DPU-based delay monitoring method and apparatus, and electronic device
By adopting DPU-based delay monitoring method in the trading platform, recording end-to-end delay and other related delays of transaction data, the problem of low delay monitoring accuracy in the existing technology is solved, and higher delay monitoring accuracy and data decoding real-time performance are achieved.
Patent Information
- Application Number
- CN202411583567.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-07
- Publication Date
- 2025-04-11
AI Technical Summary
The existing trading market data delay monitoring method has low accuracy and cannot effectively count the network delay between the exchange and the analysis equipment.
Using a DPU-based delay monitoring method, the original market data carrying the transmission time is sent to the DPU data processing platform through the first DPU network card, and the reception time is recorded to determine the end-to-end delay. At the same time, the second DPU network card decodes the original market data, records the decoding delay, and determines the software processing delay based on the sending and receiving time.
It improves the accuracy of delay monitoring, avoids the problem of end-to-end delay cannot be recorded, reduces the analysis time of original market data, ensures real-time data decoding, and improves the data management capabilities of preset databases.
Smart Images
Figure CN119806971A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of data processing technology, and in particular to a delay monitoring method, device and electronic device based on DPU. Background Art
[0002] In the scenario of high frequency trading (HFT) of futures, by analyzing the trading market data generated by the exchange, we can understand the corresponding trading conditions of various futures products. In order to be able to grasp the real-time trading conditions in a timely manner, relevant analysis equipment is required to complete data collection and data analysis in a very short time.
[0003] At present, the delay monitoring method of trading market data is usually to obtain trading market data directly from the exchange through software, and record the relevant delay in the process of parsing and analyzing the trading market data. It can be seen that this method can only count the relevant delay when the software processes the trading market data. If the network environment between the exchange and the relevant analysis equipment deteriorates, it will seriously affect the real-time acquisition of the transaction situation, but it will not be counted by the current delay monitoring method of trading market data. Therefore, the current delay monitoring method of trading market data has low accuracy. Summary of the invention
[0004] In order to solve the above technical problems, the present application provides a DPU-based delay monitoring method, device and electronic device, which can improve the accuracy of delay monitoring.
[0005] In the first aspect, the present application provides a DPU-based delay monitoring method, which is applied to a first DPU network card, including: when obtaining original market data, sending the original market data carrying a first sending time to a DPU data processing platform; the first sending time is the sending time of the original market data; the DPU data processing platform is used to generate a transaction order based on the original market data; when receiving the transaction order, recording the first receiving time; determining the end-to-end delay based on the first sending time and the first receiving time.
[0006] In the second aspect, the present application provides a DPU-based delay monitoring device, which is applied to a first DPU network card, including: a communication module, which is used to send the original market data carrying a first sending time to the DPU data processing platform when the original market data is acquired; the first sending time is the sending time of the original market data; the DPU data processing platform is used to generate a transaction order based on the original market data; a recording module, which is used to record the first receiving time when the transaction order is received; a determination module, which is used to determine the end-to-end delay based on the first sending time and the first receiving time.
[0007] In the third aspect, the present application provides a DPU-based delay monitoring method, which is applied to a DPU data processing platform, wherein the DPU data processing platform includes a second DPU network card and a CPU, and the method includes: receiving and decoding original market data through the second DPU network card to obtain decoded data, and recording the decoding delay; when transmitting the decoded data to the target application, recording the second sending time; the target application runs on the CPU of the DPU data processing platform; performing strategy calculation processing on the decoded data according to a preset strategy through the target application, generating a transaction order, and recording the strategy order delay; transmitting the transaction order to the second DPU network card; when receiving the transaction order, recording the second receiving time through the second DPU network card, and determining the software processing delay based on the second sending time and the second receiving time; sending the transaction order to the first DPU network card, and recording the order transmission delay; the first DPU network card is used to collect original market data.
[0008] In a fourth aspect, the present application provides a DPU-based delay monitoring device, which is applied to a DPU data processing platform, and the DPU data processing platform includes a second DPU network card and a CPU; the device includes: a network card processing module, which is used to: receive and decode original market data through the second DPU network card, obtain decoded data, and record the decoding delay; when transmitting the decoded data to the target application, record the second sending time; the target application runs on the CPU of the DPU data processing platform; a central processing module, which is used to: perform strategy calculation processing on the decoded data according to a preset strategy through the target application, generate a transaction order, and record the strategy order delay; transmit the transaction order to the second DPU network card; the network card processing module is also used to: when receiving the transaction order, record the second receiving time through the second DPU network card, and determine the software processing delay based on the second sending time and the second receiving time; send the transaction order to the first DPU network card, and record the order transmission delay; the first DPU network card is used to collect original market data.
