CN115277579A - Warehouse video calling method and cloud platform - Google Patents

Abstract

The application provides a warehouse video calling method and a cloud platform, wherein the method comprises the following steps: the cluster server acquires a video calling task and a task priority of the video calling task, and adds the video calling task to a task queue according to the task priority; for each execution server, when the execution server is in an idle state, the execution server performs task preemption operation on the cluster server so as to preempt the execution permission of the video retrieval task corresponding to the highest priority in the task queue; the cluster server allocates the video calling task corresponding to the highest priority to an execution server with a task preemption success; and for each execution server, under the condition that the execution server is distributed with the video calling task, executing the distributed video calling task to download the warehouse video to be called from the equipment platform of the warehouse monitoring equipment. By adopting the method and the device, the average response time of the video frequency acquisition task can be shortened.

Description

Warehouse Video Retrieval Method and Cloud Platform

technical field

This application relates to the technical field of warehouse management, in particular to a warehouse video call method and a cloud platform.

Background technique

In e-commerce sales, in order to clarify the status of the goods in the delivery process and provide after-sales service to customers, e-commerce sales companies will install multiple warehouse monitoring devices in the warehouse and on the workbench to take pictures of the warehouse and work respectively. The platform enables the warehouse video to record the product status of the product in the delivery process.

When providing after-sales service or backtracking on the shipment of goods, the staff can use the video call system to call the warehouse video from the equipment platform of the warehouse monitoring equipment to determine the product status of the product in the shipping link. However, research by the inventors has found that the prior art has a problem of too long response time when retrieving the warehouse video.

Contents of the invention

The purpose of this application is to at least solve one of the above-mentioned technical defects, especially the technical defect of too long response time in the prior art.

In the first aspect, the embodiment of the present application provides a warehouse video calling method, the method is applied to a cloud platform, and the cloud platform includes a cluster server and a plurality of execution servers, and the method includes:

The cluster server obtains the video retrieval task and the task priority of the video retrieval task, and adds the video retrieval task to the task queue according to the task priority;

For each execution server, when the execution server is in an idle state, the execution server performs a task preemption operation to the cluster server, so as to preempt the execution authority of the video call task corresponding to the highest priority in the task queue;

The cluster server assigns the video call task corresponding to the highest priority to the execution server that successfully preempts the task;

For each execution server, if assigned a video retrieval task, the execution server executes the assigned video retrieval task, so as to download the warehouse video to be retrieved from the equipment platform of the warehouse monitoring device.

In one of the embodiments, for each execution server, if the execution server is assigned a video retrieval task, it executes the assigned video retrieval task, so as to download the video retrieval task from the equipment platform of the warehouse monitoring device. The steps to call the warehouse video include:

For each execution server, the execution server obtains an effective network connection from the SDK connection pool configured by the execution server under the condition that the execution server is allocated with a video call task, and sends the monitoring device to the warehouse through the effective network connection. The device platform sends a video download request, so that the device platform returns the warehouse video to be called to the execution server;

Wherein, the SDK connection pool is used to store the information of the network connection between the cloud platform and the device platform.

In one of the embodiments, the cloud platform also includes a cloud storage object server, and the method also includes:

For each execution server, after downloading the warehouse video to be called, the execution server transcodes the warehouse video to be called to obtain the warehouse video in H264 format, and uploads the warehouse video in H264 format to the cloud storage object server.

In one of the embodiments, after the execution server uploads the warehouse video in H264 format to the cloud storage object server, it also includes:

The cluster server assigns a temporary uniform resource locator to the playback device, so that the playback device obtains the warehouse video to be played from the cloud storage object server based on the temporary uniform resource locator.

In one of the embodiments, the step of the cluster server acquiring the video calling task includes:

The cluster server receives the call request sent by the user, and the call request includes the product delivery number and the link identification of the delivery link corresponding to the warehouse video to be called;

The cluster server obtains the commodity delivery log, and queries the commodity delivery log according to the commodity delivery number and the link identification, so as to determine the equipment identification of the warehouse monitoring device used to shoot the warehouse video to be retrieved and the shooting period corresponding to the warehouse video to be called;

The cluster server generates the video retrieval task according to the device identifier and the shooting period.

In one of the embodiments, the step of obtaining the task priority of the video call task by the cluster server includes:

The cluster server determines a target user identity, where the target user identity is the user identity corresponding to the user who sent the calling request;

The cluster server determines the target task priority corresponding to the target user ID from the preset mapping relationship, and uses the target task priority as the task priority of the video call task, and the preset mapping The relationship is the mapping relationship between user identity and task priority.

