A quantum computing simulation multi-node monitoring scheduling method and device
Through the remote process call protocol, the quantum computing simulation nodes are monitored and dispatched in real time in quantum computing simulation, and the problems of slow single-thread processing speed and low task scheduling efficiency are solved, and efficient task processing is achieved through the collaborative work of multiple nodes.
Patent Information
- Application Number
- CN202210404453.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-04-18
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2042-04-18
AI Technical Summary
The existing quantum computing simulation tasks are mainly single threaded running on a single machine, and the processing speed is slow, making it difficult to monitor the status of each quantum computing simulation node in real time, resulting in inefficient scheduling of tasks from multiple nodes.
Through the remote process call protocol, the running status of the quantum computing simulation node is monitored in real time, and the task is allocated to idle nodes for calculation according to the calling level of the task, realizing multi-node scheduling and status monitoring of the quantum computing simulation nodes, improving task processing efficiency.
It improves the execution efficiency of quantum experimental tasks, solves the problem of slow processing speed of single-threaded computing nodes, and realizes correct task scheduling and real-time monitoring of multiple nodes, which significantly speeds up the processing speed.
Smart Images

Figure CN114861921B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of quantum computing monitoring and scheduling management, and in particular to a quantum computing simulation multi-node monitoring scheduling method and device. Background Art
[0002] Quantum computing simulation can realize parallel computing for data processing, which greatly improves the computing speed. At the same time, the speed of quantum computing simulation will increase exponentially with the increase of the number of entangled bits that can be manipulated in the experiment. Therefore, in the future, quantum computer simulation can be applied to scientific problems that require large-scale computing. Multi-node parallel computing for quantum computing simulation is also imperative.
[0003] At present, quantum computing simulation tasks are mainly single-threaded running tasks on a single machine. For a large number of experimental tasks, the processing of a single simulation node is relatively heavy and slow. It is also difficult to monitor the status of each quantum computing simulation node in real time, and it is not possible to correctly schedule tasks for multiple nodes, resulting in low execution efficiency of quantum experimental tasks. Summary of the invention
[0004] The embodiments of the present application provide a quantum computing simulation multi-node monitoring and scheduling method and device, which are used to solve the following technical problems: the existing single simulation node has a slow processing speed, cannot realize multi-node task scheduling, and is difficult to monitor the status of quantum computing simulation nodes in real time, resulting in low execution efficiency.
[0005] The present application embodiment adopts the following technical solutions:
[0006] On the one hand, an embodiment of the present application provides a quantum computing simulation multi-node monitoring and scheduling method and device, characterized in that the method includes: storing a number of acquired quantum experiment task information in a task data table; performing call level calculation on a number of the quantum experiment task information to obtain a call level of each of the quantum experiment task information; identifying the call level of each of the quantum experiment task information through a remote procedure call protocol, and sending the quantum experiment task information with the highest call level to a quantum computing simulation node; calculating the quantum experiment task information through the quantum computing simulation node to obtain operation result information; querying the operation result information based on a preset time interval, and storing the queried operation result information in the task data table.
[0007] In the embodiment of the present application, a plurality of quantum experiment tasks are stored in a task data table through a user terminal, and then the quantum experiment task is sent to an idle quantum computing simulation node in a running end state for calculation according to the time and the calling level and the calling level identified by the remote procedure call protocol, and the running result information is obtained and then transmitted back to the task data table. By real-time monitoring of the running status of the quantum simulation node, the quantum experiment task is called and assigned to the corresponding quantum computing simulation node, and the running result information of the quantum experiment task is monitored and queried in real time and fed back to the task data table. With the support of multiple quantum computing simulation nodes, the problem of slow processing speed of a single simulation node is solved, and through the remote procedure call protocol, the newly added quantum experiment tasks and the calculation results after the calculation can be monitored in real time, so that the correct task scheduling of multiple nodes can be better performed, and the monitoring and scheduling of the quantum computing simulation nodes can be realized, which greatly improves the execution efficiency of quantum experiment tasks.
