Computing power resource delay loading method and device of intelligent computing center

Abstract

The invention provides a computing power resource delay loading method and device for an intelligent computing center, and the method is applied to an Index service node, and comprises the steps: S1, receiving a DiskANN object creation request sent by a client; sending feedback information that the creation of the target DiskANN object is completed to the client; s2, receiving a query request sent by a client; s3, the query request is forwarded to the DiskANN service node, and a return result is received; s4, if the returned result indicates that the DiskANN object does not exist, a DiskANN object creation request is sent to the DiskANN service node; moreover, after receipt information that the creation of the target DiskANN object is completed is received, returning to the step S3; and S5, if the returned result indicates that the state information in the returned result exists, sending the state information in the returned result to the client. According to the method, delayed loading of the computing power resources is realized, and the utilization efficiency of the resources is improved.

Description

Technical Field

[0001] The present invention relates to the technical fields of intelligent computing centers, intelligent computing centers, and computing power infrastructure technologies, and in particular, to a method and device for delayed loading of computing power resources in an intelligent computing center. Background Art

[0002] With the rapid development of artificial intelligence technology, "intelligent computing centers" and "intelligent computing centers" have emerged as the times require.

[0003] An "intelligent computing center" refers to a facility that provides the required computing power, data, and algorithms for artificial intelligence applications (such as artificial intelligence deep learning model development, model training, and model inference scenarios) by using large-scale heterogeneous computing power resources, including general computing power and intelligent computing power. The intelligent computing center covers facilities, hardware, and software, and can provide full-stack capabilities from underlying computing power to top-level application enablement.

[0004] The "intelligent computing center" includes, but is not limited to, the "intelligent computing center".

[0005] An "intelligent computing center", that is, an artificial intelligence computing center, is a type of computing power infrastructure that provides computing power services, data services, and algorithm services required for artificial intelligence applications based on artificial intelligence theory and using an artificial intelligence computing architecture.

[0006] "Computing power" is the core of "intelligent computing centers" and "intelligent computing centers". It is the ability of computer devices or computing / data centers to process information. It is the ability of computer hardware and software to cooperate to jointly execute a certain computing requirement. It is the computing ability to achieve the output of target results by processing information data. It is a new type of productive force integrating information computing power, network carrying capacity, and data storage capacity, and mainly provides services to society through computing power infrastructure.

[0007] In the existing method for creating a DiskANN object, the DiskANN service node immediately creates it according to the creation request from the client forwarded by the Index service node, regardless of whether the DiskANN object exists in the DiskANN service node or whether the DiskANN object will be immediately used by the client. This causes the DiskANN service node to occupy a large amount of computing power resources for a long time to create the DiskANN object, and the utilization rate of computing power resources is low. Summary of the Invention

[0008] The present invention provides a method and apparatus for delayed loading of computing power resources in an intelligent computing center, so as to solve the problem that in the existing method for creating a DiskANN object, the DiskANN service node immediately creates a DiskANN object according to a creation request from a client forwarded by the Index service node, regardless of whether the DiskANN service node has a DiskANN object or whether the DiskANN object will be immediately used by the client, resulting in the DiskANN service node occupying a large amount of computing power resources for a long time to create the DiskANN object, and the utilization rate of computing power resources is low.

[0009] To solve the above technical problems, the present invention is implemented as follows:

[0010] In a first aspect, the present invention provides a method for delayed loading of computing power resources in an intelligent computing center, which is applied to an Index service node and includes:

[0011] Step S1: Receive a DiskANN object creation request sent by a client, where the DiskANN object creation request is used to instruct a DiskANN service node to create a target DiskANN object by using the computing power resources of the intelligent computing center; send feedback information indicating that the creation of the target DiskANN object has been completed to the client;

[0012] Step S2: Receive a query request sent by the client to query the status of the target DiskANN object;

[0013] Step S3: Forward the query request to the DiskANN service node and receive a return result sent by the DiskANN service node;

[0014] Step S4: If the return result indicates that the target DiskANN object does not exist, send the DiskANN object creation request to the DiskANN service node; and after receiving the receipt information indicating that the creation of the target DiskANN object has been completed sent by the DiskANN service node, return to Step S3;

[0015] Step S5: If the return result indicates that the target DiskANN object exists, send the status information of the target DiskANN object in the return result to the client.