In one of the embodiments, the method further includes: the cluster server writes the video calling task into a task database.

In the second aspect, the embodiment of the present application provides a cloud platform, the cloud platform includes a cluster server and a plurality of execution servers, wherein:

The cluster server is used to obtain the video retrieval task and the task priority of the video retrieval task, and add the video retrieval task to the task queue according to the task priority;

For each execution server, when the execution server is in an idle state, the execution server is used to perform a task preemption operation to the cluster server, so as to preempt the execution authority of the video call task corresponding to the highest priority in the task queue;

The cluster server is used to assign the video call task corresponding to the highest priority to the execution server whose task preemption is successful;

For each execution server, the execution server is configured to execute the assigned video retrieval task in the case of being assigned a video retrieval task, so as to download the warehouse video to be retrieved from the equipment platform of the warehouse monitoring device.

In one of the embodiments, the cloud platform further includes a cloud storage object server. For each execution server, the execution server is also used to transcode the warehouse video to be called to obtain the warehouse video in H264 format after downloading the warehouse video to be called, and convert the warehouse video in H264 format The warehouse video is uploaded to the cloud storage object server.

In one of the embodiments, the cluster server is further configured to write the video retrieval task into a task database.

In the method for invoking warehouse video and the cloud platform in the embodiments of the present application, the cloud platform may include a cluster server and multiple execution servers. Wherein, the cluster server can acquire the video retrieval task and its corresponding task priority, and add the acquired video retrieval task to the task queue according to the task priority. Each execution server in an idle state can perform a task preemption operation to the cluster server, so as to preempt the execution right of the video calling task corresponding to the highest priority in the task queue. The cluster server can assign the video retrieval task corresponding to the highest priority to the successfully preempted execution server, so as to download the warehouse video to be retrieved from the equipment platform of the warehouse monitoring device through the server resources of the successfully preempted execution server. In this way, this application can preempt each video retrieval task in the task queue according to the server dimension, and execute different video retrieval tasks through multiple execution servers, so as to realize the concurrent parallel execution of multiple video retrieval tasks, so that Accelerate the response speed of video retrieval tasks and shorten the average response time of video retrieval tasks, thereby improving user experience.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present application. Those skilled in the art can also obtain other drawings based on these drawings without any creative effort.

Fig. 1 is a schematic structural block diagram of a cloud platform in an embodiment;

Fig. 2 is a schematic flow diagram of a warehouse video calling method in an embodiment;

Fig. 3 is a schematic flow diagram of the cluster server acquiring video call task steps in one embodiment;

Fig. 4 is a schematic flowchart of the steps of the cluster server acquiring the task priority of the video retrieval task in an embodiment.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.

As mentioned in the background art, the prior art has the problem of too long response time when calling warehouse videos. According to the inventor's research, it is found that the reason for this problem is that the video retrieval system in the prior art is realized by a single machine. In other words, the existing technology uses a single server to execute all video retrieval tasks, so it does not support simultaneous execution of multiple tasks, poor concurrency performance, and poor scalability.

At the same time, when performing a single video retrieval task, the server needs to use the SDK (Software Development Kit, Software Development Kit) provided by the equipment manufacturer of the warehouse monitoring equipment to build a network connection between the server and the equipment platform of the warehouse monitoring equipment. During the construction process, steps such as initialization, login, download, release of links, and release of consumed resources are required. After the network connection is established, the server downloads the warehouse video to be retrieved from the equipment platform of the warehouse monitoring device through the network connection, and transcodes the warehouse video to be retrieved. It can be seen that the server needs to consume a large amount of CPU (Central Processing Unit, central processing unit), memory and IO (Input/Output, input/output) resources when performing a single video retrieval task, and the execution time of a single video retrieval task Long, which further increases the average response time of each video retrieval task.

In order to solve the foregoing problems, the embodiment of the present application provides a warehouse video retrieval method and a cloud platform, which preempts each video retrieval task in the task queue according to the server dimension, and executes different video retrieval tasks through multiple execution servers Tasks to implement concurrent and parallel execution of multiple video retrieval tasks, thereby speeding up the response speed of video retrieval tasks, shortening the average response time of video retrieval tasks, and improving user experience.