[0009] In the fifth aspect, the present application provides a DPU-based delay monitoring method, which is applied to a preset database, including: obtaining end-to-end delay, decoding delay, strategy order delay, software processing delay, and order transmission delay; the end-to-end delay is the delay from the first DPU network card sending the original market data to the first DPU network card receiving the transaction order corresponding to the original market data; the decoding delay is the delay corresponding to the second DPU network card decoding the original market data; the strategy order delay is the delay corresponding to the target application generating the transaction order of the original market data; the software processing delay is the delay from the second DPU network card sending the decoded data of the original market data to the second DPU network card receiving the transaction order corresponding to the original market data; the order transmission delay is the delay from the second DPU network card sending the decoded data of the original market data to the first DPU network card receiving the transaction order corresponding to the original market data; the order transmission delay is the delay from the second DPU network card sending the transaction order corresponding to the original market data to the first DPU network card; the end-to-end delay, decoding delay, strategy order delay, software processing delay, and order transmission delay are stored according to the inclusion relationship.
[0010] In a sixth aspect, the present application provides a DPU-based delay monitoring device, which is applied to a preset database, including: an acquisition module, used to obtain end-to-end delay, decoding delay, strategy order delay, software processing delay, and order transmission delay; the end-to-end delay is the delay from the first DPU network card sending the original market data to the first DPU network card receiving the transaction order corresponding to the original market data; the decoding delay is the delay corresponding to the second DPU network card decoding the original market data; the strategy order delay is the delay corresponding to the transaction order generating the original market data by the target application; the software processing delay is the delay from the second DPU network card sending the decoded data of the original market data to the second DPU network card receiving the transaction order corresponding to the original market data; the order transmission delay is the delay from the second DPU network card sending the decoded data of the original market data to the first DPU network card receiving the transaction order corresponding to the original market data; the order transmission delay is the delay from the second DPU network card sending the transaction order corresponding to the original market data to the first DPU network card; a storage module is used to store the end-to-end delay, decoding delay, strategy order delay, software processing delay, and order transmission delay according to the inclusion relationship.
[0011] In the seventh aspect, the present application provides an electronic device, comprising: a DPU network card, a processor, a memory, a programmable logic loaded on the DPU network card, and a computer program stored in the memory and runnable on the processor, wherein when the programmable logic is executed by the DPU network card, the DPU-based delay monitoring method of the first aspect is implemented; or, when the programmable logic is executed by the DPU network card and the computer program is executed by the processor, the DPU-based delay monitoring method of the third aspect is implemented; or, when the computer program is executed by the processor, the DPU-based delay monitoring method of the fifth aspect is implemented.
[0012] In an eighth aspect, the present application provides a computer-readable storage medium, comprising: a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the DPU-based delay monitoring method as in the fifth aspect is implemented.
[0013] In a ninth aspect, the present application provides a computer program product, comprising: when the computer program product runs on a computer, the computer implements the DPU-based delay monitoring method as in the fifth aspect.
[0014] The technical solution provided by the present application has the following advantages compared with the prior art: First, when the first DPU network card obtains the original market data, it sends the original market data carrying the first sending time to the second DPU network card of the DPU data processing platform. Secondly, the second DPU network card receives and decodes the original market data, obtains the decoded data, and records the decoding delay; when transmitting the decoded data to the target application running on the CPU, the second sending time is recorded. Afterwards, the target application performs strategy calculation processing on the decoded data according to the preset strategy, generates a transaction order, and records the strategy order delay; and transmits the transaction order to the second DPU network card. Afterwards, when the second DPU network card receives the transaction order, it records the second receiving time, and determines the software processing delay based on the second sending time and the second receiving time; sends the transaction order to the first DPU network card, and records the order transmission delay. Finally, when the first DPU network card receives the transaction order, it records the first receiving time, and determines the end-to-end delay based on the first sending time and the first receiving time. In addition, the preset database stores end-to-end delay, decoding delay, strategy order delay, software processing delay, and order transmission delay according to the inclusion relationship. In this way, on the one hand, the end-to-end delay from sending the original market data to receiving the transaction order corresponding to the original market data can be recorded through the first DPU network card, avoiding the problem of end-to-end delay being unable to be recorded and improving the accuracy of delay monitoring. On the other hand, the original market data can be decoded through the second DPU network card, avoiding the problem of using software to parse the original market data resulting in a long parsing time, reducing the parsing time of the original market data, and ensuring the real-time decoding of the original market data. On the third hand, each delay can also be stored according to the inclusion relationship, making the query access of the preset database more concise and improving the data management capabilities of the preset database. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0017] Figure 1 One of the application scenario schematic diagrams of the DPU-based delay monitoring method provided in the embodiment of the present application;
[0018] Figure 2 The second schematic diagram of the application scenario of the delay monitoring method based on DPU provided in the embodiment of the present application;
[0019] Figure 3 One of the flow charts of the delay monitoring method based on DPU provided in the embodiment of the present application;
[0020] Figure 4 A third schematic diagram of an application scenario of a delay monitoring method based on a DPU provided in an embodiment of the present application;
[0021] Figure 5 The second flowchart of the delay monitoring method based on DPU provided in the embodiment of the present application;
[0022] Figure 6 The third flowchart of the delay monitoring method based on DPU provided in the embodiment of the present application;
[0023] Figure 7 A fourth flowchart of a delay monitoring method based on a DPU provided in an embodiment of the present application;
[0024] Figure 8 One of the schematic diagrams of the inclusion relationship provided in the embodiment of the present application;
[0025] Fig. 9 The second schematic diagram of the inclusion relationship provided for the embodiment of the present application;
[0026] Fig.10 One of the structural schematic diagrams of the delay monitoring device based on DPU provided in the embodiment of the present application;
[0027] Fig.11 The second structural diagram of the delay monitoring device based on DPU provided in the embodiment of the present application;
[0028] Fig.12 The third structural diagram of the delay monitoring device based on DPU provided in the embodiment of the present application;
[0029] Fig.13 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0030] In order to more clearly understand the above-mentioned purposes, features and advantages of the present application, the scheme of the present application will be further described below. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.