[0016] Optionally, before sending the feedback information indicating that the creation of the target DiskANN object has been completed to the client, it includes:

[0017] Step S01: In response to the DiskANN object creation request, create a virtual object of the target DiskANN object.

[0018] Optionally, after the step S5, the following steps are included:

[0019] Step S6: Receive a loading request sent by the client, where the loading request is used to instruct the DiskANN service node to load the target DiskANN object;

[0020] Step S7: Forward the loading request to the DiskANN service node.

[0021] Optionally, after the step S5, the following steps are included:

[0022] Step S8: Receive an import data request sent by the client for the target DiskANN object, where the import data request includes first vector data that needs to be imported into the target DiskANN object; store the first vector data in the database of the Index service node, and send a first notification message indicating the completion of the import to the client after the import is completed;

[0023] Step S9: Receive a construction request sent by the client for the target DiskANN object. In response to the construction request, import the first vector data stored in the database into the target DiskANN object, and receive a second notification message sent by the DiskANN service node, where the second notification message indicates that the first vector data has been written to disk to form a file in the target DiskANN object; forward the construction request to the DiskANN service node, so that the DiskANN service node processes the first vector data imported into the target DiskANN object based on the construction request to obtain a graph index of the first vector data, and save the first vector data and the graph index.

[0024] In a second aspect, the present invention provides a computing power resource delayed loading device for an intelligent computing center, which is applied to an Index service node and includes:

[0025] A receiving module, configured to receive a DiskANN object creation request sent by a client, where the DiskANN object creation request is used to instruct a DiskANN service node to create a target DiskANN object using the computing power resources of the intelligent computing center; send feedback information indicating that the creation of the target DiskANN object has been completed to the client;

[0026] The receiving module is further configured to receive a query request sent by the client to query the status of the target DiskANN object;

[0027] A sending module, configured to forward the query request to the DiskANN service node and receive a return result sent by the DiskANN service node;

[0028] A first execution module, configured to, if the return result indicates that the target DiskANN object does not exist, send a DiskANN object creation request to the DiskANN service node; and, after receiving a receipt message sent by the DiskANN service node indicating that the creation of the target DiskANN object has been completed, return to the sending module to execute the step of forwarding the query request to the DiskANN service node and receiving a return result sent by the DiskANN service node;

[0029] A second execution module, configured to, if the return result indicates that the target DiskANN object exists, send the status information of the target DiskANN object in the return result to the client.

[0030] Optionally, the receiving module is further configured to receive a loading request sent by the client, where the loading request is used to instruct the DiskANN service node to load the target DiskANN object;

[0031] The sending module is further configured to forward the loading request to the DiskANN service node.

[0032] Optionally, it further includes a third execution module;

[0033] The receiving module is further configured to receive an import data request sent by the client for the target DiskANN object, where the import data request includes first vector data that needs to be imported into the target DiskANN object; store the first vector data in the database of the Index service node, and send a first notification message indicating that the import is completed to the client after the import is completed;

[0034] The third execution module is configured to receive a construction request sent by the client for the target DiskANN object, in response to the construction request, import the first vector data stored in the database into the target DiskANN object, and receive a second notification message sent by the DiskANN service node, where the second notification message indicates that the first vector data has been written to disk to form a file in the target DiskANN object; forward the construction request to the DiskANN service node, so that the DiskANN service node processes the first vector data imported into the target DiskANN object based on the construction request to obtain a graph index of the first vector data, and save the first vector data and the graph index.

[0035] In a third aspect, the present invention provides an electronic device, including a processor, a memory, and a program or instruction stored on the memory and executable on the processor. When the program or instruction is executed by the processor, the steps in the computing power resource delayed loading method of the intelligent computing center as described in any one of the first aspects are implemented.

[0036] In a fourth aspect, the present invention provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps in the computing power resource delayed loading method of the intelligent computing center as described in any one of the first aspects are implemented.

[0037] In a fifth aspect, the present invention provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, the steps of the computing power resource delayed loading method of the intelligent computing center as described in any one of the first aspects are implemented.

