Cloud data warehouse, data reading method, server and computer storage medium
By retrieving the location information of the target execution node in the computing node of the cloud data warehouse and selecting the path to read the data based on the results, the problem that the cloud data warehouse cannot cache shared data is solved, and query performance is improved.
Patent Information
- Application Number
- CN202411715266.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-27
- Publication Date
- 2025-03-18
AI Technical Summary
Cloud data warehouses cannot effectively cache shared data, resulting in reduced query performance.
After the computing node receives the query request, the location information of the target execution node is retrieved in the administrator node. If the target data is in the local executor node, it is read through the local file system; if it is not local, it is read in the target execution node, and a copy is written in the executor node corresponding to the computing node; if the location information of the target execution node is not retrieved, the data is read in the object store.
It effectively solves the problem that cloud data warehouse cannot cache and share, and improves query performance.
Smart Images
Figure CN119646040A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of data processing technology, and in particular to a cloud data warehouse, a data reading method, a server and a computer storage medium. Background Art
[0002] Cloud data warehouse is a new generation of data warehouse that is deployed and operated based on cloud environment. Cloud data warehouse can be divided into a three-layer architecture of cloud service layer, computing layer and storage layer. The management layer provides tenant access portal, resource management, metadata management, transaction management, parsing and execution scheduling services; the computing layer provides and manages computing resources of different specifications, and provides cache acceleration support; the storage layer provides object storage service and supports the persistence of result set cache. The storage layer is usually deployed in the IaaS layer, and the computing layer runs on K8S, realizing the separation of storage and computing. K8S: The full name is Kubernetes, which is a portable container orchestration management tool for container services. A container is an instance of an image at runtime. It has the characteristics of rapid deployment and delivery, efficient virtualization, easy migration and expansion, and simple management.
[0003] Theoretically, the object storage service provides a storage system with unlimited space, but its performance is lower than that of cloud disks or physical hard disks. In order to improve the operating efficiency of the computing cluster, the cloud data warehouse will use local memory or hard disk as the cache of the object storage service to save user hot data. In other words, the cache in the computing cluster is a local cache, but the network transmission speed between computing clusters is usually faster than the speed between computing nodes and the underlying object storage. Therefore, in actual application, the query performance of the cloud data warehouse will be reduced. Summary of the invention
[0004] In view of this, the present application provides a cloud data warehouse, a data reading method, a server and a computer storage medium to solve the problem that the cloud data warehouse cannot cache and share, and effectively improve the query performance.
[0005] The first aspect of the present application provides a cloud data warehouse, including:
[0006] N tenant spaces and manager nodes; wherein the tenant space includes M physical nodes; N and M are positive integers; each of the physical nodes includes a computing node and an executor node; the computing node and the executor node are deployed on the physical node through affinity configuration;
[0007] The computing node is used to receive a query request; wherein the query request includes at least one target data; the target data is the data that needs to be read this time;
[0008] The computing node is further used to retrieve the location information of the target execution node in the manager node to obtain a search result; wherein the location information of the target execution node is the location information of the executor node containing the target data;
[0009] If the search result indicates that the location information of the target execution node is the location information of the executor node corresponding to the computing node, the computing node reads the target data through the local file system;
[0010] If the search result indicates that the location information of the target execution node is not the location information of the executor node corresponding to the computing node, the computing node is further used to read the target data in the target execution node, and after the computing node reads the target data in the target execution node, writes a copy of the target data in the executor node corresponding to the computing node;
[0011] If the search result indicates that the location information of the target execution node is not retrieved, the computing node is further configured to read the target data in the object storage.
[0012] Optionally, the manager node is responsible for servicing computing node requests in the cloud data warehouse and managing global metadata of all executor nodes.
[0013] Optionally, the manager node is deployed on an independent physical machine.
[0014] Optionally, the executor node is responsible for running the tasks assigned by the manager node and managing and allocating local resources.
[0015] Optionally, the configuration file of the executor node includes the address information of the manager node, and after the executor node is started, a registration application is made to the manager node.
[0016] Optionally, the manager node detects the running status of the executor node through a heartbeat mechanism.
[0017] Optionally, if the search result indicates that the location information of the target execution node is not retrieved, the manager node entrusts the target node to read the target data in the object storage.