[0031] In the following description, many specific details are set forth to facilitate a full understanding of the present application, but the present application may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only part of the embodiments of the present application, rather than all of the embodiments.
[0032] The current delayed monitoring methods for transaction market data are as follows: Figure 1 As shown, the software in the software analysis system 200 usually obtains the trading market data from the trading platform 100 through the Internet communication with the trading platform 100 of the exchange, and then the software analysis system 200 records the relevant delay in the process of parsing and analyzing the trading market data. It can be seen that this method can only count the relevant delay when the software analysis system 200 processes the trading market data. If the network environment between the trading platform 100 and the software analysis system 200 deteriorates, it will seriously affect the real-time nature of obtaining the transaction situation, but it will not be counted by the software analysis system 200. Therefore, the current delay monitoring method of trading market data has low accuracy.
[0033] In response to the above problems, Figure 2 As shown, the present application provides a DPU-based delay monitoring system, including a trading platform 300, a DPU data processing platform 400, and a preset database 500. Among them, the trading platform 300 includes at least a first DPU network card; the DPU data processing platform 400 includes at least a second DPU network card and a CPU, and the second DPU network card and the CPU can be connected through a bus.
[0034] In some embodiments, the first DPU network card in the trading platform 300 and the second DPU network card in the DPU data processing platform 400 can communicate via optical fiber.
[0035] In some embodiments, the first DPU network card and the DPU data processing platform 400 can asynchronously store data in a preset database 500 .
[0036] It should be noted that the DPU-based delay monitoring method provided in the embodiment of the present application can be executed through the above-mentioned first DPU network card, or through the above-mentioned DPU data processing platform 400, or through the above-mentioned preset database 500, or can be jointly executed by the first DPU network card, the DPU data processing platform 400, and the preset database 500, and the present application does not limit this.
[0037] Exemplarily, when the DPU-based delay monitoring method provided in the embodiment of the present application is jointly executed by the first DPU network card, the DPU data processing platform 400, and the preset database 500, the first DPU network card may send the original market data carrying the first sending time to the second DPU network card of the DPU data processing platform 400 when acquiring the original market data. The second DPU network card receives and decodes the original market data, obtains the decoded data, and records the decoding delay; when transmitting the decoded data to the target application running on the CPU, the second sending time is recorded. Afterwards, the target application performs policy calculation processing on the decoded data according to the preset strategy, generates a transaction order, and records the policy order delay; and transmits the transaction order to the second DPU network card. When the second DPU network card receives the transaction order, it records the second receiving time, and determines the software processing delay according to the second sending time and the second receiving time; sends the transaction order to the first DPU network card, and records the order transmission delay. Finally, when the first DPU network card receives the transaction order, it records the first receiving time, and determines the end-to-end delay according to the first sending time and the first receiving time. In addition, the preset database 500 stores the end-to-end delay, decoding delay, strategy order delay, software processing delay, and order transmission delay according to the inclusion relationship. In this way, the problem of the end-to-end delay being unable to be recorded is avoided, the accuracy of delay monitoring is improved, the analysis time of the original market data is reduced, the real-time decoding of the original market data is guaranteed, and the data management capability of the preset database is improved.
[0038] Figure 3 A flow chart of a delay monitoring method based on a DPU provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, the delay monitoring method based on DPU may include the following steps:
[0039] S11. The first DPU network card determines that original market data has been obtained.
[0040] Specifically, the first DPU network card is located in the trading platform and can monitor whether a transaction occurs in the trading center of the trading platform. When a transaction occurs, delayed monitoring is started and the original market data corresponding to the transaction is obtained; when no transaction occurs in the trading center, delayed monitoring is not started.
[0041] S12. The first DPU network card sends original market data carrying a first sending time to the DPU data processing platform.
[0042] Among them, the first sending time is the sending time of the original market data; the DPU data processing platform is used to generate a transaction order based on the original market data.
[0043] Specifically, when the first DPU network card obtains the original market data, it intercepts the hardware timestamp of the current moment as the first sending moment, and attaches the first sending moment to the original market data and sends it to the DPU data processing platform. The first DPU network card sends the original market data to the DPU data processing platform in a manner that the first DPU network card sends the original market data to the DPU data processing platform via optical fiber communication.