[0038] In the implementation of the present invention, through steps S1 to S5, the present invention realizes that in the case where the customer requirements are not met, the execution process of creating a target DiskANN object by using the computing power resources of the intelligent computing center is triggered, that is, creating the target DiskANN object on demand. Instead of immediately instructing the DiskANN service node to create the target DiskANN object by using the computing power resources of the intelligent computing center as long as a DiskANN object creation request is received from the client, the on-demand creation of the target DiskANN object in the present invention realizes the delayed creation of the target DiskANN object, thereby realizing the delayed loading of the computing power resources of the intelligent computing center, reducing the occupation of computing power resources, and improving the utilization efficiency of computing power resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] By reading the detailed description of the preferred embodiments below, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0040] Figure 1 is a schematic flowchart of the computing power resource delayed loading method of the intelligent computing center of the present invention;

[0041] Figure 2 is a schematic execution flowchart;

[0042] Figure 3 is a schematic block diagram of the computing power resource delayed loading device of the intelligent computing center of the present invention;

[0043] Figure 4This is the principle block diagram of the electronic device of the present invention. Detailed implementation manners

[0044] Next, the technical solutions of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts fall within the protection scope of the present invention.

[0045] The terms "first", "second", etc. in the present invention are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that such terms can be interchanged under appropriate circumstances so that the present invention can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first" and "second" are usually of the same category, and the number of objects is not limited. For example, the first object can be one or multiple. In addition, "or" in the present invention means at least one of the connected objects. For example, "A or B" covers three scenarios, namely, Scenario 1: including A and not including B; Scenario 2: including B and not including A; Scenario 3: including both A and B. The character " / " generally indicates an "or" relationship between the associated objects before and after.

[0046] In addition, the technical features involved in different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0047] It should be noted that in the technical solutions of the present invention, in terms of the collection, gathering, updating, analysis, processing, use, transmission, storage, etc. of personal information, they all comply with the provisions of relevant laws and regulations, are used for legal purposes, and do not violate public order and good customs. Necessary measures are taken for personal information to prevent illegal access to personal information data and to maintain personal information security and network security.

[0048] First, the technical terms involved in the present invention will be briefly described below.

[0049] The "computing power" described in the present invention refers to: the ability of a computer device or a computing / data center to process information, the ability of computer hardware and software to cooperate to jointly execute a certain computing requirement, the computing ability to process information data and output a target result, a new type of productive force integrating information computing power, network carrying capacity, and data storage capacity, and mainly providing services to society through computing power infrastructure.

[0050] The "Computational Power (CP)" described in the present invention refers to: the ability of a data center server to process data and output results, which is a comprehensive indicator for measuring the computing power of a data center and includes general computing power, supercomputing power, and intelligent computing power. The commonly used measurement unit is the number of floating-point operations per second (FLOPS, 1 EFLOPS = 10^18 FLOPS), and the larger the value, the stronger the comprehensive computing power. It is estimated that 1 EFLOPS is approximately the computing power output of 5 Tianhe-2A computers or 500,000 mainstream server CPUs or 2 million mainstream laptops. The calculation formula is: CP = CP 通用 + CP 智能 + CP 超级 。

[0051] The "Network Power (NP)" described in the present invention refers to: the manifestation of the data transmission ability of computing power facilities, which is a comprehensive ability including network architecture, network bandwidth, transmission delay, intelligent management and scheduling, etc., and involves network transmission within and between data centers, and is a comprehensive indicator for measuring network transmission scheduling ability.

[0052] The "Storage Power (SP)" described in the present invention refers to: the comprehensive ability of a data center in four aspects: data storage capacity, performance, security and reliability, and green and low-carbon, which is a comprehensive indicator for measuring the data storage ability of a data center and includes external storage devices such as storage arrays and server internal storage devices. The commonly used measurement unit for storage capacity is exabyte (EB, 1 EB = 2^60 bytes), the commonly used measurement unit for performance is the number of read and write operations per second per unit capacity (IOPS / TB, Input / Output Operations Per Second / TB), and the disaster recovery ratio is an important manifestation of security and reliability.