[0008] In a feasible implementation, the acquired quantum experiment task information is stored in a task data table, specifically including: converting the uploaded quantum experiment tasks into the quantum experiment task information through a user terminal; storing the quantum experiment task information in the task data table; wherein the task data table includes at least any one of the following items: quantum experiment task ID, quantum experiment task content, quantum experiment task cycle, quantum experiment task call level, quantum experiment task remaining maximum time limit, and operation result.
[0009] In a feasible implementation, the calling level calculation is performed on a plurality of the quantum experiment task information to obtain the calling level of each of the quantum experiment task information, specifically including: Get the average processing time T of the i-th quantum experiment task i,ave ; Among them, Q i is the estimated amount of computation required for the ith quantum experiment task, A ave is the average computing power of the quantum computing simulation node, B i is the content parameter of the ith quantum experimental task, C ave Average quantum experiment task completion time; based on Get the urgency of the i-th quantum experiment task U i ; Among them, T i,max is the maximum remaining time limit of the i-th quantum experiment task; according to K i =λ1 L i +λ2 U i +λ3 Q i , get the information call level K of the i-th quantum experiment task i Among them, L iis the period of the i-th quantum experiment task, λ1, λ2, λ3∈[0,1] are weight parameters, and λ1+λ2+λ3=1; the information call level of each quantum experiment task is stored in the corresponding position of the task data table.
[0010] In a feasible implementation, the remote procedure call protocol includes at least any one of the following: a calling server and a calling client; the calling server is used to operate the quantum computing simulation node, and to obtain the operating status and operating results of the quantum computing simulation node in real time and send them to the calling client; wherein the operating status includes a running status and an end-of-operation status; the calling client is used to identify and transmit the calling level of quantum experiment task information in the user terminal, and to parse the quantum computing simulation node operating data obtained by the calling server to realize the operation monitoring of the quantum computing simulation node; wherein the operating data is the operating status and operating result information of the quantum computing simulation node.
[0011] The embodiment of the present application realizes real-time monitoring and scheduling of user terminals and quantum computing simulation nodes through a remote procedure call protocol. First, the running status of the quantum computing simulation node is obtained by calling the server and sent to the calling client. After the client identifies the running status, the calling level of the quantum experiment task information in the task data table is queried. Then, according to the calling level, the quantum experiment task information with a high calling level is sent to the quantum computing simulation node in the running end state. After the processing is completed, the above operation is repeated to realize real-time monitoring and scheduling of the quantum computing simulation node, improve the execution efficiency, solve the problem of slow processing of single-threaded computing nodes, and greatly speed up the node processing speed.
[0012] In a feasible implementation manner, before identifying the call level of each quantum experiment task information through a remote procedure call protocol and sending the quantum experiment task information with the highest call level to a quantum computing simulation node, the method further includes: querying the operating status of several quantum computing simulation nodes obtained by the calling client; if there is a quantum computing simulation node whose operating status is an end-of-operation state, initiating an update query on the task data table.
[0013] In a feasible implementation, the call level of each quantum experiment task information is identified through a remote procedure call protocol, and the quantum experiment task information with the highest call level is sent to a quantum computing simulation node, specifically including: updating and identifying the call level of each quantum experiment task information in the task data table through an update query of the task data table; and sending the quantum experiment task information with the highest call level to the corresponding quantum computing simulation node according to the updated and identified call level of each quantum experiment task information.
[0014] In a feasible implementation, the quantum experiment task information is calculated through the quantum computing simulation node to obtain operation result information, which specifically includes: parsing the quantum experiment task information through the task running node in the quantum computing simulation node to obtain quantum experiment task information data; and changing the operation state of the quantum computing simulation node to a running state, and sending it to a calling client; calculating the quantum experiment task information data to obtain operation result data; and caching the operation result data in a node database to obtain operation result information.