[0044] S13. The second DPU network card receives and decodes the original market data, obtains decoded data and records the decoding delay.
[0045] First, the second DPU network card receives and decodes the original market data to obtain decoded data.
[0046] Specifically, a hardware decoding acceleration module can be generated by a field programmable gate array (FPGA) and loaded onto the second DPU network card to decode the original market data to obtain decoded data, thereby improving the decoding speed of the original market data.
[0047] Secondly, the second DPU network card records the decoding delay.
[0048] Specifically, the second DPU network card records the decoding delay by recording the decoding start time and the decoding end time of the original market data, and calculating the decoding delay of the original market data by the decoding start time and the decoding end time. The duration between the decoding start time and the decoding end time can be determined as the decoding delay of the original market data.
[0049] S14. When the second DPU network card transmits the decoded data to the target application, the second sending time is recorded.
[0050] The target application runs on the CPU of the DPU data processing platform.
[0051] In some embodiments, Figure 4 As shown, the DPU data processing platform also includes a driver module for data transmission between the second DPU network card and the CPU in the DPU data processing platform. That is, the second DPU network card can transmit decoded data to the target application through the driver module.
[0052] In some embodiments, the second DPU network card carries the second sending time when transmitting decoded data to the target application.
[0053] S15. The target application performs strategy calculation processing on the decoded data according to the preset strategy, generates a transaction order, and records the strategy order delay.
[0054] First, the target application performs strategy calculation processing on the decoded data according to the preset strategy to generate a transaction order. The preset strategy is pre-written and used to guide the generation of the transaction order.
[0055] For example, Figure 4 As shown, the CPU of the DPU data processing platform may include a software development kit (SDK) module and a policy module. Users can configure preset policies in the policy module. When the target application receives decoded data, it can read the decoded data through the SDK module, and perform policy calculation processing on the decoded data according to the preset policies in the policy module to generate a transaction order corresponding to the decoded data.
[0056] Secondly, the target application records the strategy order delay.
[0057] For example, Figure 4 As shown, the target application records the strategy order delay by recording the calculation start time when reading the decoded data through the SDK module; and recording the calculation end time when generating the transaction order corresponding to the decoded data; finally, the strategy order delay corresponding to the decoded data is calculated according to the calculation start time and calculation end time. Among them, the time length between the calculation start time and the calculation end time can be determined as the strategy order delay corresponding to the decoded data.
[0058] S16. The target application transmits the transaction order to the second DPU network card.
[0059] Specifically, the target application may transmit the transaction order to the second DPU network card through the driver module.
[0060] S17. When receiving the transaction order, the second DPU network card records the second receiving time, and determines the software processing delay according to the second sending time and the second receiving time.
[0061] Specifically, the second DPU network card determines the software processing delay according to the second sending time and the second receiving time by determining the duration between the second sending time and the second receiving time as the software processing delay corresponding to when the CPU processes decoded data.
[0062] S18. The second DPU network card sends a transaction order to the first DPU network card and records the order transmission delay.
[0063] Among them, the first DPU network card is used to collect original market data.
[0064] First, the second DPU network card sends a transaction order to the first DPU network card. Specifically, the second DPU network card sends the transaction order to the first DPU network card by sending original market data to the first DPU network card through optical fiber communication.
[0065] Secondly, the second DPU network card records the order transmission delay. Specifically, the second DPU network card records the order transmission delay in the following manner: when the second DPU network card starts to send the transaction order to the first DPU network card, it records the time when the order starts to be sent; when the transaction order is sent, it records the time when the order ends to be sent; finally, the order transmission delay of the transaction order is calculated according to the time when the order starts to be sent and the time when the order ends to be sent. Among them, the duration between the time when the order starts to be sent and the time when the order ends to be sent can be determined as the order transmission delay of the transaction order.
[0066] S19. When receiving the transaction order, the first DPU network card records the first receiving time.
[0067] S20. The first DPU network card determines the end-to-end delay according to the first sending time and the first receiving time.
[0068] Specifically, the first DPU network card can determine the duration between the first sending moment and the first receiving moment as the high-frequency trading end-to-end delay (Trade Round Trip Time, trtt), that is, the end-to-end delay in this application.