[0053] The "computing power infrastructure" described in the present invention refers to: a new type of information infrastructure that integrates information computing power, network carrying power, and data storage power, which can realize the centralized computing, storage, transmission, and application of information, and presents characteristics such as multi-element ubiquitous, intelligent and agile, safe and reliable, green and low-carbon, etc., and is of great significance for boosting industrial transformation and upgrading, empowering China's scientific and technological innovation, meeting people's beautiful life, and realizing high-efficiency social governance.

[0054] The "new information infrastructure" described in the present invention refers to: mainly including network infrastructures such as 5G networks, fiber broadband networks, backbone networks, international communication networks, satellite Internet, etc., computing power infrastructures such as data centers, general computing power centers, intelligent computing centers, supercomputing centers, etc., and new technology facilities such as artificial intelligence, blockchain, and quantum computing. With the emergence and popularization of new general technologies, the form of the new information infrastructure will be more diverse.

[0055] The "computing power" described in the present invention includes: general computing power, intelligent computing power, and super computing power.

[0056] The "general computing power" described in the present invention refers to: the computing power provided by servers based on CPU (Central Processing Unit) chips, which is used to support basic general computing such as cloud computing and edge computing.

[0057] The "intelligent computing power" described in the present invention refers to: for various artificial intelligence innovation applications, a computing platform deployed on a large scale based on dedicated chips such as GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), and ASIC (Application Specific Integrated Circuit), such as natural language processing, machine vision, etc.

[0058] The "super computing power" described in the present invention refers to: mainly the computing power provided by high-performance computing clusters such as supercomputers. It utilizes the centralized computing resources of multiple computer systems working in parallel and processes extremely complex or data-intensive problems through a dedicated operating system, mainly used for computing in cutting-edge scientific fields, such as planetary simulation, drug molecule design, gene analysis, etc.

[0059] The "intelligent computing center" described in the present invention refers to: a facility that mainly provides the required computing power, data, and algorithms for artificial intelligence applications (such as scenarios of artificial intelligence deep learning model development, model training, and model inference) by using large-scale heterogeneous computing power resources, including general computing power (CPU) and intelligent computing power (GPU, FPGA, ASIC, etc.). The intelligent computing center covers facilities, hardware, and software, and can provide full-stack capabilities from underlying computing power to top-level application enabling.

[0060] The "intelligent computing center" described in the present invention includes but is not limited to the "intelligent computing center".

[0061] The "intelligent computing center" described in the present invention, namely the artificial intelligence computing center, is a type of computing power infrastructure that provides computing power services, data services, and algorithm services required for artificial intelligence applications based on artificial intelligence theory and using an artificial intelligence computing architecture.

[0062] The "computing power center" described in the present invention refers to: a facility mainly composed of infrastructure such as wind, fire, water, and electricity and IT software and hardware devices, with computing power, carrying capacity, and storage capacity, including general data centers, intelligent computing centers, supercomputing centers, etc.

[0063] The "supercomputing center" described in the present invention refers to: that is, a supercomputing data center, which is a data center based on supercomputers or large-scale computing clusters, and can provide functions such as large-scale computing, storage, and network services, and is widely used in application scenarios such as aerospace, national defense, oil exploration, climate modeling, and genome sequencing.

[0064] The "computing power resources" described in the present invention refer to: technologies and facilities with information computing, transmission, storage, and application capabilities required for the development of the digital society, including but not limited to computing resources such as CPUs and GPUs, network resources such as switches and routers, storage resources such as storage arrays and distributed storage, security resources such as firewalls and intrusion detection systems, and support and guarantee resources such as wind, fire, water, and electricity.

[0065] The "DiskANN" described in the present invention is a vector retrieval engine based on distributed storage, which can store and retrieve vector data at the billion level on a single computer. Compared with traditional vector retrieval algorithms, DiskANN has higher storage efficiency and faster retrieval speed.

[0066] The present invention provides a method for delayed loading of computing power resources of an intelligent computing center, which is applied to an Index service node. See Figure 1 as shown Figure 1 is a schematic flow diagram of the method for delayed loading of computing power resources of the intelligent computing center of the present invention, including:

[0067] Step S1: Receive a DiskANN object creation request sent by a client, where the DiskANN object creation request is used to instruct the DiskANN service node to create a target DiskANN object using the computing power resources of the intelligent computing center; send feedback information indicating that the creation of the target DiskANN object has been completed to the client;

[0068] Step S2: Receive a query request sent by the client to query the status of the target DiskANN object;

[0069] Step S3: Forward the query request to the DiskANN service node and receive the return result sent by the DiskANN service node;

[0070] Step S4: if the returned result indicates that the target DiskANN object does not exist, send a DiskANN object creation request to the DiskANN service node; and after receiving a receipt message from the DiskANN service node indicating that the target DiskANN object has been created, return to step S3;

[0071] Step S5: If the returned result indicates that the target DiskANN object exists, the status information of the target DiskANN object in the returned result is sent to the client.