[0015] In a feasible implementation manner, after calculating the quantum experiment task information through the quantum computing simulation node to obtain the operation result information, the method also includes: changing the operation state of the quantum computing simulation node to an operation end state; and sending the operation end state corresponding to the quantum computing simulation node to the calling client of the remote procedure call protocol through the calling server in the remote procedure call protocol.
[0016] In a feasible implementation, based on a preset time interval, the operation result information is queried, and the queried operation result information is stored in the task data table, specifically including: by calling the server, the operation result information newly added to the node database after the last query time is obtained; wherein the remote procedure call protocol includes the calling server; and the obtained operation result information is stored in the task data table through the remote procedure call protocol; wherein the operation result information corresponds one-to-one to the quantum experiment task ID in the task data table.
[0017] On the other hand, an embodiment of the present application also provides a quantum computing simulation multi-node monitoring scheduling device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor so that the at least one processor can execute a quantum computing simulation multi-node monitoring scheduling method described in any of the above embodiments.
[0018] The embodiment of the present application provides a quantum computing simulation multi-node monitoring and scheduling method and device, which realizes real-time monitoring and calling of quantum computing simulation nodes through the remote procedure call protocol, monitors according to the real-time running status of the quantum computing simulation node, and sends the newly added quantum experiment task information to the corresponding executable quantum simulation computing node in real time according to the calling level, solves the problem of slow running and processing speed of single-threaded quantum simulation computing nodes, and ensures execution efficiency with the support of multiple quantum computing simulation nodes, and can also better execute the scheduling of quantum experiment tasks. By real-time monitoring and scheduling of quantum simulation computing nodes, the processing efficiency of quantum experiment tasks is improved, and the speed of processing a large number of quantum experiment tasks is accelerated. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:
[0020] Figure 1 A flow chart of a quantum computing simulation multi-node monitoring scheduling method provided in an embodiment of the present application;
[0021] Figure 2 A schematic diagram of the structure of a quantum computing simulation multi-node monitoring and scheduling device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0022] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.
[0023] The present application embodiment provides a quantum computing simulation multi-node monitoring scheduling method, such as Figure 1 As shown, the quantum computing simulation multi-node monitoring scheduling method specifically includes steps S101-S105:
[0024] S101. The user terminal stores the acquired information of several quantum experiment tasks in a task data table.
[0025] Specifically, through the user terminal, the uploaded quantum experiment tasks are converted into quantum experiment task information. Then, the quantum experiment task information is stored in a task data table. The task data table includes at least one of the following items: quantum experiment task ID, quantum experiment task content, quantum experiment task cycle, quantum experiment task call level, quantum experiment task remaining maximum time limit, and operation result.
[0026] In one embodiment, the user uploads several newly added quantum experiment tasks to the system backend through a user terminal such as a cloud platform or client software, and then converts them into several quantum experiment task information through the analysis of the system backend, and stores them in the task data table of the backend database. Each quantum experiment task in the task data table corresponds to the quantum experiment task ID, quantum experiment task content, quantum experiment task cycle, quantum experiment task call level, quantum experiment task remaining maximum time limit, and operation results.
[0027] S102: The user terminal calculates the calling level of a number of quantum experiment task information to obtain the calling level of each quantum experiment task information.
[0028] Specifically, according to Get the average processing time T of the i-th quantum experiment task i,ave Among them, Q i is the estimated amount of computation required for the ith quantum experiment task, A ave is the average computing power of quantum computing simulation nodes, B i is the content parameter of the ith quantum experimental task, C ave Average quantum experiment task completion time. Get the urgency of the i-th quantum experiment task U i ; Among them, T i,max is the maximum remaining time limit for the i-th quantum experiment task.