[0069] In the above scheme, first, when the first DPU network card obtains the original market data, it sends the original market data carrying the first sending time to the second DPU network card of the DPU data processing platform. Secondly, the second DPU network card receives and decodes the original market data, obtains the decoded data, and records the decoding delay; when transmitting the decoded data to the target application running on the CPU, the second sending time is recorded. Afterwards, the target application performs strategy calculation processing on the decoded data according to the preset strategy, generates a transaction order, and records the strategy order delay; and transmits the transaction order to the second DPU network card. Afterwards, when the second DPU network card receives the transaction order, it records the second receiving time, and determines the software processing delay based on the second sending time and the second receiving time; sends the transaction order to the first DPU network card, and records the order transmission delay. Finally, when the first DPU network card receives the transaction order, it records the first receiving time, and determines the end-to-end delay based on the first sending time and the first receiving time. In this way, on the one hand, the end-to-end delay from sending the original market data to receiving the transaction order corresponding to the original market data can be recorded through the first DPU network card, avoiding the problem of the inability to record the end-to-end delay and improving the accuracy of delay monitoring. On the other hand, the original market data can be decoded through the second DPU network card, avoiding the problem of using software to parse the original market data resulting in a long parsing time, reducing the parsing time of the original market data, and ensuring the real-time decoding of the original market data. In addition, the first DPU network card directly encapsulates the raw package of the original market data into a frame at the data link layer and sends it to the second DPU network card for processing. The traditional method of using the network layer to capture packets is no longer used, avoiding the need to unify the format of the original market data and data packaging according to the network protocol when transmitting through the network layer, greatly improving the transmission speed.
[0070] In some embodiments, the original market data includes a second market identifier, and the transaction order includes at least one transaction order. Figure 5 As shown, the second DPU network card determines the software processing delay according to the second sending time and the second receiving time, which may include the following steps:
[0071] S171. Determine the order identifier of each transaction order from at least one received transaction order.
[0072] In actual applications, multiple transactions of different types may occur on the trading platform in the same time period. Therefore, there is a high probability that the second DPU network card will receive a large number of trading orders in a relatively short period of time. Therefore, the market identification is usually carried in the original market data, and it is guaranteed not to be lost during the decoding and strategy settlement stages, so as to identify the original market data corresponding to the decoded data and trading orders.
[0073] In some embodiments, the market information identifier may be an identifier that can uniquely identify the corresponding transaction, such as a transaction number, a transaction name, etc.
[0074] In some embodiments, the market information may be more than one. For example, the market information may be a combination of a transaction contract number and a transaction time.
[0075] S172. Determine the transaction order containing the order identifier corresponding to the second market identifier as the second target order.
[0076] The second target order is a trading order corresponding to the original market data; and the order identifier is a market identifier in the original market data corresponding to the trading order.
[0077] Specifically, the transaction order with the order identifier identical to the second market identifier in the original market data is determined as the transaction order corresponding to the original market data.
[0078] S173. Determine the software processing delay according to the second sending time and the second receiving time corresponding to the second target order.
[0079] Specifically, the duration between the second sending moment and the second receiving moment corresponding to the second target order may be determined as the software processing delay.
[0080] In the above scheme, when there are multiple trading orders, the trading order corresponding to the original market data can be identified based on the market identifier in the original market data and the order identifier in the trading order, thereby avoiding the problem of inaccurate software processing delay caused by the inability to identify the trading order corresponding to the original market data.
[0081] In some embodiments, the original market data includes a first market identifier, and the transaction order includes at least one transaction order. Figure 6 As shown, the manner in which the first DPU network card determines the end-to-end delay according to the first sending time and the first receiving time may include the following steps:
[0082] S201. Determine an order identifier of each transaction order from at least one received transaction order.
[0083] Specifically, the method of determining the order identification of each transaction order from at least one received transaction order is the same as the method of determining the order identification of each transaction order from at least one received transaction order in step S171, and is not repeated here.
[0084] S202: Determine the transaction order containing the order identifier corresponding to the first market identifier as the first target order.
[0085] Among them, the first target order is the transaction order corresponding to the original market data.
[0086] S203: Determine the end-to-end delay according to the first sending time and the first receiving time corresponding to the first target order.
[0087] Specifically, the duration between the first sending moment and the first receiving moment corresponding to the first target order may be determined as the end-to-end delay.
[0088] In the above scheme, when there are multiple transaction orders, the transaction order corresponding to the original market data can be identified based on the market identifier in the original market data and the order identifier in the transaction order, thereby avoiding the problem of inaccurate end-to-end delay determined due to the inability to identify the transaction order corresponding to the original market data.
[0089] In some embodiments, Figure 7 As shown, the delay monitoring method based on DPU may also include the following steps:
[0090] S31. After recording the order transmission delay, the DPU data processing platform asynchronously stores the decoding delay, strategy order delay, software processing delay, and order transmission delay in the preset database.
[0091] In some embodiments, Figure 4 As shown, the CPU of the DPU data processing platform may include a public tool module, which can establish a communication connection with a preset database and asynchronously store the decoding delay, strategy order delay, software processing delay, and order transmission delay to the preset database through a pre-established communication connection.
[0092] S32. After determining the end-to-end delay according to the first sending time and the first receiving time, the first DPU network card asynchronously stores the end-to-end delay in a preset database.
[0093] In some embodiments, an asynchronous storage module can be generated by FPGA and loaded onto the first DPU network card. The asynchronous storage module can establish a communication connection with a preset database and asynchronously store the end-to-end delay in the preset database.
[0094] S33. After obtaining the end-to-end delay, decoding delay, strategy order delay, software processing delay, and order transmission delay, the preset database stores the end-to-end delay, decoding delay, strategy order delay, software processing delay, and order transmission delay according to the inclusion relationship.