[0072] In step S1 of the present invention, after receiving the DiskANN object creation request sent by the client, the Index service node does not send the DiskANN object creation request to the DiskANN service node. In other words, the DiskANN service node does not use the computing power resources of the intelligent computing center to create the target DiskANN object, but directly reports the creation completion to the client (that is, sending feedback information of the completion of the creation of the target DiskANN object to the client in the present invention).

[0073] In the above case, if the client initiates a query request for status query of the target DiskANN object (corresponding to step S2), then step S3 is entered, and the Index service node forwards the query request to the DiskANN service node and receives the return result sent by the DiskANN service node.

[0074] Afterwards, step S4: if the returned result indicates that the target DiskANN object does not exist, send a DiskANN object creation request to the DiskANN service node; and after receiving the receipt information sent by the DiskANN service node indicating that the creation of the target DiskANN object has been completed, return to step S3. That is to say, if the returned result indicates that the target DiskANN object does not exist, the Index service node instructs the DiskANN service node to use the computing power resources of the intelligent computing center to create the target DiskANN object. The present invention realizes that when the customer demand (the client initiates a query request for status query of the target DiskANN object) is not met (the returned result indicates that the target DiskANN object does not exist), the execution process of using the computing power resources of the intelligent computing center to create the target DiskANN object is triggered, that is, the target DiskANN object is created on demand. Instead of immediately instructing the DiskANN service node to use the computing power resources of the intelligent computing center to create the target DiskANN object upon receiving the DiskANN object creation request sent by the client, the present invention delays the creation of the target DiskANN object, realizes the delayed loading of the computing power resources of the intelligent computing center, reduces the occupancy of computing power resources (especially memory resources), and improves the utilization efficiency of computing power resources.

[0075] After that, step S5: If the return result indicates the existence of a target DiskANN object, send the status information of the target DiskANN object in the return result to the client. That is to say, if there is a target DiskANN object, the query request from the client is completed.

[0076] In the present invention, the status information of the DiskANN object includes at least one of the following:

[0077] Dataset information (e.g., the number of data points, the dimension of each data point), index structure (e.g., the index type used (e.g., graph-based index or tree structure), the construction parameters of the index (such as the number of neighbors, distance metric, etc.)), query parameters (e.g., the number of queries, the type of queries), performance metrics (e.g., query time, memory usage, recall rate, and precision rate (if there is labeled data)), status flags (e.g., whether it has been initialized, whether data has been loaded, whether a query is being executed), log information (e.g., the timestamp of the operation, error, and warning information).

[0078] In the implementation of the present invention, through steps S1 to S5, the present invention realizes the execution process of creating a target DiskANN object by triggering the use of the computing power resources of the intelligent computing center only when the customer requirements are not met, that is, creating the target DiskANN object on demand. Instead of immediately instructing the DiskANN service node to create the target DiskANN object by using the computing power resources of the intelligent computing center as long as a DiskANN object creation request is received from the client, the on-demand creation of the target DiskANN object in the present invention realizes the delayed creation of the target DiskANN object, thereby realizing the delayed loading of the computing power resources of the intelligent computing center, reducing the occupancy of the computing power resources, and improving the utilization efficiency of the computing power resources.

[0079] In some embodiments of the present invention, optionally, before sending the feedback information indicating that the creation of the target DiskANN object has been completed to the client, it includes:

[0080] Step S01: In response to the DiskANN object creation request, create a virtual object of the target DiskANN object.

[0081] It should be noted that the virtual object of the target DiskANN object can be a virtual graph and does not have any functions of the target DiskANN object. In the present invention, the role of creating the virtual object of the target DiskANN object in the Index service node is to display it to the user associated with the client; or, it is used to virtually display the graph of the target DiskANN object in the set area according to the classification settings preset by the user in the Index service node, thereby improving the user experience.

