointer is incremented by one, thereby greatly reducing resource consumption. The new identification output by the request identification substitution unit is composed of a dedicated code and a ring buffer write pointer. When the request identification is completed, the original request information is combined to generate a target identification transmission request, and the target identification transmission request is transmitted to the request arbitration unit and then sent to the slave device. When the read response from the slave device returns to the response filtering unit, the response filtering unit will parse the read response identification, and according to the pre-set identification processing rule of the identification without advance transmission capability, for example, there is a flag bit 1 in the highest bit, the response identification is judged, if the identification does not have advance transmission capability, the response is directly transmitted to the response arbitration unit and then returned to the master device; if the identification has advance transmission capability, the response is sent to the response reordering unit, and the dedicated code and the ring buffer pointer of the identification are further parsed. If the pointer is exactly the read pointer of the ring buffer corresponding to the dedicated code, the dedicated code is decoded into the original request identification, i.e. the request identification corresponding to the basic identification transmission request, and the response data is sent to the response arbitration unit and then returned to the master device, and the request identification substitution unit is notified to increment the read pointer of the corresponding ring buffer. If the pointer is not equal to the read pointer of the ring buffer corresponding to the dedicated code, the response data is temporarily stored in the response reordering unit, waiting for the next slave device response to return, until the pointer parsed by the identification and the read pointer of the corresponding ring buffer are consistent. After the response is returned to the master device and the read pointer of the corresponding ring buffer is incremented by one, the read response identification cached in the response reordering unit is compared with the pointer again, and the above operation is repeated to ensure that the same identification request is returned in sequence.

[0069] If the target work type in the target scenario type is non-blocking mode and the target channel type is write request type, the work flow of request processing and response processing is consistent with that when the target work type in the target scenario type is non-blocking mode and the target channel type is read request type, the only difference is that the channel adaptation unit also receives write data from the master device and transmits it to the slave device.

[0070] If the target work type in the target scene type is the blocking mode and the target channel type is the read request type, the request identifier filtering unit first compares the request identifier in the basic identifier transmission request based on the advance transmission identifier rule. If the request identifier does not have the advance transmission capability, the request is not processed and is directly transmitted to the request arbitration unit. The request and the blocked request from the request identifier replacement unit are sent to the slave device based on the polling arbitration mechanism in the request arbitration unit according to the arbitration result. If the request identifier has the advance transmission capability, the request is transmitted to the request identifier replacement unit. The request identifier replacement unit generates a dedicated code for each identifier with the advance transmission capability and manages a ring buffer for controlling the request sending order in the blocking mode. At this time, the ring buffer is used not only to manage the read and write pointers but also to record the request data. If there is still a request with a response not returned in the ring buffer before the request, the request is temporarily stored in the ring buffer until the responses of the requests in front of it are returned. The request identifier replacement unit replaces the new identifier composed of the dedicated code and the pointer of the request with the original request information to generate the target identifier transmission request, and transmits the target identifier transmission request to the request arbitration unit. The request and the blocked request in the ring buffer with the dedicated code of the other identifier with the advance transmission capability from the request identifier replacement unit are arbitrated in the request arbitration unit based on the polling arbitration mechanism, and each request is sent to the slave device according to the arbitration result. When the read response is returned from the slave device to the response filtering unit, the response filtering unit analyzes the read response identifier and judges the identifier based on the pre-set identifier processing rule without the advance transmission capability. If the identifier does not have the advance transmission capability, the response is directly transmitted to the response arbitration unit and then returned to the master device. If the identifier has the advance transmission capability, the response is sent to the response reordering unit for further analysis of the dedicated code of the identifier. Since the request with the dedicated code in the blocking mode is blocked, the pointer corresponding to the identifier must be the read pointer of the ring buffer corresponding to the dedicated code. Therefore, it is not necessary to compare the read pointer of the ring buffer corresponding to the dedicated code and to temporarily store the response data in the response reordering unit. Only the dedicated code of the original request identifier, i.e., the request identifier corresponding to the basic identifier transmission request, is decoded and sent to the response arbitration unit together with the response data to return to the master device, and the read pointer of the corresponding ring buffer is incremented by one to release the next blocked request in the ring buffer with the same identifier.