[0095] In some embodiments, the inclusion relationship may include only the end-to-end delay, decoding delay, policy order delay, software processing delay, and order transmission delay. Figure 8As shown, the end-to-end delay includes decoding delay, strategy order delay, software processing delay, and order transmission delay; the software processing delay includes the strategy order delay.
[0096] In some embodiments, the inclusion relationship may also include other user-defined delays and end-to-end delays, decoding delays, policy order delays, software processing delays, and order transmission delays. Fig. 9 As shown, the software processing delay includes not only the strategy order delay, but also the CPU transmission delay of decoded data and transaction orders between the second DPU network card and the end-to-end delay includes not only the decoding delay, strategy order delay, software processing delay, and order transmission delay, but also the data transmission delay of original market data between the first DPU network card and the second DPU network card, and the CPU transmission delay of decoded data and transaction orders between the second DPU network card and the end-to-end delay.
[0097] In the above scheme, first, after recording the order transmission delay, the second DPU network card asynchronously stores the decoding delay, strategy order delay, software processing delay, and order transmission delay in the preset database. Secondly, after the first DPU network card determines the end-to-end delay according to the first sending time and the first receiving time, it asynchronously stores the end-to-end delay in the preset database. Finally, the preset database stores the end-to-end delay, decoding delay, strategy order delay, software processing delay, and order transmission delay according to the inclusion relationship. In this way, the preset database can store each delay according to the inclusion relationship, making the query access of the preset database more concise and improving the data management capability of the preset database.
[0098] In some embodiments, the end-to-end delay, decoding delay, strategy order delay, software processing delay, and order transmission delay stored in the preset database can be used to analyze the performance of the trading platform and the DPU data processing platform. For example, the average value, standard deviation, and quantile of the delay at each stage are analyzed, and the weight of the delay at each stage on the trtt is analyzed, so as to determine the performance bottleneck of the entire trtt.
[0099] The embodiment of the present application can divide the functional modules of the DPU-based delay monitoring device according to the above method example. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules. For example, in the first DPU network card and the second DPU network card, the integrated module can be implemented in the form of hardware. For example, the corresponding module can be generated by FPGA and loaded into the corresponding DPU network card; in the CPU of the DPU data processing platform, the integrated module can be implemented in the form of software functional modules. It should be noted that the division of modules in the embodiment of the present application is schematic and is only a logical functional division. There may be other division methods in actual implementation.
[0100] like Fig.10 As shown, a DPU-based delay monitoring device provided in an embodiment of the present application is applied to a first DPU network card. The DPU-based delay monitoring device includes a communication module 71, a recording module 72 and a determination module 73.
[0101] The communication module 71 is used to send the original market data carrying the first sending time to the DPU data processing platform when the original market data is obtained; the first sending time is the sending time of the original market data; the DPU data processing platform is used to generate a transaction order based on the original market data; the recording module 72 is used to record the first receiving time when the transaction order is received; the determination module 73 is used to determine the end-to-end delay based on the first sending time and the first receiving time.
[0102] In some embodiments, the original market data includes a first market identifier; the transaction order includes at least one transaction order; a determination module 73 is specifically used to determine the order identifier of each transaction order from at least one received transaction order; the transaction order containing the order identifier corresponding to the first market identifier is determined as the first target order; the first target order is the transaction order corresponding to the original market data; and the end-to-end delay is determined based on the first sending time and the first receiving time corresponding to the first target order.
[0103] In some embodiments, the determination module 73 is further configured to asynchronously store the end-to-end delay in a preset database after determining the end-to-end delay according to the first sending time and the first receiving time.
[0104] The DPU-based delay monitoring device provided in this embodiment can execute the DPU-based delay monitoring method provided in the above method embodiment. Its implementation principle and technical effects are similar to those of the above method and will not be repeated here.
[0105] like Fig.11As shown, a DPU-based delay monitoring device provided in an embodiment of the present application is applied to a DPU data processing platform. The DPU data processing platform includes a second DPU network card and a CPU. The DPU-based delay monitoring device includes a network card processing module 81 and a central processing module 82.
[0106] The network card processing module 81 is used to: receive and decode the original market data through the second DPU network card, obtain decoded data, and record the decoding delay; when transmitting the decoded data to the target application, record the second sending time; the target application runs on the CPU of the DPU data processing platform; the central processing module 82 is used to: perform strategy calculation processing on the decoded data according to the preset strategy through the target application, generate a transaction order, and record the strategy order delay; transmit the transaction order to the second DPU network card; the network card processing module 81 is also used to: when receiving the transaction order, record the second receiving time through the second DPU network card, and determine the software processing delay according to the second sending time and the second receiving time; send the transaction order to the first DPU network card, and record the order transmission delay; the first DPU network card is used to collect original market data.
[0107] In some embodiments, the network card processing module 81 is specifically used to: determine the order identifier of each transaction order from at least one received transaction order; determine the transaction order containing the order identifier corresponding to the second market identifier as the second target order; the second target order is the transaction order corresponding to the original market data; determine the software processing delay based on the second sending time and the second receiving time corresponding to the second target order.