[0082] In some embodiments of the present invention, optionally, after step S5, the following steps are included:

[0083] Step S6: Receive a loading request sent by the client, where the loading request is used to instruct the DiskANN service node to load the target DiskANN object;

[0084] Step S7: Forward the loading request to the DiskANN service node.

[0085] The loading task refers to the client sending a loading request to the DiskANN service node through the Index service node, and the DiskANN service node loading at least part of the graph index of the vector data of the DiskANN object into the memory based on the loading request. After the loading is completed, the DiskANN service node notifies the client through the Index service node.

[0086] In the present invention, through steps S6 and S7, it is possible to perform a loading task on the target DiskANN object, that is, load at least part of the graph index of the vector data of the target DiskANN object into the memory.

[0087] In some embodiments of the present invention, optionally, after step S5, the following steps are included:

[0088] Step S8: Receive an import data request for the target DiskANN object sent by the client, where the import data request includes first vector data that needs to be imported into the target DiskANN object; store the first vector data in the database of the Index service node, and send a first notification message indicating the completion of the import to the client after the import is completed;

[0089] Step S9: Receive a construction request for the target DiskANN object sent by the client. In response to the construction request, import the first vector data stored in the database into the target DiskANN object, and receive a second notification message sent by the DiskANN service node, where the second notification message indicates that the first vector data has been written to disk and formed a file in the target DiskANN object; forward the construction request to the DiskANN service node, so that the DiskANN service node processes the first vector data imported into the target DiskANN object based on the construction request to obtain the graph index of the first vector data, and save the first vector data and the graph index.

[0090] In the present invention, in the task of importing vector data to the DiskANN object, the client sends an ImportData request (i.e., an import data request) containing vector data to the Index service node, and the Index service node temporarily stores the vector data in the ImportData request in the database of the Index service node. The Index service node notifies the client after completing the data import.

[0091] After the data import task, the client can send a build request to the Index service node. The Index service node executes the Push Data task according to the build request, that is, the Index service node imports the vector data in the database of the Index service node to the DiskANN service node. The DiskANN service node notifies the Index service node after the vector data is stored on the disk to form a file.

[0092] In the present invention, the Index service node forwards the build (Bulid) request to the DiskANN service node, and the DiskANN service node performs a build (Bulid) task based on the build (Bulid) request. The build (Bulid) task refers to the DiskANN service node processing the vector data of the imported DiskANN object based on the build (Bulid) request sent by the Index service node, obtaining the graph index of the vector data, and saving the vector data and the graph index. After completing the build (Bulid) task, the DiskANN service node notifies the client through the Index service node.

[0093] The following is a description with reference to specific embodiments.

[0094] See also Figure 2 As shown, Figure 2 The execution process is shown in the following figure. The execution process includes:

[0095] The client initiates a create request (i.e., a DiskANN object creation request) to the Index service node, and the Index service node creates a virtual object.

[0096] The Index service node creates a network connection with the DiskANN service node, initiates a Status RPC (i.e., a query request) to the DiskANN service node, and returns a failure because the object does not exist on the DiskANN service node.

[0097] The Index service node initiates a request to create an object, and the DiskANN service node creates a DIskANN object (ie, the target DIskANN object).

[0098] The Index service node automatically retries and initiates the Status RPC (i.e., the query request) to the DiskANN service node again, and returns to the normal state.

[0099] For the client, the above execution process is imperceptible.

[0100] The present invention provides a device for delaying the loading of computing power resources in an intelligent computing center, which is applied to the Index service node. See Figure 3 as shown in Figure 3 is the principle block diagram of the device for delaying the loading of computing power resources in the intelligent computing center of the present invention. The device 30 for delaying the loading of computing power resources in the intelligent computing center includes:

[0101] A receiving module 31, configured to receive a DiskANN object creation request sent by a client, where the DiskANN object creation request is used to instruct the DiskANN service node to create a target DiskANN object by using the computing power resources of the intelligent computing center; and send feedback information indicating that the creation of the target DiskANN object has been completed to the client.

[0102] The receiving module 31 is further configured to receive a query request sent by the client to query the status of the target DiskANN object.