[0018] A second aspect of the present application provides a data reading method, which is applied to a cloud data warehouse, wherein the cloud data warehouse includes N tenant spaces and a manager node; wherein the tenant space includes M physical nodes; N and M are positive integers; each of the physical nodes includes a computing node and an executor node; the computing node and the executor node are deployed on the physical node through affinity configuration, and the data reading method includes:
[0019] The computing node receives a query request; wherein the query request includes at least one target data; the target data is the data that needs to be read this time;
[0020] The computing node retrieves the location information of the target execution node in the manager node to obtain a search result; wherein the location information of the target execution node is the location information of the executor node containing the target data;
[0021] If the search result indicates that the location information of the target execution node is the location information of the executor node corresponding to the computing node, the computing node reads the target data through the local file system;
[0022] If the search result indicates that the location information of the target execution node is not the location information of the executor node corresponding to the computing node, the computing node reads the target data in the target execution node;
[0023] After the computing node reads the target data in the target execution node, the computing node writes a copy of the target data in the executor node corresponding to the computing node;
[0024] If the search result indicates that the location information of the target execution node is not retrieved, the computing node reads the target data in the object storage.
[0025] Optionally, the manager node is responsible for servicing computing node requests in the cloud data warehouse and managing global metadata of all executor nodes.
[0026] Optionally, the manager node is deployed on an independent physical machine.
[0027] Optionally, the executor node is responsible for running the tasks assigned by the manager node and managing and allocating local resources.
[0028] Optionally, the configuration file of the executor node includes the address information of the manager node, and after the executor node is started, a registration application is made to the manager node.
[0029] Optionally, the manager node detects the running status of the executor node through a heartbeat mechanism.
[0030] Optionally, if the search result indicates that the location information of the target execution node is not retrieved, the manager node entrusts the target node to read the target data in the object storage.
[0031] A third aspect of the present application provides a server, including:
[0032] one or more processors;
[0033] a storage device having one or more programs stored thereon;
[0034] When the one or more programs are executed by the one or more processors, the one or more processors implement the data reading method as described in any one of the second aspects.
[0035] A fourth aspect of the present application provides a computer storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the data reading method as described in any one of the second aspects is implemented.
[0036] As can be seen from the above scheme, the present application provides a cloud data warehouse, a data reading method, a server and a computer storage medium, wherein computing nodes and executor nodes are deployed on physical nodes through affinity configuration, and when the computing node receives a query request, the location information of the target execution node is retrieved from the manager node to obtain a search result; wherein the location information of the target execution node is the location information of the executor node containing the target data; if the search result shows that the location information of the target execution node is the location information of the executor node corresponding to the computing node, the computing node reads the target data through the local file system; if the search result shows that the location information of the target execution node is not the location information of the executor node corresponding to the computing node, the computing node reads the target data in the target execution node; after the computing node reads the target data in the target execution node, a copy of the target data is written in the executor node corresponding to the computing node; if the search result shows that the location information of the target execution node is not retrieved, the computing node reads the target data in the object storage. Thus, the problem that the cloud data warehouse cannot cache and share is solved, and the query performance is effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] In order to more clearly illustrate the embodiments of the present invention 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, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.
[0038] Figure 1 A specific flow chart of a data reading method provided in an embodiment of the present application;
[0039] Figure 2 A schematic diagram showing the difference between a prior art and the present invention provided for another embodiment of the present application;
[0040] Figure 3A schematic diagram of a cloud data warehouse provided in another embodiment of the present application;
[0041] Figure 4 A schematic diagram of a server for implementing a method for reading data provided in another embodiment of the present application. DETAILED DESCRIPTION
[0042] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0043] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.
[0044] It should be noted that the information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.
[0045] It should be noted that the concepts such as "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.
[0046] It should be noted that the modifications of "one" and "plurality" mentioned in the present application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".
[0047] The present application embodiment provides a data reading method, which is applied to a cloud data warehouse, wherein the cloud data warehouse includes N tenant spaces and a manager node; wherein the tenant space includes M physical nodes; N and M are positive integers; each physical node includes a computing node and an executor node; the computing node and the executor node are deployed on the physical node through affinity configuration, such as Figure 1 As shown, the method for reading data includes:
[0048] S101. A computing node receives a query request.
[0049] The query request includes at least one target data; the target data is the data that needs to be read this time.
[0050] S102: The computing node searches the manager node for location information of the target execution node and obtains a search result.
[0051] The location information of the target execution node is the location information of the executor node containing the target data.
[0052] S103. If the search result indicates that the location information of the target execution node is the location information of the executor node corresponding to the computing node, the computing node reads the target data through the local file system.
[0053] It is understandable that when the data that the computing node of the cloud data warehouse needs to read has been cached on the local executor (Worker) node, it is a local cache hit. The computing node retrieves the location information of the Worker node that stores the data from the manager (Master) node. If the local Worker node stores the data, the computing node can directly read the file through the local file system.