[0029] According to K i =λ1L i +λ2U i +λ3Q i , get the information call level K of the i-th quantum experiment task i Among them, L i is the period of the i-th quantum experiment task, λ1, λ2, λ3∈[0,1] are weight parameters, and λ1+λ2+λ3=1. Finally, the information call level of each quantum experiment task is stored in the corresponding position of the task data table.
[0030] As a feasible implementation method, the user terminal calculates each quantum experiment task call information in the task data. The call level of each quantum experiment task information is calculated by weight parameters, the cycle of the quantum experiment task, the urgency of the quantum experiment task, and the estimated calculation amount of the quantum experiment task content, and the obtained call level of each quantum experiment task information is stored in the corresponding position in the task data table.
[0031] S103. Identify the call level of each quantum experiment task information through the remote procedure call protocol, and send the quantum experiment task information with the highest call level to the quantum computing simulation node.
[0032] Specifically, the remote procedure call protocol includes at least any of the following: a calling server and a calling client. The calling server is used to run the quantum computing simulation node, and to obtain the running status and running results of the quantum computing simulation node in real time and send them to the calling client. The running status includes the running status and the running end status. The calling client is used to identify and transmit the calling level of the quantum experiment task information in the user terminal, and to parse the quantum computing simulation node running data obtained by the calling server to realize the operation monitoring of the quantum computing simulation node. The running data is the running status and running result information of the quantum computing simulation node.
[0033] Furthermore, according to the remote procedure call protocol, the running status of several quantum computing simulation nodes obtained by the calling client is queried. If the running status of any quantum computing simulation node is in the running end state, an update query of the task data table is initiated.
[0034] Furthermore, after the task data table is updated and queried, the calling level of each quantum experiment task information in the task data table is updated and identified. According to the calling level of each quantum experiment task information after the update and identification, the quantum experiment task information with the highest calling level is sent to the corresponding quantum computing simulation node.
[0035] In one embodiment, according to the remote procedure call protocol, the operation status of several quantum computing simulation nodes obtained by the calling client in the remote procedure call protocol is first queried, wherein the calling client exists in the user terminal, and is used to identify and transmit the calling level of the quantum experiment task information in the user terminal, and parse the quantum computing simulation node operation data obtained by the calling server to realize the operation monitoring of the quantum computing simulation node. When it is found that the operation status of the quantum computing simulation node is the end of operation, the calling client is started to update the task data table, and the quantum experiment task information with the highest calling level in the current task data table is identified, and the quantum experiment task information is sent to the quantum computing simulation node in the end of operation at this time, wherein the quantum computing simulation node is combined with the calling server in the remote procedure call protocol, that is, the quantum experiment task information with the highest calling level is sent to the calling server of the corresponding quantum computing simulation node.
[0036] S104. Calculate the quantum experiment task information through the quantum computing simulation node to obtain the operation result information.
[0037] Specifically, the quantum experiment task information is parsed through the task running node in the quantum computing simulation node to obtain the quantum experiment task information data. The running state of the quantum computing simulation node is changed to the running state and sent to the calling client. Then, the quantum experiment task information data is calculated to obtain the running result data. Finally, the running result data is cached in the node database to obtain the running result information.
[0038] Furthermore, after obtaining the operation result information, the operation state of the quantum computing simulation node is changed to the operation end state. The operation end state corresponding to the quantum computing simulation node is sent to the calling client of the remote procedure call protocol through the calling server in the remote procedure call protocol.
[0039] In one embodiment, the quantum experiment task information in the calling server is sent to the task running node in the quantum computing simulation node, and the quantum experiment task information is parsed to obtain quantum experiment task information data. When the calling server monitors that the task running node obtains the quantum experiment task information data, the current running state of the quantum computing simulation node is changed to the running state, and the data information of the running state is sent to the calling client, and the calling client will stop sending quantum experiment task information to the calling server of the quantum computing simulation node.