[0103] A sending module 32, configured to forward the query request to the DiskANN service node and receive a return result sent by the DiskANN service node.

[0104] A first execution module 33, configured to, if the return result indicates that the target DiskANN object does not exist, send the DiskANN object creation request to the DiskANN service node; and, after receiving a receipt message sent by the DiskANN service node indicating that the creation of the target DiskANN object has been completed, return to the sending module to execute the steps of forwarding the query request to the DiskANN service node and receiving a return result sent by the DiskANN service node.

[0105] A second execution module 34, configured to, if the return result indicates that the target DiskANN object exists, send the status information of the target DiskANN object in the return result to the client.

[0106] In some embodiments of the present invention, optionally,

[0107] The receiving module 31 is further configured to create a virtual object of the target DiskANN object in response to the DiskANN object creation request.

[0108] In some embodiments of the present invention, optionally,

[0109] The receiving module 31 is further configured to receive a loading request sent by the client, where the loading request is used to instruct the DiskANN service node to load the target DiskANN object;

[0110] The sending module 32 is further configured to forward the loading request to the DiskANN service node.

[0111] In some embodiments of the present invention, optionally, a third execution module is further included;

[0112] The receiving module 31 is further configured to receive an import data request sent by the client for the target DiskANN object, where the import data request includes first vector data to be imported into the target DiskANN object; store the first vector data in the database of the Index service node, and send a first notification message indicating the import is completed to the client after the import is completed;

[0113] The third execution module is configured to receive a construction request sent by the client for the target DiskANN object, in response to the construction request, import the first vector data stored in the database into the target DiskANN object, and receive a second notification message sent by the DiskANN service node, where the second notification message indicates that the first vector data has been written to disk to form a file in the target DiskANN object; forward the construction request to the DiskANN service node, so that the DiskANN service node processes the first vector data imported into the target DiskANN object based on the construction request to obtain a graph index of the first vector data, and save the first vector data and the graph index.

[0114] The computing power resource delayed loading device of the intelligent computing center provided by the present invention can implement Figures 1 to 2 each process implemented by the method embodiments and achieve the same technical effects. To avoid repetition, they will not be elaborated here.

[0115] The present invention provides an electronic device 40, see Figure 4 as shown, Figure 4 is a principle block diagram of the electronic device 40 of the present invention, including a processor 41, a memory 42, and a program or instruction stored in the memory 42 and executable on the processor 41. When the program or instruction is executed by the processor, the steps in any of the computing power resource delayed loading methods of the intelligent computing center of the present invention are implemented.

[0116] The present invention provides a readable storage medium, on which a program or instructions are stored. When the program or instructions are executed by a processor, the various processes of the embodiment of the computing power resource delayed loading method of the intelligent computing center as described in any one of the above are implemented, and the same technical effects can be achieved. To avoid repetition, they will not be elaborated here.

[0117] Among them, the readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc, etc. In some examples, the readable storage medium may be a non-transitory readable storage medium.

[0118] The present invention also provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, the various processes of the embodiment of the computing power resource delayed loading method of the intelligent computing center as described in any one of the above are implemented, and the same technical effects can be achieved. To avoid repetition, they will not be elaborated here.

[0119] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the purpose of the present invention and the scope protected by the claims, and all of them fall within the protection scope of the present invention.

Claims

1. A method for delayed loading computing resources of an intelligent computing center, characterized in that: Applied to Index service nodes, including: Step S1: receiving a DiskANN object creation request sent by a client, wherein the DiskANN object creation request is used to instruct the DiskANN service node to use computing resources of an intelligent computing center to create a target DiskANN object; and sending feedback information that the target DiskANN object has been created to the client; Step S2: receiving a query request sent by the client to query the status of the target DiskANN object; Step S3: forwarding the query request to the DiskANN service node, and receiving a return result sent by the DiskANN service node; Step S4: if the returned result indicates that the target DiskANN object does not exist, send the DiskANN object creation request to the DiskANN service node; and after receiving the receipt information sent by the DiskANN service node indicating that the target DiskANN object has been created, return to step S3; Step S5: If the returned result indicates that the target DiskANN object exists, the status information of the target DiskANN object in the returned result is sent to the client.