[0054] S104. If the search result indicates that the location information of the target execution node is not the location information of the executor node corresponding to the computing node, the computing node reads the target data in the target execution node.
[0055] It is understandable that when the target data requested by the cloud data warehouse computing node is not on the local Worker node, but on a remote Worker node in the computing cluster, the cloud data warehouse computing node will read the data from the remote Worker node.
[0056] S105. After the computing node reads the target data in the target execution node, the computing node writes a copy of the target data in the executor node corresponding to the computing node.
[0057] Specifically, when the cloud data warehouse computing node completes data reading, the Master node will instruct the local Worker node (if any) to write a copy locally so that when there is a request to access the same data in the future, it can be read from the local memory.
[0058] It should be noted that the data reading speed in the case of a remote cache hit can reach the local network transmission speed. Since the network speed between Worker nodes is usually faster than the speed between Worker nodes and the underlying object storage, this application enables the cloud data warehouse computing node to read data from the remote Worker node storage first.
[0059] S106. If the search result indicates that the location information of the target execution node is not retrieved, the computing node reads the target data in the object storage.
[0060] That is to say, the target data is not in the space of all Worker nodes, that is, if the request misses the cache, the cloud data warehouse computing node will have to read the data from the underlying object storage.
[0061] Optionally, in another embodiment of the present application, if the search result indicates that the location information of the target execution node is not retrieved, the manager node entrusts the target node to read the target data in the object storage.
[0062] Among them, the target node has the highest priority to select the local Worker node of the computing node. When the cache misses, since the computing node of the cloud data warehouse must read the data from the underlying object storage system, it generally causes a large delay. Cache misses usually occur when reading data for the first time.
[0063] Optionally, in another embodiment of the present application, the manager node is responsible for servicing computing node requests in the cloud data warehouse and managing the global metadata of all executor nodes.
[0064] The global metadata of the Worker node includes, but is not limited to, file system metadata (such as index node tree), data block metadata (such as block location), Worker capacity metadata (free and used space), etc., which are not limited here.
[0065] Optionally, in another embodiment of the present application, the manager node is deployed on an independent physical machine.
[0066] Optionally, in another embodiment of the present application, the executor node is responsible for running the tasks assigned by the manager node and managing and allocating local resources, including but not limited to RAMDisk, SSD, HDD, etc., which are not limited here.
[0067] Optionally, in another embodiment of the present application, the configuration file of the executor node includes the address information of the manager node, and a registration application is made to the manager node after the executor node is started.
[0068] Optionally, in another embodiment of the present application, the manager node detects the running status of the executor node through a heartbeat mechanism.
[0069] like Figure 2As shown, it is a schematic diagram of the difference between the prior art and the present invention. In order to improve the operating efficiency of the computing cluster, in the prior art, the cloud data warehouse will use local memory or hard disk as the cache of the object storage service to save the user's hot data, that is, the cache in the computing cluster is a local cache, but the network transmission speed between the computing clusters is usually faster than the speed between the computing node and the underlying object storage. Therefore, in the actual application process, the query performance of the cloud data warehouse will be reduced. When the data that the cloud data warehouse computing node needs to read has been cached on the local Worker, it is a local cache hit. The computing node retrieves the Worker location where the data is stored from the Master. If the local Worker has the data, the computing node can read the file directly through the local file system. When the data requested by the cloud data warehouse computing node is not on the local Worker, but on a remote Worker in the computing cluster, the cloud data warehouse computing node will read the data from the remote Worker. When the cloud data warehouse computing node completes the data reading, the Master will instruct the local Worker (if it exists) to write a copy locally so that when there is a request to access the same data in the future, it can be read from the local memory. The data reading speed in the case of a remote cache hit can reach the local network transmission speed. Since the network speed between Workers is usually faster than the speed between Workers and the underlying object storage, this device enables the cloud data warehouse computing node to prioritize reading data from the remote Worker storage. If the requested data is not in all Worker spaces, that is, the request misses the cache, the cloud data warehouse computing node will have to read the data from the underlying object storage. The Master will delegate the read request to a Worker (preferentially the local Worker of the computing node) to read and cache the data from the underlying object storage. When the cache misses, since the cloud data warehouse computing node must read the data from the underlying object storage system, it generally causes a large delay. Cache misses usually occur when reading data for the first time.