In one embodiment, the present application provides a cloud platform. As shown in FIG. 1 , the cloud platform includes a cluster server 102 and multiple execution servers 104. It can be understood that the specific number of execution servers 104 can be determined according to actual conditions, and this application does not specifically limit it.

Wherein, the cluster server 102 refers to a server with task management and task allocation functions, which may be a Redis cluster server, or other servers with Redis cluster characteristics, which is not specifically limited in this application. The execution server 104 refers to a server that can be used to execute video retrieval tasks, and one or more threads can run therein, and the execution server 104 can perform task preemption operations through the threads running on it. The cluster server 102 can be used to manage the threads on each execution server 104, and assign tasks accordingly.

In an embodiment, the cloud platform may further include a cloud storage object server, such as an OSS (Operation Support Systems, operation support system) cloud storage object server. The cloud storage object server can be used to store the warehouse video to be retrieved obtained by each execution server 104 after executing the video retrieval task, so as to realize massive video storage.

For the cluster server 102, the execution server 104, and the cloud storage object server involved in the above-mentioned embodiments, the execution steps of each server can be as described in any embodiment of the following warehouse video retrieval method, and the above-mentioned embodiments can be combined with the following Any embodiment of the warehouse video calling method should be understood.

In one embodiment, the present application provides a warehouse video calling method. As shown in Figure 2, the method specifically includes the following steps:

S202. The cluster server acquires a video retrieval task and a task priority of the video retrieval task, and adds the video retrieval task to a task queue according to the task priority.

Wherein, the video retrieval task refers to a task for retrieving the warehouse video of a corresponding time period from the equipment platform of the warehouse monitoring equipment. In one of the embodiments, the video retrieval task may include the equipment identification of the warehouse monitoring equipment used to shoot the warehouse video to be retrieved, and the shooting period corresponding to the warehouse video to be retrieved, so that the execution server can retrieve the task according to the video. The task information is retrieved via video, and then the task of video retrieval is completed.

The video retrieval task may correspond to a task priority, and the task priority may be used to reflect the task execution order of the video retrieval task. The task queue can be used to maintain video retrieval tasks. When the task queue includes multiple video retrieval tasks, each video retrieval task is sorted according to the task priority, for example, it can be sorted according to the priority from high to low, or according to Sort in order of priority from low to high.

When the cluster server obtains the video retrieval task, it can add the video retrieval task to the corresponding position of the task queue according to the task priority of the video retrieval task, so that in the added task queue, each video retrieval task Tasks can be sorted according to a preset priority order.

In one of the embodiments, the method further includes: the cluster server writes the video calling task into a task database. In other words, when the video retrieval task is obtained, the cluster server can perform double writing, so as to respectively add the video retrieval task to the task queue and write the video retrieval task into the task database. In this way, each video call task can be recorded through the task database, so as to facilitate task traceability in the later stage.

S204. For each execution server, when the execution server is in an idle state, the execution server performs a task preemption operation to the cluster server, so as to preempt the execution right of the video calling task corresponding to the highest priority in the task queue.

Wherein, the idle state refers to a state in which the execution server has enough idle resources to execute the video retrieval task, for example, it may be a state in which the execution server does not execute the task. When the task queue includes at least one video retrieval task, for the video retrieval task corresponding to the highest priority in the task queue, each execution server in the idle state in the cloud platform will perform a task preemption operation on the video retrieval task, to try to obtain the execution permission of the video calling task.

S206. The cluster server assigns the video retrieval task corresponding to the highest priority to the execution server whose task preemption succeeds.

In one of the embodiments, after allocating the video retrieval task corresponding to the highest priority, the cluster server can delete the allocated video retrieval task from the task queue, so as to update the task queue in real time according to the task allocation situation, In order for the execution server of the cloud platform to continue preempting the next video retrieval task according to step S204, until no video retrieval task is included in the task queue.

S208. For each execution server, if the execution server is assigned a video retrieval task, execute the assigned video retrieval task, so as to download the warehouse video to be retrieved from the equipment platform of the warehouse monitoring device.

Specifically, when an execution server is assigned the execution authority of the video retrieval task, the execution server can execute the assigned video retrieval task through its own resources, so as to obtain the warehouse to be retrieved from the equipment platform of the warehouse monitoring equipment. video.