[0071] If the target work type in the target scene type is the blocking mode and the target channel type is the write request type, the work flow of the request processing and the response processing is consistent with the work flow of the target work type being the blocking mode and the target channel type being the read request type in the target scene type, but in the target scene type, the channel adaptation unit can first cache all the write data from the master device, and according to the arbitration result of the request arbitration unit, adjust the delivery order of the write data, and then, according to the delivery order, transmit the write data to the slave device, so as to ensure that the write data transmission order is strictly consistent with the actual write request delivery order.

[0072] Therefore, through the integrated mechanism of request substitution, identifier re-allocation and response reordering, a closed loop is formed from request screening, identifier reconstruction to response calibration, which effectively avoids the conflict of the same identifier request and improves the accuracy of the response order.

[0073] Embodiment three

[0074] Figure 4 A structural schematic diagram of a scheduling strategy generation system based on the same identifier request is provided for the third embodiment of the application. As shown in the figure, the system comprises an identifier request processing device 310, a data decision module 320 and a basic processing unit set 330. Figure 4

[0075] The identifier request processing device 310 is configured to acquire the basic identifier transmission request corresponding to a target master device and the current identifier transmission demand, perform scene matching on the current identifier transmission demand based on a preset scene division rule, determine the target scene type corresponding to the target master device, and send the target scene type to the data decision module 320.

[0076] The data decision module 320 is configured to determine the target scheduling strategy corresponding to the basic identifier transmission request and the basic processing unit set based on the target scene type, and send the target scheduling strategy to the basic processing unit set 330.

[0077] The basic processing unit set 330 is configured to perform enhanced processing on the basic identifier transmission request based on the target scheduling strategy, obtain the target identifier transmission request corresponding to the basic identifier transmission request, and execute the target identifier transmission request to obtain the target response result corresponding to the target master device.

[0078] ​The technical scheme of the embodiment of the application comprises: the identification request processing device 310 is used for obtaining a basic identification transmission request corresponding to a target master device and a current identification transmission demand, performing scene matching on the current identification transmission demand based on a preset scene division rule, determining a target scene type corresponding to the target master device, and sending the target scene type to the data decision module 320; the data decision module 320 is used for determining a target scheduling strategy corresponding to the basic identification transmission request and a basic processing unit set based on the target scene type, and sending the target scheduling strategy to the basic processing unit set 330; and the basic processing unit set 330 is used for performing strengthened processing on the basic identification transmission request based on the target scheduling strategy, obtaining a target identification transmission request corresponding to the basic identification transmission request, and executing the target identification transmission request to obtain a target response result corresponding to the target master device. Due to the use of the differentiated screening mechanism, the identification with the advanced transmission capability and the identification without the advanced transmission capability can be accurately identified, the request and the response of the latter are directly transmitted, and the latter does not need to enter the processes such as substitution, reallocation, sorting and the like, unnecessary resource occupation is significantly reduced, system resource utilization efficiency is improved, the problem of low processing efficiency of the same-identification advanced transmission request is solved, the same-identification advanced transmission request can be efficiently dynamically scheduled and managed on the premise of maintaining the correctness of the request response sequence, and the processing efficiency of the same-identification advanced transmission request is improved.

[0079] Optionally, the target scene type comprises a target work type and a target channel type; the target work type comprises a non-blocking mode or a blocking mode; and the target channel type comprises a read request type or a write request type.

[0080] Optionally, the identification request processing device 310 comprises a parameter configuration unit, a mode switching unit, a channel adaptation unit, an exception processing unit and a basic processing unit set; and the basic processing unit set comprises a request identification filtering unit, a request identification substitution unit, a request arbitration unit, a response filtering unit, a response reordering unit and a response arbitration unit.

[0081] Optionally, the mode switching unit is specifically used for performing mode matching on the current identification transmission demand based on a preset mode division rule, determining a target work type corresponding to the target master device, and sending the target work type to the data decision module.

[0082] The channel adaptation unit is specifically used for performing channel division on the current identification transmission demand based on a preset channel division rule, determining a target channel type corresponding to the target master device, and sending the target channel type to the data decision module.