[0040] In one embodiment, when the calling server monitors that the operation result data is cached in the node database and obtains the operation result information, the operation state of the quantum computing simulation node is changed, that is, the operation end state, and the data information of the operation end state is transmitted to the calling client sent to the remote procedure call protocol, so that the calling client re-updates and identifies the task data table, and then sends the quantum experiment task information with the highest call level at this time to the calling server of the quantum computing simulation node, and then repeats the above operation. Real-time monitoring and scheduling of the state of quantum computing simulation nodes is realized, the utilization efficiency of quantum computing simulation nodes is improved, the completion speed of quantum experiment tasks is accelerated, and the processing efficiency is greatly improved.
[0041] S105. Based on a preset time interval, the remote procedure call protocol queries the operation result information, and stores the queried operation result information in the task data table.
[0042] Specifically, by calling the server, the newly added operation result information of the node database after the last query time is obtained. Among them, the remote procedure call protocol includes calling the server. And the obtained operation result information is stored in the task data table through the remote procedure call protocol. Among them, the operation result information corresponds to the quantum experiment task ID in the task data table.
[0043] As a feasible implementation method, the newly added operation result information of the node database in the quantum computing simulation node is queried in real time, and the newly added operation result information is sent to the operation result in the user terminal task data table through the remote procedure call protocol. The operation result corresponds to the quantum experiment task ID to realize real-time monitoring of the operation result information and complete the update of the task data table content.
[0044] In addition, the present application embodiment also provides a computing simulation multi-node monitoring scheduling device, such as Figure 2 As shown, the computing simulation multi-node monitoring and scheduling device 200 specifically includes:
[0045] At least one processor 201. And a memory 202 in communication with the at least one processor 201. The memory 202 stores instructions that can be executed by the at least one processor 201, so that the at least one processor 201 can execute:
[0046] The obtained information of several quantum experiment tasks is stored in the task data table;
[0047] Calculate the calling level of several quantum experiment task information to obtain the calling level of each quantum experiment task information;
[0048] Through the remote procedure call protocol, the calling level of each quantum experiment task information is identified, and the quantum experiment task information with the highest calling level is sent to the quantum computing simulation node;
[0049] Through quantum computing simulation nodes, the quantum experiment task information is calculated to obtain the operation result information;
[0050] Based on the preset time interval, query the operation result information, and store the queried operation result information in the task data table.
[0051] The present application proposes a quantum computing simulation multi-node monitoring and scheduling method and device, which solves the problem of slow running and processing speed of single-threaded quantum simulation computing nodes, and with the support of multiple quantum computing simulation nodes, ensures execution efficiency, and can better perform the scheduling of quantum experimental tasks. By real-time monitoring and scheduling of quantum simulation computing nodes, the processing efficiency of quantum experimental tasks is improved, and the speed of processing a large number of quantum experimental tasks is accelerated.
[0052] Each embodiment in this application is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.
[0053] The above describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0054] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the embodiments of the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiments of the present application should be included in the scope of the claims of the present application.
Claims
1. A quantum computing simulation multi-node monitoring scheduling method, characterized in that: The method comprises: The obtained information of several quantum experiment tasks is stored in the task data table; The calling level calculation is performed on a plurality of the quantum experiment task information to obtain the calling level of each of the quantum experiment task information, specifically including: according to , get the average processing time of the i-th quantum experiment task ;in, is the estimated amount of computation required for the i-th quantum experiment task, is the average computing power of the quantum computing simulation nodes, is the content parameter of the i-th quantum experimental task, Average quantum experiment task completion time; according to , get the urgency of the i-th quantum experiment task ;in, is the maximum remaining time limit for the i-th quantum experiment task; according to , get the information call level of the i-th quantum experiment task ;in, is the period of the ith quantum experimental task, is the weight parameter, and ; The calling level of each quantum experiment task information is stored in a corresponding position of the task data table; Through a remote procedure call protocol, the calling level of each quantum experiment task information is identified, and the quantum experiment task information with the highest calling level is sent to the quantum computing simulation node; The quantum experiment task information is calculated by the quantum computing simulation node to obtain operation result information; Based on a preset time interval, the operation result information is queried, and the queried operation result information is stored in the task data table.