The embodiment of the present application can preempt each video retrieval task in the task queue according to the server dimension, and execute different video retrieval tasks through multiple execution servers, so as to realize concurrent and parallel execution of multiple video retrieval tasks, so that Accelerate the response speed of video retrieval tasks and shorten the average response time of video retrieval tasks, thereby improving user experience. In addition, adopting the solution of this application can solve the problem of low release efficiency, improve self-operation and maintenance capabilities, and improve performance.

In one embodiment, for each execution server, the execution server executes the assigned video retrieval task in the case of being assigned a video retrieval task, so as to download the video retrieval task from the equipment platform of the warehouse monitoring device. The steps to take the warehouse video include:

For each execution server, the execution server obtains an effective network connection from the SDK connection pool configured by the execution server under the condition that the execution server is allocated with a video call task, and sends the monitoring device to the warehouse through the effective network connection. The device platform sends a video download request, so that the device platform returns the warehouse video to be called to the execution server;

Wherein, the SDK connection pool is used to store the information of the network connection between the cloud platform and the device platform.

Specifically, if the execution server directly uses the SDK provided by the equipment manufacturer to build the network connection between it and the equipment platform of the warehouse monitoring equipment, then in the construction process, initialization, login, download, release link and release It takes a long time to consume resources and other steps. In addition, since the SDK provided by the equipment manufacturer is implemented based on C++, and the execution server is implemented based on Java, the incompatibility between C++ and Java will further increase the time-consuming construction of network connections, and further increase the task of video retrieval average response time.

In order to further speed up the response speed of video retrieval tasks and shorten the average response time of video retrieval tasks, this application can maintain the network connection between the cloud platform and the device platform through the SDK connection pool, and realize the multiplexing of network connections. When the execution server executes the video call task, it can directly obtain a valid network connection from the SDK connection pool, without the aforementioned steps of initialization, login, release link release and resource consumption, so as to avoid repeated login and improve the parallelism of task execution This can further speed up the response speed of the video retrieval task and shorten the average response time of the video retrieval task.

Specifically, the SDK connection pool can be used to store network connection information between the cloud platform and the device platform, and is configured in each execution server. When the execution server is assigned a video call task, it can obtain a valid network connection from the SDK connection pool, and send a video download request to the device platform of the warehouse monitoring device through the obtained valid network connection, so that the device platform can The video download request was received. When the device platform receives a video download request, it can respond to the request and return the corresponding warehouse video to the execution server that sent the request, and the corresponding warehouse video is the warehouse video to be called.

In one embodiment, the cloud platform further includes a cloud storage object server, and the method further includes: for each execution server, after downloading the video of the warehouse to be called, the execution server The video is transcoded to obtain the warehouse video in H264 format, and the warehouse video in H264 format is uploaded to the cloud storage object server.

The warehouse video to be retrieved returned by the device platform is usually in the H265 format with a high compression ratio. However, most of the current mainstream players and video playback plug-ins do not support the H265 format, and cannot directly play videos in this format. Therefore, in order to make the warehouse video to be retrieved playable on the playback device, after the execution server downloads the warehouse video to be retrieved in a high compression ratio format from the device platform, it needs to transcode the video to obtain the warehouse in H264 format video. After obtaining the warehouse video in H264 format, the execution server can upload the warehouse video in H264 format to the cloud storage object server, so as to realize massive video storage through the cloud storage object server.

In one embodiment, after the execution server uploads the warehouse video in H264 format to the cloud storage object server, it further includes: the cluster server assigns a temporary uniform resource locator to the playback device, so that The playback device acquires the warehouse video to be played from the cloud storage object server based on the temporary uniform resource locator.

Specifically, after the execution server uploads the warehouse video in H264 format to the cloud storage object server, the playback device can obtain the corresponding warehouse video from the cloud storage object server and play it. At this point, the cluster server can assign a temporary URL (Uniform Resource Locator, Uniform Resource Locator) to the playback device, so that the playback device can pull streaming data from the cloud storage object server through the temporary URL to buffer the warehouse video to be played. In this way, the video buffering time of the playback device can be shortened, and the video processing speed can be further improved. In one of the embodiments, the playback device can play the warehouse video through its configured H5 video playback plug-in.

In one embodiment, as shown in Figure 3, the step of the cluster server acquiring the video calling task includes:

S302. The cluster server receives the retrieval request sent by the user, and the retrieval request includes the commodity dispatch number and the link identifier of the dispatch link corresponding to the warehouse video to be retrieved;

S304. The cluster server acquires the commodity delivery log, and queries the commodity delivery log according to the commodity delivery number and the link identifier, so as to determine the location of the warehouse monitoring device used to shoot the warehouse video to be retrieved. Equipment identification and the shooting period corresponding to the warehouse video to be called;

S306. The cluster server generates the video calling task according to the device identifier and the shooting period.