[0070] As can be seen from the above scheme, the present application provides a method for reading data, which retrieves the location information of the target execution node in the manager node after the computing node receives the query request, and obtains the retrieval result; wherein, the location information of the target execution node is the location information of the executor node containing the target data; if the retrieval result shows that the location information of the target execution node is the location information of the executor node corresponding to the computing node, the computing node reads the target data through the local file system; if the retrieval result shows that the location information of the target execution node is not the location information of the executor node corresponding to the computing node, the computing node reads the target data in the target execution node; after the computing node reads the target data in the target execution node, a copy of the target data is written in the executor node corresponding to the computing node; if the retrieval result shows that the location information of the target execution node is not retrieved, the computing node reads the target data in the object storage. This solves the problem that the cloud data warehouse cannot cache and share, and effectively improves the query performance.
[0071] The present application embodiment provides a cloud data warehouse, such as Figure 3 As shown, specifically including:
[0072] N tenant spaces 10 and a manager node 20.
[0073] The tenant space includes M physical nodes 11; N and M are positive integers; each physical node 11 includes a computing node 12 and an executor node 13; the computing node 12 and the executor node 13 are deployed on the physical node 11 through affinity configuration.
[0074] In the specific implementation process of the present application, the computing node 12 is used to receive a query request; wherein the query request includes at least one target data; the target data is the data that needs to be read this time.
[0075] The computing node 12 is further used to retrieve the location information of the target execution node in the manager node 20 to obtain a search result; wherein the location information of the target execution node is the location information of the executor node containing the target data.
[0076] If the search result indicates that the location information of the target execution node is the location information 13 of the executor node corresponding to the computing node 12, the computing node 12 reads the target data through the local file system.
[0077] If the search result indicates that the location information of the target execution node is not the location information 13 of the executor node corresponding to the computing node 12, the computing node is also used to read the target data in the target execution node. After the computing node 12 reads the target data in the target execution node, a copy of the target data is written in the executor node 13 corresponding to the computing node 12.
[0078] If the search result indicates that the location information of the target execution node is not retrieved, the computing node 12 is further used to read the target data in the object storage.
[0079] For the specific working process of the units disclosed in the above embodiments of the present application, please refer to the corresponding method embodiments, such as Figure 1 As shown, no further description is given here.
[0080] Optionally, in another embodiment of the present application, if the search result indicates that the location information of the target execution node is not retrieved, the manager node entrusts the target node to read the target data in the object storage.
[0081] Among them, the target node has the highest priority to select the local Worker node of the computing node. When the cache misses, since the computing node of the cloud data warehouse must read the data from the underlying object storage system, it generally causes a large delay. Cache misses usually occur when reading data for the first time.
[0082] Optionally, in another embodiment of the present application, the manager node is responsible for servicing computing node requests in the cloud data warehouse and managing the global metadata of all executor nodes.
[0083] The global metadata of the Worker node includes, but is not limited to, file system metadata (such as index node tree), data block metadata (such as block location), Worker capacity metadata (free and used space), etc., which are not limited here.
[0084] Optionally, in another embodiment of the present application, the manager node is deployed on an independent physical machine.
[0085] Optionally, in another embodiment of the present application, the executor node is responsible for running the tasks assigned by the manager node and managing and allocating local resources, including but not limited to RAMDisk, SSD, HDD, etc., which are not limited here.
[0086] Optionally, in another embodiment of the present application, the configuration file of the executor node includes the address information of the manager node, and a registration application is made to the manager node after the executor node is started.
[0087] Optionally, in another embodiment of the present application, the manager node detects the running status of the executor node through a heartbeat mechanism.
[0088] It can be seen from the above scheme that the present application provides a cloud data warehouse, which includes N tenant spaces 10 and a manager node 20. Among them, the tenant space includes M physical nodes 11; N and M are positive integers; each physical node 11 includes a computing node 12 and an executor node 13; the computing node 12 and the executor node 13 are deployed on the physical node 11 through affinity configuration. The computing node 12 is used to receive query requests; wherein the query request includes at least one target data; the target data is the data that needs to be read this time. The computing node 12 is also used to retrieve the location information of the target execution node in the manager node 20 to obtain a retrieval result; wherein the location information of the target execution node is the location information of the executor node containing the target data. If the retrieval result shows that the location information of the target execution node is the location information 13 of the executor node corresponding to the computing node 12, the computing node 12 reads the target data through the local file system. If the search result shows that the location information of the target execution node is not the location information 13 of the executor node corresponding to the computing node 12, the computing node is also used to read the target data in the target execution node. After the computing node 12 reads the target data in the target execution node, a copy of the target data is written in the executor node 13 corresponding to the computing node 12. If the search result shows that the location information of the target execution node is not retrieved, the computing node 12 is also used to read the target data in the object storage. This solves the problem that the cloud data warehouse cannot cache and share, and effectively improves the query performance.