[0108] In some embodiments, the central processing module 82 is used to asynchronously store the decoding delay, the strategy order delay, the software processing delay, and the order transmission delay in a preset database after recording the order transmission delay.
[0109] The DPU-based delay monitoring device provided in this embodiment can execute the DPU-based delay monitoring method provided in the above method embodiment. Its implementation principle and technical effects are similar to those of the above method and will not be repeated here.
[0110] like Fig.12 As shown, a DPU-based delay monitoring device provided in an embodiment of the present application is applied to a preset database. The DPU-based delay monitoring device includes an acquisition module 91 and a storage module 92.
[0111] The acquisition module 91 is used to obtain the end-to-end delay, decoding delay, strategy order delay, software processing delay, and order transmission delay; the end-to-end delay is the delay from the first DPU network card sending the original market data to the first DPU network card receiving the transaction order corresponding to the original market data; the decoding delay is the delay corresponding to the second DPU network card decoding the original market data; the strategy order delay is the delay corresponding to the target application generating the transaction order of the original market data; the software processing delay is the delay from the second DPU network card sending the decoded data of the original market data to the second DPU network card receiving the transaction order corresponding to the original market data; the order transmission delay is the delay from the second DPU network card sending the transaction order corresponding to the original market data to the first DPU network card; the storage module 92 is used to store the end-to-end delay, decoding delay, strategy order delay, software processing delay, and order transmission delay according to the inclusion relationship.
[0112] The DPU-based delay monitoring device provided in this embodiment can execute the DPU-based delay monitoring method provided in the above method embodiment. Its implementation principle and technical effects are similar to those of the above method and will not be repeated here.
[0113] Fig.13 An electronic device is shown according to an exemplary embodiment. The electronic device may include a DPU network card 905 and a processor 902, wherein the DPU network card 905 is used to execute programmable logic, and the processor 902 is used to execute application code, thereby implementing the DPU-based delay monitoring method in this application.
[0114] The processor 902 may be a central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the program of the present application.
[0115] The DPU network card 905 includes programmable logic, which allows the user to configure it to implement the DPU-based delay monitoring method in the present application.
[0116] like Fig.13 As shown, the electronic device may further include a memory 903. The memory 903 is used to store application program codes for executing the solution of the present application, and the execution is controlled by the processor 902.
[0117] The memory 903 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 903 may exist independently and be connected to the processor 902 via the bus 904. The memory 903 may also be integrated with the processor 902.
[0118] like Fig.13 As shown, the electronic device may further include a communication interface 901, wherein the communication interface 901, the processor 902, and the memory 903 may be coupled to each other, for example, via a bus 904. The communication interface 901 is used to exchange information with other devices, for example, to support information exchange between the electronic device and other devices.
[0119] It should be pointed out that Fig.13 The device structure shown in the figure does not constitute a limitation on the electronic device, except Fig.13 In addition to the components shown, the electronic device may include more or fewer components than shown, or combine certain components, or arrange the components differently. The electronic device provided in this embodiment can execute the DPU-based delay monitoring method provided in the above method embodiment, and its implementation principle and technical effects are similar to the above method, which will not be repeated here.
[0120] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the various processes of the DPU-based delay monitoring method in the above method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.
[0121] The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.
[0122] An embodiment of the present application provides a computer program product, which stores a computer program. When the computer program is executed by a processor, each process of the DPU-based delay monitoring method in the above method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.
[0123] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media that include computer-usable program code.
[0124] In this application, memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.
[0125] In this application, computer-readable media includes permanent and non-permanent, removable and non-removable storage media. Storage media can be implemented by any method or technology to store information, and the information can be computer-readable instructions, data structures, modules of programs or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. According to the definition in this article, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data and carrier waves.
[0126] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0127] The above description is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A delay monitoring method based on DPU, characterized in that: Applied to the first DPU network card, including: When the original market data is acquired, the original market data carrying the first sending time is sent to the DPU data processing platform; the first sending time is the sending time of the original market data; the DPU data processing platform is used to generate a transaction order according to the original market data; When receiving a transaction order, record the first receiving time; An end-to-end delay is determined according to the first sending time and the first receiving time.
2. The delay monitoring method based on DPU according to claim 1 is characterized in that: The original market data includes a first market identifier; the transaction order includes at least one transaction order; and the end-to-end delay is determined according to the first sending time and the first receiving time, including: Determining an order identifier of each transaction order from at least one received transaction order; Determine the transaction order containing the order identifier corresponding to the first market identifier as the first target order; the first target order is the transaction order corresponding to the original market data; The end-to-end delay is determined according to the first sending time and the first receiving time corresponding to the first target order.
3. The delay monitoring method based on DPU according to claim 1 or 2, characterized in that: After determining the end-to-end delay according to the first sending time and the first receiving time, the method further includes: The end-to-end delay is asynchronously stored in a preset database.
