CN113590415B - Port management system, method, equipment and medium of deep learning training platform - Google Patents
Port management system, method, equipment and medium of deep learning training platform Download PDFInfo
- Publication number
- CN113590415B CN113590415B CN202110744300.0A CN202110744300A CN113590415B CN 113590415 B CN113590415 B CN 113590415B CN 202110744300 A CN202110744300 A CN 202110744300A CN 113590415 B CN113590415 B CN 113590415B
- Authority
- CN
- China
- Prior art keywords
- port
- deep learning
- range
- pool
- learning training
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3041—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is an input/output interface
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Quality & Reliability (AREA)
- Computer And Data Communications (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The application discloses a port management system, a method, electronic equipment and a computer readable storage medium of a deep learning training platform, wherein the system comprises: the port Chi Chushi initialization module is used for initializing available ports into a port pool according to configuration information of a port range when the deep learning training platform is started; the port management module is used for responding to the port range modification instruction after the deep learning training platform operates, and dynamically adjusting the port range in the port pool; responding to a port acquisition instruction, and selecting a target port from a port pool for the current task; and the monitoring module is used for creating an event monitor after the deep learning training platform operates so as to acquire the information of the deleted task in real time and release the port occupied by the deleted task. The application can help to autonomously manage the port range and dynamically manage the use of the port in the deep learning platform, and greatly improves the training efficiency during deep learning training.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a port management system and method for a deep learning training platform, an electronic device, and a computer readable storage medium.
Background
Currently, artificial intelligence techniques typified by deep learning have been rapidly developed, and these techniques are being applied to various industries. With the wide application of deep learning, many fields generate a great deal of strong, efficient and convenient demands for training artificial intelligent models, and the training depends on a deep learning training platform.
The current industry does not have a better solution to the port management function of the deep learning platform, most of the current industry is a port range used by the port segment control task of the NodePort based on kubernetes, the port range is exposed to the outside by an operating system, the mode is relatively rough, on one hand, the cost of modifying the NodePort based on kubernetes once is relatively high, the port range is generally set larger, the subsequent mode is not modified, the mode is very inflexible, on the other hand, the port range is generally synchronous with the firewall rule, if the port range is too large, the safety of the operating system is not guaranteed, and the mode is unsafe from the aspect of system safety.
In view of this, it has been a great need for a person skilled in the art to provide a solution to the above-mentioned technical problems.
Disclosure of Invention
The application aims to provide a port management system, a port management method, electronic equipment and a computer readable storage medium of a deep learning training platform, so that autonomous management and dynamic management of the deep learning training platform on the port can be effectively realized, and the training efficiency of the platform can be improved.
In order to solve the technical problems, in one aspect, the application discloses a port management system of a deep learning training platform, comprising:
the port Chi Chushi initialization module is used for initializing an available port into a port pool according to configuration information of a port range when the deep learning training platform is started;
the port management module is used for responding to a port range modification instruction after the deep learning training platform operates, and dynamically adjusting the port range in the port pool; responding to a port acquisition instruction, and selecting a target port from the port pool for the current task to use;
and the monitoring module is used for creating an event monitor after the deep learning training platform operates so as to acquire the information of the deleted task in real time and release the port occupied by the deleted task.
Optionally, the port Chi Chushi module is further configured to:
and checking ports in the current port pool, and deleting invalid ports.
Optionally, the port management module is specifically configured to, when dynamically adjusting a port range in the port pool:
performing range verification on the new port range specified in the port range modification instruction;
and if the new port range is a subset of the parent set of the old port range and the kubernetes nodePort port range, dynamically adjusting the port range in the port pool according to the new port range.
Optionally, after dynamically adjusting the port range in the port pool, the port management module is further configured to:
and synchronously modifying the updated open port range to firewall rules of all hosts in the cluster by executing the shell script.
Optionally, the port management module is further configured to, after selecting a target port from the port pool for use by a current task:
and storing the corresponding relation between the target port and the current task, and modifying the state data of the target port in the port pool into an used state.
Optionally, the listening module is further configured to:
and deleting the event monitor after the deep learning training platform stops running.