[0089] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.
[0090] Another embodiment of the present application provides a server, such as Figure 4 As shown, including:
[0091] One or more processors 401 .
[0092] The storage device 402 stores one or more programs.
[0093] When the one or more programs are executed by the one or more processors 401 , the one or more processors 401 implement the data reading method as described in any one of the above embodiments.
[0094] Another embodiment of the present application provides a computer storage medium on which a computer program is stored, wherein when the computer program is executed by a processor, the data reading method as described in any one of the above embodiments is implemented.
[0095] In the context of the present application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
[0096] It should be noted that the computer-readable medium mentioned above in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.
[0097] Another embodiment of the present application provides a computer program product. When the computer program product is executed, it is used to execute the above-mentioned data reading method.
[0098] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device, or installed from a ROM. When the computer program is executed by a processing device, the above-mentioned functions defined in the method of the embodiment of the present application are executed.
[0099] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined in this application is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing this application.
[0100] Although several specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present application. Certain features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination.
[0101] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of application involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above application concept. For example, the above features are replaced with (but not limited to) technical features with similar functions applied in the present application.
Claims
1. A cloud data warehouse, characterized in that: include: N tenant spaces and manager nodes; wherein the tenant space includes M physical nodes; N and M are positive integers; each of the physical nodes includes a computing node and an executor node; the computing node and the executor node are deployed on the physical node through affinity configuration; The computing node is used to receive a query request; wherein the query request includes at least one target data; the target data is the data that needs to be read this time; The computing node is further used to retrieve the location information of the target execution node in the manager node to obtain a search result; wherein the location information of the target execution node is the location information of the executor node containing the target data; If the search result indicates that the location information of the target execution node is the location information of the executor node corresponding to the computing node, the computing node reads the target data through the local file system; If the search result indicates that the location information of the target execution node is not the location information of the executor node corresponding to the computing node, the computing node is further used to read the target data in the target execution node, and after the computing node reads the target data in the target execution node, writes a copy of the target data in the executor node corresponding to the computing node; If the search result indicates that the location information of the target execution node is not retrieved, the computing node is further configured to read the target data in the object storage.
2. The cloud data warehouse according to claim 1, characterized in that: The manager node is responsible for servicing computing node requests in the cloud data warehouse and managing the global metadata of all executor nodes.
3. The cloud data warehouse according to claim 1, characterized in that: The manager node is deployed on an independent physical machine.
4. The cloud data warehouse according to claim 1, characterized in that: The executor node is responsible for running the tasks assigned by the manager node and managing and allocating local resources.
5. The cloud data warehouse according to claim 1, characterized in that: The configuration file of the executor node includes the address information of the manager node, and after the executor node is started, a registration application is made to the manager node.
6. The cloud data warehouse according to claim 1, characterized in that: The manager node detects the running status of the executor node through a heartbeat mechanism.
7. The cloud data warehouse according to claim 1, characterized in that: Also includes: If the search result indicates that the location information of the target execution node is not retrieved, the manager node entrusts the target node to read the target data in the object storage.
8. A method for reading data, characterized in that: Applied to a cloud data warehouse, the cloud data warehouse includes N tenant spaces and a manager node; wherein the tenant space includes M physical nodes; N and M are positive integers; each of the physical nodes includes a computing node and an executor node; the computing node and the executor node are deployed on the physical node through affinity configuration, and the data reading method includes: The computing node receives a query request; wherein the query request includes at least one target data; the target data is the data that needs to be read this time; The computing node retrieves the location information of the target execution node in the manager node to obtain a search result; wherein the location information of the target execution node is the location information of the executor node containing the target data; If the search result indicates that the location information of the target execution node is the location information of the executor node corresponding to the computing node, the computing node reads the target data through the local file system; If the search result indicates that the location information of the target execution node is not the location information of the executor node corresponding to the computing node, the computing node reads the target data in the target execution node; After the computing node reads the target data in the target execution node, the computing node writes a copy of the target data in the executor node corresponding to the computing node; If the search result indicates that the location information of the target execution node is not retrieved, the computing node reads the target data in the object storage.
9. A server, characterized in that: include: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the data reading method as claimed in claim 8 .
10. A computer storage medium, characterized in that: A computer program is stored thereon, wherein when the computer program is executed by a processor, the data reading method as claimed in claim 8 is implemented.