4. A delay monitoring method based on DPU, characterized in that: Applied to a DPU data processing platform, the DPU data processing platform includes a second DPU network card and a CPU, and the method includes: Receiving and decoding the original market data through the second DPU network card to obtain decoded data, and recording the decoding delay; when transmitting the decoded data to the target application, recording the second sending time; the target application runs on the CPU of the DPU data processing platform; Performing strategy calculation processing on the decoded data according to a preset strategy through the target application, generating a transaction order, and recording the strategy order delay; transmitting the transaction order to the second DPU network card; When a transaction order is received, the second receiving time is recorded through the second DPU network card, and the software processing delay is determined according to the second sending time and the second receiving time; the transaction order is sent to the first DPU network card, and the order transmission delay is recorded; the first DPU network card is used to collect original market data.
5. The delay monitoring method based on DPU according to claim 4 is characterized in that: The original market data includes a second market identifier; the transaction order includes at least one transaction order; and the software processing delay is determined according to the second sending time and the second receiving time, including: Determining an order identifier of each transaction order from at least one received transaction order; Determine the transaction order containing the order identifier corresponding to the second market identifier as the second target order; the second target order is the transaction order corresponding to the original market data; The software processing delay is determined according to the second sending time and the second receiving time corresponding to the second target order.
6. The delay monitoring method based on DPU according to claim 4 or 5, characterized in that: After recording the order transmission delay, the method further includes: The decoding delay, the strategy order delay, the software processing delay, and the order transmission delay are asynchronously stored in a preset database.
7. A delay monitoring method based on DPU, characterized in that: Applies to preset databases, including: Obtain end-to-end delay, decoding delay, strategy order delay, software processing delay, and order transmission delay; the end-to-end delay is the delay from the first DPU network card sending the original market data to the first DPU network card receiving the transaction order corresponding to the original market data; the decoding delay is the delay corresponding to the second DPU network card decoding the original market data; the strategy order delay is the delay corresponding to the target application generating the transaction order of the original market data; the software processing delay is the delay from the second DPU network card sending the decoded data of the original market data to the second DPU network card receiving the transaction order corresponding to the original market data; the order transmission delay is the delay from the second DPU network card sending the transaction order corresponding to the original market data to the first DPU network card; The end-to-end delay, the decoding delay, the policy order delay, the software processing delay, and the order transmission delay are stored according to an inclusion relationship.
8. A delay monitoring device based on DPU, characterized in that: Applied to the first DPU network card, including: A communication module, configured to send the original market data carrying a first sending time to the DPU data processing platform when the original market data is acquired; the first sending time is the sending time of the original market data; the DPU data processing platform is configured to generate a transaction order according to the original market data; A recording module, used to record the first receiving time when receiving the transaction order; A determination module is used to determine the end-to-end delay according to the first sending time and the first receiving time.
9. A delay monitoring device based on DPU, characterized in that: Applied to a DPU data processing platform, the DPU data processing platform includes a second DPU network card and a CPU; the device includes: The network card processing module is used to: receive and decode the original market data through the second DPU network card, obtain decoded data, and record the decoding delay; when transmitting the decoded data to the target application, record the second sending time; the target application runs on the CPU of the DPU data processing platform; The central processing module is used to: perform strategy calculation processing on the decoded data according to a preset strategy through the target application, generate a transaction order, and record the strategy order delay; transmit the transaction order to the second DPU network card; The network card processing module is also used to: when receiving a transaction order, record the second receiving time through the second DPU network card, and determine the software processing delay according to the second sending time and the second receiving time; send the transaction order to the first DPU network card, and record the order transmission delay; the first DPU network card is used to collect original market data.
10. A delay monitoring device based on DPU, characterized in that: Applies to preset databases, including: An acquisition module is used to acquire end-to-end delay, decoding delay, strategy order delay, software processing delay, and order transmission delay; the end-to-end delay is the delay from the first DPU network card sending the original market data to the first DPU network card receiving the transaction order corresponding to the original market data; the decoding delay is the delay corresponding to the second DPU network card decoding the original market data; the strategy order delay is the delay corresponding to the target application generating the transaction order of the original market data; the software processing delay is the delay from the second DPU network card sending the decoded data of the original market data to the second DPU network card receiving the transaction order corresponding to the original market data; the order transmission delay is the delay from the second DPU network card sending the transaction order corresponding to the original market data to the first DPU network card; A storage module is used to store the end-to-end delay, the decoding delay, the policy order delay, the software processing delay, and the order transmission delay according to an inclusion relationship.
11. An electronic device, characterized in that: include: A DPU network card, a processor, a memory, a programmable logic loaded on the DPU network card, and a computer program stored in the memory and executable on the processor, wherein the programmable logic, when executed by the DPU network card, implements the DPU-based delay monitoring method as described in any one of claims 1 to 3; or, the programmable logic, when executed by the DPU network card and the computer program, when executed by the processor, implements the DPU-based delay monitoring method as described in any one of claims 4 to 6; or, the computer program, when executed by the processor, implements the DPU-based delay monitoring method as described in claim 7.