[0083] Optionally, the parameter configuration unit is specifically configured to: before the target master device corresponding basic identification transmission request and current identification transmission demand are acquired by the identification request processing device, construct an advanced transmission identification rule based on a preset identification set, and synchronously process the advanced transmission identification rule, and transmit the advanced transmission identification rule to the basic processing unit set.

[0084] Optionally, the data decision module 320 is specifically configured to: if the target work type is a non-blocking mode and the target channel type is a read request type, the first request processing strategy is taken as the target scheduling strategy; if the target work type is a non-blocking mode and the target channel type is a write request type, the second request processing strategy is taken as the target scheduling strategy; if the target work type is a blocking mode and the target channel type is a read request type, the third request processing strategy is taken as the target scheduling strategy; and if the target work type is a blocking mode and the target channel type is a write request type, the fourth request processing strategy is taken as the target scheduling strategy.

[0085] Optionally, the request identification filtering unit is specifically configured to: perform identification matching on the basic identification transmission request based on the advanced transmission identification rule, determine an identification matching result corresponding to the basic identification transmission request, and determine a target processing unit set corresponding to the basic identification transmission request based on the identification matching result.

[0086] The target processing unit set is specifically configured to: perform enhanced processing on the basic identification transmission request based on the target scheduling strategy, obtain a target identification transmission request corresponding to the basic identification transmission request, and execute the target identification transmission request to obtain a target response result corresponding to the target master device.

[0087] Optionally, the request identification filtering unit is specifically configured to: if the identification matching result is consistent with matching, the request identification replacement unit, the request arbitration unit, the response filtering unit, the response reordering unit and the response arbitration unit are taken as the target processing unit set corresponding to the basic identification transmission request; and if the identification matching result is inconsistent with matching, the request arbitration unit, the response filtering unit and the response arbitration unit are taken as the target processing unit set corresponding to the basic identification transmission request.

[0088] Optionally, the exception processing unit is specifically configured to: after the target scheduling strategy is generated by the target scheduling strategy generation unit, the target scheduling strategy is sent to the base processing unit set, and the base processing unit set performs enhanced processing on the base identification transmission request based on the target scheduling strategy to obtain a target identification transmission request corresponding to the base identification transmission request; and the target identification transmission request is executed to obtain a target response result corresponding to the target master device.

[0089] The scheduling strategy generation system based on the same-identification request provided in the embodiments of the present application can execute the scheduling strategy generation method based on the same-identification request provided in any of the embodiments of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0090] It should be understood that the various forms of flow shown above can be used to reorder, add or delete steps. For example, each step described in the present application can be executed in parallel, sequentially or in a different order, as long as the desired results of the technical solutions of the present application can be achieved, which is not limited herein.

[0091] The above specific embodiments do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for generating scheduling policies based on requests with the same identifier, characterized in that, include: The basic identifier transmission request and current identifier transmission requirement corresponding to the target master device are obtained through the identifier request processing device. The current identifier transmission requirement is matched according to the preset scenario division rules to determine the target scenario type corresponding to the target master device, and the target scenario type is sent to the data decision module. The data decision module determines the target scheduling strategy and basic processing unit set corresponding to the basic identifier transmission request based on the target scenario type, and sends the target scheduling strategy to the basic processing unit set. The basic processing unit set enhances the basic identifier transmission request based on the target scheduling strategy to obtain the target identifier transmission request corresponding to the basic identifier transmission request, and executes the target identifier transmission request to obtain the target response result corresponding to the target master device.

2. The method according to claim 1, characterized in that, The target scenario types include: target work type and target channel type; the target work type includes: non-blocking mode or blocking mode; the target channel type includes: read request type or write request type.

3. The method according to claim 2, characterized in that, The identifier request processing device includes: a parameter configuration unit, a mode switching unit, a channel adaptation unit, an exception handling unit, and a set of basic processing units; the set of basic processing units includes: a request identifier filtering unit, a request identifier replacement unit, a request arbitration unit, a response filtering unit, a response reordering unit, and a response arbitration unit.