In yet another aspect, the present application further discloses a port management method of the deep learning training platform, which is characterized in that the method includes:
when the deep learning training platform is started, initializing an available port into a port pool based on configuration information of a port range by a port pool initialization module;
after the deep learning training platform operates, dynamically adjusting a port range in the port pool based on a port management module responding to a port range modification instruction;
selecting a target port from the port pool for use by a current task based on the port management module responding to a port acquisition instruction;
and creating an event monitor based on the monitoring module to acquire the information of the deleted task in real time and releasing the port occupied by the deleted task.
Optionally, after the port-based pool initialization module initializes the available ports into the port pool according to the configuration information of the port range, the method further includes:
and checking ports in the current port pool, and deleting invalid ports.
Optionally, the port-based management module dynamically adjusts the port range in the port pool in response to a port range modification instruction, specifically including:
performing range verification on the new port range specified in the port range modification instruction;
and if the new port range is a subset of the parent set of the old port range and the kubernetes nodePort port range, dynamically adjusting the port range in the port pool according to the new port range.
Optionally, after the port-based management module dynamically adjusts the port range in the port pool in response to the port range modification instruction, the method further includes:
and synchronously modifying the updated open port range to firewall rules of all hosts in the cluster by executing the shell script.
Optionally, after the selecting, based on the port management module in response to a port acquisition instruction, a target port from the port pool for use by a current task, the method further includes:
and storing the corresponding relation between the target port and the current task, and modifying the state data of the target port in the port pool into an used state.
Optionally, the method further comprises:
and deleting the event monitor based on the monitoring module after the deep learning training platform stops running.
In yet another aspect, the present application also discloses an electronic device, including:
a memory for storing a computer program;
a processor for executing the computer program to implement the steps of any of the port management methods of the deep learning training platform described above.
In yet another aspect, the present application also discloses a computer readable storage medium having stored therein a computer program which when executed by a processor is configured to implement the steps of the port management method of any of the deep learning training platforms described above.
The port management system, the method, the electronic equipment and the computer readable storage medium of the deep learning training platform provided by the application have the beneficial effects that: the application can effectively realize port management matters such as port use, port release, k8s service port segment adaptation and the like, help autonomously manage the port range in the deep learning platform and dynamically manage the use of the port, and greatly improve the training efficiency in deep learning training.
Drawings
In order to more clearly illustrate the technical solutions in the prior art and the embodiments of the present application, the following will briefly describe the drawings that need to be used in the description of the prior art and the embodiments of the present application. Of course, the following drawings related to embodiments of the present application are only a part of embodiments of the present application, and it will be obvious to those skilled in the art that other drawings can be obtained from the provided drawings without any inventive effort, and the obtained other drawings also fall within the scope of the present application.
Fig. 1 is a block diagram of a port management system of a deep learning training platform according to an embodiment of the present application;
FIG. 2 is a schematic diagram of port initialization according to an embodiment of the present application;
FIG. 3 is a schematic diagram of a port range according to an embodiment of the present application;
FIG. 4 is a flowchart of a method for managing ports of a deep learning training platform according to an embodiment of the present application;
fig. 5 is a block diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The application aims at providing a port management system, a port management method, electronic equipment and a computer readable storage medium of a deep learning training platform so as to effectively realize autonomous management and dynamic management of the deep learning training platform on the port and improve the training efficiency of the platform.
In order to more clearly and completely describe the technical solutions in the embodiments of the present application, the technical solutions in the embodiments of the present application will be described below with reference to the accompanying drawings in the embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
At present, the industry does not have a better solution to the port management of the deep learning platform, most of the solutions are port ranges used by the port segment control task of the NodePort based on kubernetes, the mode of the operating system is relatively rough to the external exposed port range, on one hand, the cost of modifying the NodePort of kubernetes once is relatively high, so that the port range is generally set relatively large, the modification is not carried out later, and the mode is very inflexible.
The Kubernetes is abbreviated as k8s, is an open-source application for managing containerization on a plurality of hosts in a cloud platform, and provides a mechanism for deploying, planning, updating and maintaining the application in order to enable the application deploying containerization to be simple and efficient. Kubernetes NodePort is one way of external traffic accessing the service entries in the k8s cluster (another way is LoadBalancer), nodeIP, nodePort is the entry provided to external traffic accessing the service in the k8s cluster.
On the other hand, this port range is generally synchronous with firewall rules, and if the port range is too large, there is no guarantee for the security of the operating system, and this way is also unsafe from the aspect of system security. In view of the above, the present application provides a port management scheme for a deep learning training platform, which can effectively solve the above-mentioned problems.