Specifically, when a user has a demand for warehouse video retrieval, he or she can send a retrieval demand to the cluster server, which may include the product shipment number and the warehouse video corresponding to the warehouse video that needs to be retrieved. Link identification, so that the cluster server can determine the video information of the warehouse video to be retrieved accordingly. Wherein, the product delivery number may be a unique number of the product in the delivery process, for example, it may be an order number or a waybill number. Since the product needs to go through multiple links (such as picking, packing, and conveying) during the shipping process, it takes a relatively long time, and in some cases, the user only wants to make sure that the product is in the There is no need to obtain the complete shipping video for the status of a certain link or several links. Therefore, the call request may include the link identifier of the outbound link corresponding to the warehouse video to be called, so that the warehouse video segment corresponding to the link identifier is called, such as the warehouse video corresponding to the packing link or the picking link. Corresponding warehouse video.

The cluster server can obtain the product delivery log, which records the information corresponding to each product in each delivery link, for example, records the product delivery number, the identification of each link, and the product status corresponding to the product delivery number. Corresponding to the time period of the shipping link. After the cluster server obtains the retrieval requirement, it can extract the product delivery number and link identifier from the retrieval requirement, and use the commodity delivery number and link identifier as the log query keywords to query the pending delivery log from the commodity delivery log. Call the equipment identification and shooting period corresponding to the warehouse video, and generate a video retrieval task according to the equipment identification and shooting period.

In one embodiment, the cluster server can determine the target workbench ID according to the product delivery number and link ID, and determine the equipment ID according to the target workbench ID. Wherein, the target workbench identifier is the identifier of the workbench where the product is located in the outbound link corresponding to the warehouse video to be retrieved.

In this embodiment, the cluster server can automatically determine the equipment identification of the warehouse monitoring device to be retrieved from the warehouse video and the shooting period corresponding to the warehouse video to be retrieved based on the retrieval request sent by the user, and generate a video retrieval task accordingly. The user does not need to manually determine detailed information such as device identification and shooting time, so that the response speed of video retrieval tasks can be further accelerated, and the average response time of video retrieval tasks can be shortened.

In one embodiment, as shown in FIG. 4, the step for the cluster server to obtain the task priority of the video retrieval task includes:

S402. The cluster server determines a target user identity, where the target user identity is the user identity corresponding to the user who sent the calling request;

S404. The cluster server determines the target task priority corresponding to the target user identity from the preset mapping relationship, and uses the target task priority as the task priority of the video call task, and the preset Let the mapping relationship be the mapping relationship between user identity and task priority.

Specifically, the cluster server may pre-store the mapping relationship between different user identities and their corresponding task priorities. When obtaining the call request sent by the user, the cluster server can determine the task priority corresponding to the user ID from the pre-stored mapping relationship according to the user ID corresponding to the user who sent the call request, and use This is the task priority of the video retrieval task. For example, when the identity of the target user is the identification of customer service personnel, the video retrieval task may correspond to the highest priority; when the identity of the target user is the identity of the quality control personnel, the video retrieval task may correspond to the second highest priority; When the identity of the target user is identified by other relevant personnel, the video retrieval task may correspond to the lowest priority. In this way, the task priority can be determined according to the user identity corresponding to the user who sent the calling request, so that the user with high response requirements can be given priority to respond, thereby further improving the user experience.

Finally, it should also be noted that in this text, relational terms such as first and second etc. are only used to distinguish one entity or operation from another, and do not necessarily require or imply that these entities or operations, any such actual relationship or order exists. Furthermore, the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article or apparatus comprising a set of elements includes not only those elements, but also includes elements not expressly listed. other elements of or also include elements inherent in such a process, method, article, or device. Without further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article or apparatus comprising said element. As used herein, "a", "an", "the", "the" and "it" may also include plural forms unless the context clearly dictates otherwise. A plurality refers to at least two situations, such as 2, 3, 5 or 8 and so on. "And/or" includes any and all combinations of the associated listed items.

Each embodiment in this specification is described in a progressive manner. Each embodiment focuses on the difference from other embodiments. The various embodiments can be combined as needed, and the same and similar parts can be referred to each other. .