2. The method for delayed loading computing resources of an intelligent computing center according to claim 1, characterized in that: Sending feedback information that the target DiskANN object has been created to the client, including: Step S01: In response to the DiskANN object creation request, a virtual object of the target DiskANN object is created.

3. The method for delayed loading computing resources of an intelligent computing center according to claim 1, characterized in that: The step S5 then includes: Step S6: receiving a loading request sent by the client, wherein the loading request is used to instruct the DiskANN service node to load the target DiskANN object; Step S7: forwarding the loading request to the DiskANN service node.

4. The method for delayed loading computing resources of an intelligent computing center according to claim 1, characterized in that: The step S5 then includes: Step S8: receiving an import data request for the target DiskANN object sent by the client, wherein the import data request includes first vector data to be imported into the target DiskANN object; storing the first vector data in a database of the Index service node, and sending a first notification message of import completion to the client after the import is completed; Step S9: receiving a build request for the target DiskANN object sent by the client, importing the first vector data stored in the database into the target DiskANN object in response to the build request, and receiving a second notification message sent by the DiskANN service node, wherein the second notification message indicates that the first vector data has been stored on the disk to form a file of the target DiskANN object; forwarding the build request to the DiskANN service node, so that the DiskANN service node processes the first vector data imported into the target DiskANN object based on the build request, obtains a graph index of the first vector data, and saves the first vector data and the graph index.

5. A computing resource delayed loading device for an intelligent computing center, characterized in that: Applied to Index service nodes, including: A receiving module, configured to receive a DiskANN object creation request sent by a client, wherein the DiskANN object creation request is used to instruct the DiskANN service node to use the computing power resources of the intelligent computing center to create a target DiskANN object; and to send feedback information to the client indicating that the creation of the target DiskANN object has been completed; The receiving module is further configured to receive a query request sent by the client to query the status of the target DiskANN object; A sending module, used for forwarding the query request to the DiskANN service node and receiving a return result sent by the DiskANN service node; The first execution module is used to send the DiskANN object creation request to the DiskANN service node if the return result indicates that the target DiskANN object does not exist; and after receiving the receipt information sent by the DiskANN service node that the target DiskANN object has been created, return to the sending module to execute the steps of forwarding the query request to the DiskANN service node and receiving the return result sent by the DiskANN service node; The second execution module is configured to send the status information of the target DiskANN object in the returned result to the client if the returned result indicates that the target DiskANN object exists.

6. The computing resource delayed loading device of the intelligent computing center according to claim 5, characterized in that: The receiving module is further used to receive a loading request sent by the client, wherein the loading request is used to instruct the DiskANN service node to load the target DiskANN object; The sending module is further used to forward the loading request to the DiskANN service node.

7. The computing resource delayed loading device of the intelligent computing center according to claim 5, characterized in that: Also includes a third execution module; The receiving module is further used to receive an import data request for the target DiskANN object sent by the client, wherein the import data request includes first vector data that needs to be imported into the target DiskANN object; store the first vector data in a database of the Index service node, and send a first notification message of import completion to the client after the import is completed; The third execution module is used to receive a construction request for the target DiskANN object sent by the client, import the first vector data stored in the database into the target DiskANN object in response to the construction request, and receive a second notification message sent by the DiskANN service node, wherein the second notification message indicates that the first vector data has been stored in the target DiskANN object to form a file; The construction request is forwarded to the DiskANN service node, so that the DiskANN service node processes the first vector data imported into the target DiskANN object based on the construction request, obtains a graph index of the first vector data, and saves the first vector data and the graph index.

8. An electronic device, characterized in that: It includes a processor, a memory, and a program or instruction stored in the memory and executable on the processor. When the program or instruction is executed by the processor, the steps in the method for delayed loading of computing resources of an intelligent computing center as described in any one of claims 1 to 4 are implemented.

9. A readable storage medium, characterized in that: The readable storage medium stores programs or instructions, and when the programs or instructions are executed by the processor, the steps in the method for delayed loading of computing resources of an intelligent computing center as described in any one of claims 1 to 4 are implemented.

10. A computer program product, characterized in that It comprises computer instructions, which, when executed by a processor, implement the steps of the method for delayed loading of computing resources of an intelligent computing center as described in any one of claims 1 to 4.

Images