Referring to fig. 1, the embodiment of the application discloses a port management system of a deep learning training platform, which mainly comprises:
the port Chi Chushi initialization module 101 is configured to initialize an available port into a port pool according to configuration information of a port range when the deep learning training platform is started;
the port management module 102 is configured to dynamically adjust a port range in the port pool in response to a port range modification instruction after the deep learning training platform is operated; responding to a port acquisition instruction, and selecting a target port from a port pool for the current task;
and the monitoring module 103 is used for creating an event monitor after the deep learning training platform runs so as to acquire the information of the deleted task in real time and release the port occupied by the deleted task.
The deep learning platform can be an AIstation, provides intelligent AI containerization deployment and more efficient distributed training, is an artificial intelligent development resource platform oriented to artificial intelligent enterprise training scenes, can realize containerization deployment, visual development, centralized management and the like, provides extremely high-performance AI computing resources for users, realizes efficient computational power support, accurate resource management and scheduling, agile data integration and acceleration, and flow AI scene and service integration, effectively opens up development environments, computing resources and data resources, and improves development efficiency.
The user can create different deep learning framework environments based on the AIstation platform, can freely develop a model, debug the model in a command line mode, and quickly submit the model to the training platform through the development platform so as to develop and train an integrated solution. In the model debugging and training stage, the AIstation can enable the user to use the functions of Jupyter, SSH and the like to perform script debugging and the like by creating related tasks and exposing related ports. These rely on the port management mechanism of the deep learning platform, through which port channels between users, tasks, operating systems, containers can be opened.
Therefore, the application provides a port management system of a deep learning training platform, which comprises a port Chi Chushi module 101, a port management module 102 and a monitoring module 103, and can effectively realize autonomous dynamic management of the port and meet the port use requirement of the deep learning training platform.
Specifically, when the deep learning platform is started, the port Chi Chushi module 101 acquires configuration information of a port range, initializes an available port into a port pool, and finally persists into a port pool table. Reference may be made to the schematic diagram of fig. 2. Therein, it is readily understood that the port range specified by the configuration information should be a subset of the kubernetes nodePort port range.
As a specific embodiment, on this basis, the port Chi Chushi module 101 can verify the port data of the existing port in the current port pool, and delete the verified invalid data, that is, delete the invalid port.
After the deep learning platform operates, the port management module 102 may dynamically adjust the port range in the port pool in real time in the system configuration according to the user's port range modification instruction, and correspondingly adjust the relevant port data in the port pool, so as to increase the number of ports in the port pool in real time according to the user's needs.
Meanwhile, considering that the user needs to use the port after creating the task, the port management module 102 may select the target port according to the port acquisition instruction of the user and return the target port to the current task for the user to use. It will be readily appreciated that when a port is used by a task, its state data in the port pool can be modified to a used state and bound to the task.
Further, as a specific embodiment, the port management module 102 may specifically select in a random manner when selecting the target port, and specifically, may implement randomly selecting the port through Sql "order by RAND () limit 1".
After the ports in the port pool are used, correspondingly, after the task operation is completed, the problem of port release exists. Therefore, after the deep learning training platform is started, the listening module 103 starts a portreleaselecter, i.e. a port release listener, acquires the information of the deleted task by listening kubernetes service event, i.e. the task event message, and releases the port occupied by the deleted task. Of course, the relevant data of the released port is modified accordingly, the state of the released port is restored to be the unused state, and the released port can be reused by other tasks in the port pool.
Therefore, the port management system of the deep learning training platform provided by the application can effectively realize port management matters such as port use, port release, k8s service port segment adaptation and the like, help to autonomously manage the port range in the deep learning platform and dynamically manage the use of the port, and greatly improve the training efficiency in deep learning training.
As a specific embodiment, the port management system of the deep learning training platform provided by the embodiment of the present application is specifically configured to, based on the above content, when the port management module 102 dynamically adjusts the port range in the port pool:
performing range verification on a new port range specified in the port range modification instruction;
if the new port range is a subset of the parent set of the old port range, kubernetes nodePort port ranges, then the port ranges in the port pool are dynamically adjusted according to the new port range.
With specific reference to fig. 3, fig. 3 shows a range schematic of kubernetes nodePort port range, new port range, and old port range.