4. The method according to claim 3, characterized in that, The identifier request processing device performs scenario matching on the current identifier transmission request based on preset scenario segmentation rules, determines the target scenario type corresponding to the target master device, and sends the target scenario type to the data decision module, including: The mode switching unit performs mode matching on the current identifier transmission requirement based on preset mode division rules, determines the target working type corresponding to the target master device, and sends the target working type to the data decision module; The channel adaptation unit performs channel division on the current identifier transmission requirement based on preset channel division rules, determines the target channel type corresponding to the target master device, and sends the target channel type to the data decision module.

5. The method according to claim 3, characterized in that, Before obtaining the basic identifier transmission request and current identifier transmission requirement corresponding to the target master device through the identifier request processing device, the method further includes: The parameter configuration unit constructs advance transmission identification rules based on a preset identification set, processes the advance transmission identification rules synchronously, and transmits the advance transmission identification rules to the basic processing unit set.

6. The method according to claim 2, characterized in that, The step of determining the target scheduling strategy corresponding to the basic identifier transmission request based on the target scenario type through the data decision module includes: If the target working type is non-blocking mode and the target channel type is read request type, then the first request processing strategy is used as the target scheduling strategy through the data decision module. If the target working type is non-blocking mode and the target channel type is write request type, then the second request processing strategy is used as the target scheduling strategy through the data decision module. If the target working type is blocking mode and the target channel type is read request type, then the third request processing strategy is used as the target scheduling strategy through the data decision module. If the target working type is blocking mode and the target channel type is write request type, then the fourth request processing strategy is used as the target scheduling strategy through the data decision module.

7. The method according to claim 5, characterized in that, The step of enhancing the basic identifier transmission request based on the target scheduling strategy using a set of basic processing units to obtain the target identifier transmission request corresponding to the basic identifier transmission request, and executing the target identifier transmission request to obtain the target response result corresponding to the target master device includes: The request identifier filtering unit performs identifier matching on the basic identifier transmission request based on the advance transmission identifier rule, determines the identifier matching result corresponding to the basic identifier transmission request, and determines the target processing unit set corresponding to the basic identifier transmission request based on the identifier matching result. The target processing unit set enhances the basic identifier transmission request based on the target scheduling strategy to obtain the target identifier transmission request corresponding to the basic identifier transmission request, and executes the target identifier transmission request to obtain the target response result corresponding to the target master device.

8. The method according to claim 7, characterized in that, The request identifier filtering unit determines the set of target processing units corresponding to the basic identifier transmission request based on the identifier matching result, including: If the identifier matching result is a match, then the request identifier filtering unit, the request identifier substitution unit, the request arbitration unit, the response filtering unit, the response reordering unit, and the response arbitration unit are used as the target processing unit set corresponding to the basic identifier transmission request. If the identifier matching result is inconsistent, the request arbitration unit, response filtering unit, and response arbitration unit are used as the target processing unit set corresponding to the basic identifier transmission request through the request identifier filtering unit.

9. The method according to claim 3, characterized in that, After the basic processing unit set enhances the basic identifier transmission request based on the target scheduling strategy to obtain the target identifier transmission request corresponding to the basic identifier transmission request, and executes the target identifier transmission request to obtain the target response result corresponding to the target master device, the method further includes: The exception handling unit obtains the request sending time corresponding to the basic identifier transmission request and the response receiving time corresponding to the target response result, and performs a status judgment on the request sending time and response receiving time based on a preset time threshold to obtain the status evaluation result corresponding to the target response result.

10. A scheduling policy generation system based on same identifier requests, characterized in that, include: The identifier request processing device is used to obtain the basic identifier transmission request and current identifier transmission requirement corresponding to the target master device, perform scenario matching on the current identifier transmission requirement based on the preset scenario division rules, determine the target scenario type corresponding to the target master device, and send the target scenario type to the data decision module; The data decision module is used to determine the target scheduling strategy and the set of basic processing units corresponding to the basic identifier transmission request based on the target scenario type, and send the target scheduling strategy to the set of basic processing units. The basic processing unit set is used to enhance the basic identifier transmission request based on the target scheduling strategy, obtain the target identifier transmission request corresponding to the basic identifier transmission request, execute the target identifier transmission request, and obtain the target response result corresponding to the target master device.

Images