The above description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the application. Therefore, the present application will not be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (10)

1. The warehouse video calling method is applied to a cloud platform, the cloud platform comprises a cluster server and a plurality of execution servers, and the method comprises the following steps:

the cluster server acquires a video calling task and a task priority of the video calling task, and adds the video calling task to a task queue according to the task priority;

for each execution server, when the execution server is in an idle state, the execution server performs task preemption operation on the cluster server so as to preempt the execution permission of the video retrieval task corresponding to the highest priority in the task queue;

the cluster server allocates the video calling task corresponding to the highest priority to an execution server with a task preemption success;

and for each execution server, under the condition that the execution server is distributed with the video calling task, executing the distributed video calling task to download the warehouse video to be called from the equipment platform of the warehouse monitoring equipment.

2. The warehouse video retrieval method according to claim 1, wherein the step of executing, for each execution server, the video retrieval task allocated to the execution server to download the warehouse video to be retrieved from the device platform of the warehouse monitoring device comprises:

for each execution server, under the condition that the execution server is allocated with a video calling task, acquiring an effective network connection from an SDK connection pool configured by the execution server, and sending a video downloading request to an equipment platform of the warehouse monitoring equipment through the effective network connection so as to enable the equipment platform to return the warehouse video to be called to the execution server;

the SDK connection pool is used for storing information of network connection between the cloud platform and the equipment platform.

3. The warehouse video retrieval method of claim 1, wherein the cloud platform further comprises a cloud storage object server, the method further comprising:

for each execution server, after downloading the warehouse video to be called, the execution server transcodes the warehouse video to be called to obtain the warehouse video in the H264 format, and uploads the warehouse video in the H264 format to the cloud storage object server.

4. The warehouse video retrieval method according to claim 3, further comprising, after the step of the executive server uploading the warehouse video in H264 format to the cloud storage object server:

the cluster server allocates a temporary uniform resource locator to the playing device, so that the playing device obtains the warehouse video to be played from the cloud storage object server based on the temporary uniform resource locator.

5. The warehouse video retrieval method of claim 1, wherein the step of the cluster server obtaining the video retrieval task comprises:

the cluster server receives a calling demand sent by a user, wherein the calling demand comprises a commodity ex-warehouse number and a link identification of an ex-warehouse link corresponding to a to-be-called warehouse video;

the cluster server acquires a commodity ex-warehouse log, and inquires the commodity ex-warehouse log according to the commodity ex-warehouse serial number and the link identification so as to determine an equipment identification of warehouse monitoring equipment for shooting the warehouse video to be called and a shooting time period corresponding to the warehouse video to be called;

and the cluster server generates the video calling task according to the equipment identification and the shooting time interval.

6. The warehouse video retrieval method according to claim 5, wherein the step of the cluster server obtaining the task priority of the video retrieval task includes:

the cluster server determines a target user identity, wherein the target user identity is a user identity corresponding to the user sending the calling requirement;

the cluster server determines a target task priority corresponding to the target user identity from a preset mapping relation, the target task priority is used as a task priority of the video calling task, and the preset mapping relation is the mapping relation between the user identity and the task priority.

7. The warehouse video retrieval method of any of claims 1 to 6, wherein the method further comprises:

and the cluster server writes the video calling task into a task database.

8. The cloud platform is characterized by comprising a cluster server and a plurality of execution servers, wherein:

the cluster server is used for acquiring a video calling task and the task priority of the video calling task, and adding the video calling task to a task queue according to the task priority;

aiming at each execution server, when the execution server is in an idle state, the execution server is used for performing task preemption operation on the cluster server so as to preempt the execution permission of the video calling task corresponding to the highest priority in the task queue;

the cluster server is used for distributing the video calling task corresponding to the highest priority to the execution server which is successful in task preemption;

and for each execution server, the execution server is used for executing the distributed video calling task under the condition that the video calling task is distributed, so as to download the warehouse video to be called from the equipment platform of the warehouse monitoring equipment.

9. The cloud platform of claim 8, wherein the cloud platform further comprises a cloud storage object server;

and for each execution server, the execution server is further configured to, after downloading the warehouse video to be called, transcode the warehouse video to be called to obtain a warehouse video in an H264 format, and upload the warehouse video in the H264 format to the cloud storage object server.

10. The cloud platform of claim 8, wherein the cluster server is further configured to write the video retrieval task to a task database.

Images