As a specific embodiment, the port management system of the deep learning training platform provided by the embodiment of the present application is based on the above, where after dynamically adjusting the port range in the port pool, the port management module 102 is further configured to:
and synchronously modifying the updated open port range to firewall rules of all hosts in the cluster by executing the shell script.
Specifically, the embodiment of the application can automatically manage the use authority of the port in the port pool in each host machine operating system by automatically modifying the firewall rules of the cluster host machines, and can timely give the newly-added port use authority to each host machine.
As a specific embodiment, the port management system of the deep learning training platform provided by the embodiment of the present application is based on the above, where after the port management module 102 selects the target port from the port pool for use by the current task, the port management system is further configured to:
and storing the corresponding relation between the target port and the current task, and modifying the state data of the target port in the port pool into a used state.
Specifically, the use order of the ports can be ensured by modifying the related data of the ports in the port pool.
As a specific embodiment, the port management system of the deep learning training platform provided by the embodiment of the present application is based on the above content, and the monitoring module 103 is further configured to:
and deleting the event monitor after the deep learning training platform stops running.
Specifically, the event monitor can be destroyed after the deep learning training platform is operated, so as to release related resources.
Referring to fig. 4, the embodiment of the application discloses a port management method of a deep learning training platform, which mainly comprises the following steps:
s201: when the deep learning training platform is started, the available ports are initialized into the port pool based on the port pool initialization module according to the configuration information of the port range.
S202: after the deep learning training platform operates, the port range in the port pool is dynamically adjusted based on the port management module responding to the port range modification instruction.
S203: the port management module selects a target port from the port pool for use by the current task in response to the port acquisition instruction.
S204: and creating an event monitor based on the monitoring module to acquire the information of the deleted task in real time and release the port occupied by the deleted task.
Therefore, the port management method of the deep learning training platform disclosed by the embodiment of the application can effectively realize port management matters such as port use, port release, k8s service port segment adaptation and the like, help to autonomously manage the port range in the deep learning platform and dynamically manage the use of the port, and greatly improve the training efficiency in deep learning training.
For the specific content of the port management method of the deep learning training platform, reference may be made to the foregoing detailed description of the port management system of the deep learning training platform, which is not repeated herein.
As a specific embodiment, the port management method of the deep learning training platform disclosed in the embodiment of the present application further includes, based on the above content, after initializing the available ports into the port pool based on the configuration information of the port range by the port pool initialization module:
and checking ports in the current port pool, and deleting invalid ports.
As a specific embodiment, the port management method of the deep learning training platform disclosed in the embodiment of the present application is based on the above, and based on the port management module responding to the port range modification instruction, dynamically adjusts the port range in the port pool, and specifically includes:
performing range verification on a new port range specified in the port range modification instruction;
if the new port range is a subset of the parent set of the old port range, kubernetes nodePort port ranges, then the port ranges in the port pool are dynamically adjusted according to the new port range.
As a specific embodiment, the port management method of the deep learning training platform disclosed in the embodiment of the present application, based on the above content, further includes, after dynamically adjusting the port range in the port pool based on the port management module responding to the port range modification instruction:
and synchronously modifying the updated open port range to firewall rules of all hosts in the cluster by executing the shell script.
As a specific embodiment, the port management method of the deep learning training platform disclosed in the embodiment of the present application, based on the above content, further includes, after selecting, based on the port management module in response to the port acquisition instruction, a target port from the port pool for use by a current task:
and storing the corresponding relation between the target port and the current task, and modifying the state data of the target port in the port pool into a used state.
As a specific embodiment, the port management method of the deep learning training platform disclosed in the embodiment of the present application further includes, based on the above content:
and deleting the event monitor based on the monitoring module after the deep learning training platform stops running.
Referring to fig. 5, an embodiment of the present application discloses an electronic device, including:
a memory 301 for storing a computer program;
a processor 302 for executing the computer program to implement the steps of any of the port management methods of the deep learning training platform as described above.
Further, the embodiment of the application also discloses a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is used for realizing the steps of the port management method of any deep learning training platform when being executed by a processor.
For the details of the electronic device and the computer readable storage medium, reference may be made to the foregoing detailed description of the port management system of the deep learning training platform, which is not repeated herein.