2. A quantum computing simulation multi-node monitoring scheduling method according to claim 1, characterized in that: The acquired quantum experiment task information is stored in the task data table, including: Converting the uploaded quantum experiment tasks into a plurality of quantum experiment task information through a user terminal; The quantum experiment task information is stored in the task data table; wherein the task data table includes at least one of the following items: quantum experiment task ID, quantum experiment task content, quantum experiment task cycle, quantum experiment task call level, quantum experiment task remaining maximum time limit, and operation result.
3. A quantum computing simulation multi-node monitoring scheduling method according to claim 1, characterized in that: The remote procedure call protocol includes at least any one of the following: calling a server and calling a client; The calling server is used to operate the quantum computing simulation node, and obtain the operating status and operating results of the quantum computing simulation node in real time and send them to the calling client; The running state includes a running state and a running end state; The calling client is used to identify and transmit the calling level of quantum experiment task information in the user terminal, and to parse the quantum computing simulation node operation data obtained by the calling server to realize the operation monitoring of the quantum computing simulation node; wherein the operation data is the operation status and operation result information of the quantum computing simulation node.
4. A quantum computing simulation multi-node monitoring scheduling method according to claim 3, characterized in that: Before identifying the call level of each quantum experiment task information through the remote procedure call protocol and sending the quantum experiment task information with the highest call level to the quantum computing simulation node, the method further includes: Querying the operating status of the plurality of quantum computing simulation nodes obtained by the calling client; If the running state of the quantum computing simulation node exists and is in the running end state, an update query of the task data table is started.
5. A quantum computing simulation multi-node monitoring scheduling method according to claim 4, characterized in that: Through the remote procedure call protocol, the calling level of each quantum experiment task information is identified, and the quantum experiment task information with the highest calling level is sent to the quantum computing simulation node, specifically including: By updating and querying the task data table, updating and identifying the call level of each quantum experiment task information in the task data table; According to the updated and identified calling level of each quantum experiment task information, the quantum experiment task information with the highest calling level is sent to the corresponding quantum computing simulation node.
6. A quantum computing simulation multi-node monitoring scheduling method according to claim 1, characterized in that: The quantum experiment task information is calculated through the quantum computing simulation node to obtain operation result information, which specifically includes: The quantum experiment task information is parsed through the task running node in the quantum computing simulation node to obtain quantum experiment task information data; and changing the running state of the quantum computing simulation node to a running state, and sending the state to the calling client; Calculate the quantum experiment task information data to obtain operation result data; The operation result data is cached in a node database to obtain operation result information.
7. A quantum computing simulation multi-node monitoring scheduling method according to claim 6, characterized in that: After calculating the quantum experiment task information through the quantum computing simulation node to obtain the operation result information, the method further includes: Changing the running state of the quantum computing simulation node to a running end state; The running end status corresponding to the quantum computing simulation node is sent to the calling client of the remote procedure call protocol through the calling server in the remote procedure call protocol.
8. A quantum computing simulation multi-node monitoring scheduling method according to claim 1, characterized in that: Based on a preset time interval, querying the operation result information, and storing the queried operation result information in the task data table, specifically includes: By calling the server, the operation result information newly added to the node database after the last query time is obtained; wherein the remote procedure call protocol includes the calling server; The obtained operation result information is stored in the task data table through the remote procedure call protocol; wherein the operation result information corresponds one-to-one to the quantum experiment task ID in the task data table.
9. A quantum computing simulation multi-node monitoring and scheduling device, characterized in that: The device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, so that the at least one processor can execute a quantum computing simulation multi-node monitoring scheduling method according to any one of claims 1-8.