In the application, each embodiment is described in a progressive manner, and each embodiment is mainly used for illustrating the difference from other embodiments, and the same similar parts among the embodiments are mutually referred. For the apparatus disclosed in the examples, since it corresponds to the method disclosed in the examples, the description is relatively simple, and the relevant points are referred to in the description of the method section.
It should also be noted that in this document, relational terms such as "first" and "second" are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The technical scheme provided by the application is described in detail. The principles and embodiments of the present application have been described herein with reference to specific examples, the description of which is intended only to facilitate an understanding of the method of the present application and its core ideas. It should be noted that it will be apparent to those skilled in the art that the present application may be modified and practiced without departing from the spirit of the present application.
Claims (10)
1. A port management system for a deep learning training platform, comprising:
the port Chi Chushi initialization module is used for initializing an available port into a port pool according to configuration information of a port range when the deep learning training platform is started; wherein the port range specified by the configuration information is a subset of the kubernetes nodePort port range;
the port management module is used for responding to a port range modification instruction after the deep learning training platform operates, and dynamically adjusting the port range in the port pool; responding to a port acquisition instruction, and selecting a target port from the port pool for the current task to use;
and the monitoring module is used for creating an event monitor after the deep learning training platform operates so as to acquire the information of the deleted task in real time and release the port occupied by the deleted task.
2. The port management system of claim 1, wherein the port Chi Chushi module is further to:
and checking ports in the current port pool, and deleting invalid ports.
3. The port management system of claim 2, wherein the port management module, when dynamically adjusting the port range in the port pool, is specifically configured to:
performing range verification on the new port range specified in the port range modification instruction;
and if the new port range is a subset of the parent set of the old port range and the kubernetes nodePort port range, dynamically adjusting the port range in the port pool according to the new port range.
4. The port management system of claim 3, wherein the port management module, after dynamically adjusting port ranges in the port pool, is further to:
and synchronously modifying the updated open port range to firewall rules of all hosts in the cluster by executing the shell script.
5. The port management system of claim 4, wherein the port management module, after selecting a target port from the port pool for use by a current task, is further to:
and storing the corresponding relation between the target port and the current task, and modifying the state data of the target port in the port pool into an used state.
6. The port management system of claim 5, wherein the listening module is further configured to:
and deleting the event monitor after the deep learning training platform stops running.
7. The port management method of the deep learning training platform is characterized by comprising the following steps of:
when the deep learning training platform is started, initializing an available port into a port pool based on configuration information of a port range by a port pool initialization module; wherein the port range specified by the configuration information is a subset of the kubernetes nodePort port range;
after the deep learning training platform operates, dynamically adjusting a port range in the port pool based on a port management module responding to a port range modification instruction;
selecting a target port from the port pool for use by a current task based on the port management module responding to a port acquisition instruction;
and creating an event monitor based on the monitoring module to acquire the information of the deleted task in real time and releasing the port occupied by the deleted task.
8. The method for port management of a deep learning training platform of claim 7, further comprising, after the port pool-based initialization module initializes available ports into a port pool according to port range configuration information:
and checking ports in the current port pool, and deleting invalid ports.
9. An electronic device, comprising:
a memory for storing a computer program;
a processor for executing the computer program to implement the steps of the port management method of the deep learning training platform of claim 7 or 8.
10. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program which, when executed by a processor, is adapted to carry out the steps of the port management method of the deep learning training platform according to claim 7 or 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110744300.0A CN113590415B (en) | 2021-06-30 | 2021-06-30 | Port management system, method, equipment and medium of deep learning training platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110744300.0A CN113590415B (en) | 2021-06-30 | 2021-06-30 | Port management system, method, equipment and medium of deep learning training platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113590415A CN113590415A (en) | 2021-11-02 |
| CN113590415B true CN113590415B (en) | 2023-09-22 |
Family
ID=78245665
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110744300.0A Active CN113590415B (en) | 2021-06-30 | 2021-06-30 | Port management system, method, equipment and medium of deep learning training platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113590415B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110287029A (en) * | 2019-06-27 | 2019-09-27 | 中国—东盟信息港股份有限公司 | A method of it is adjusted based on kubernetes container resource dynamic |
| CN111190805A (en) * | 2019-12-29 | 2020-05-22 | 广东浪潮大数据研究有限公司 | Training log visualization system, method and equipment of deep learning training platform |
| CN111782486A (en) * | 2020-07-03 | 2020-10-16 | 上海浦东发展银行股份有限公司 | Alarm implementation method and system based on dynamic configuration |
| CN112104486A (en) * | 2020-08-31 | 2020-12-18 | 中国—东盟信息港股份有限公司 | Kubernetes container-based network endpoint slicing method and system |
| WO2021074668A1 (en) * | 2019-10-14 | 2021-04-22 | Telefonaktiebolaget Lm Ericsson (Publ) | Rtps discovery in kubernetes |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11494633B2 (en) * | 2017-12-30 | 2022-11-08 | Intel Corporation | Techniques to manage training or trained models for deep learning applications |
-
2021
- 2021-06-30 CN CN202110744300.0A patent/CN113590415B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110287029A (en) * | 2019-06-27 | 2019-09-27 | 中国—东盟信息港股份有限公司 | A method of it is adjusted based on kubernetes container resource dynamic |
| WO2021074668A1 (en) * | 2019-10-14 | 2021-04-22 | Telefonaktiebolaget Lm Ericsson (Publ) | Rtps discovery in kubernetes |
| CN111190805A (en) * | 2019-12-29 | 2020-05-22 | 广东浪潮大数据研究有限公司 | Training log visualization system, method and equipment of deep learning training platform |
| CN111782486A (en) * | 2020-07-03 | 2020-10-16 | 上海浦东发展银行股份有限公司 | Alarm implementation method and system based on dynamic configuration |
| CN112104486A (en) * | 2020-08-31 | 2020-12-18 | 中国—东盟信息港股份有限公司 | Kubernetes container-based network endpoint slicing method and system |
Non-Patent Citations (2)
| Title |
|---|
| 基于容器云的深度学习平台设计与实现;朱天放;梁明;王家乐;杨志浩;李亚敏;;电子设计工程(第09期);27-31 * |
| 基于容器技术的管控融合一体化系统;赵昱;中国优秀硕士学位论文全文数据库;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113590415A (en) | 2021-11-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10303450B2 (en) | Systems and methods for a policy-driven orchestration of deployment of distributed applications | |
| CN102571698B (en) | Access authority control method, system and device for virtual machine | |
| CN105684357A (en) | Management of addresses in virtual machines | |
| CN111555913A (en) | Simulation method, system, electronic device and storage medium for simulating real network environment based on virtualization | |
| CN108462746A (en) | A kind of container dispositions method and framework based on openstack | |
| DE112011100094T5 (en) | Method and system for abstracting a non-functional requirement based use of virtual machines | |
| CN104765620A (en) | Programming module deploying method and system | |
| CN111026723B (en) | Big data cluster management control method and device, computer equipment and storage medium | |
| CN110502217B (en) | ROS-based robot cloud platform design method | |
| US11894983B2 (en) | Simulation and testing of infrastructure as a service scale using a container orchestration engine | |
| CN111880738A (en) | Method for automatically creating and mounting LVM (logical volume manager) volume in K8s environment | |
| CN114341847A (en) | Correspondence of external operations on containers to mutation events | |
| CN110278101A (en) | A kind of method for managing resource and equipment | |
| CN115248692A (en) | Device and method for supporting cloud deployment of multiple deep learning framework models | |
| CN106371931B (en) | A kind of high-performance geoscience computing service system based on Web frame | |
| DE112021005927T5 (en) | PATCHING WORKFLOWS | |
| CN113806097A (en) | Data processing method and device, electronic equipment and storage medium | |
| DE112021003803T5 (en) | POOL MANAGEMENT FOR APPLICATION LAUNCH IN AN ONBOARD UNIT | |
| CN113590415B (en) | Port management system, method, equipment and medium of deep learning training platform | |
| CN116974689A (en) | Cluster container scheduling method, device, equipment and computer readable storage medium | |
| CN111352664A (en) | Distributed machine learning task starting method, system, equipment and storage medium | |
| CN109739666A (en) | Striding course call method, device, equipment and the storage medium of singleton method | |
| CN115202815A (en) | AI teaching practical training system and method based on cloud protogenesis | |
| DE102022108636A1 (en) | LIFE CYCLE MANAGEMENT FOR WORKLOADS IN HETEROGENOUS INFRASTRUCTURES | |
| CN114564211A (en) | Cluster deployment method, cluster deployment device, equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |