CN107450976A - A kind of user Explore of Unified Management Ideas of high performance computing system - Google Patents
A kind of user Explore of Unified Management Ideas of high performance computing system Download PDFInfo
- Publication number
- CN107450976A CN107450976A CN201710849994.8A CN201710849994A CN107450976A CN 107450976 A CN107450976 A CN 107450976A CN 201710849994 A CN201710849994 A CN 201710849994A CN 107450976 A CN107450976 A CN 107450976A
- Authority
- CN
- China
- Prior art keywords
- user
- servers
- unified management
- explore
- job scheduling
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F15/00—Digital computers in general; Data processing equipment in general
- G06F15/16—Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs
- G06F15/161—Computing infrastructure, e.g. computer clusters, blade chassis or hardware partitioning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F15/00—Digital computers in general; Data processing equipment in general
- G06F15/16—Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs
- G06F15/163—Interprocessor communication
- G06F15/173—Interprocessor communication using an interconnection network, e.g. matrix, shuffle, pyramid, star, snowflake
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/56—Provisioning of proxy services
- H04L67/566—Grouping or aggregating service requests, e.g. for unified processing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Multi Processors (AREA)
Abstract
The present invention discloses a kind of user's Explore of Unified Management Ideas of high performance computing system, including：NIS servers are built in (SuSE) Linux OS；AD domain servers are built in Windows operating system；WebService services are disposed in job scheduling system server, NIS servers and AD domain servers；Job scheduling system is managed to user, generates user profile；Serviced by WebService and send user profile to NIS servers and AD domain servers；NIS servers and AD domain servers are managed based on user's information.User's Explore of Unified Management Ideas of the present invention, different types of operating system user is managed collectively, while the user management in job scheduling system, the user in different types of operating system can be managed collectively simultaneously, the user in high performance computing system is realized to be managed collectively, solve the problems, such as the user management of the calculate node of different type operating system in high performance system, realize the simplification to user management mode.
Description
Technical field
The present invention relates to user management technical field.It is unified more particularly, to a kind of user of high performance computing system
Management method.
Background technology
High-performance calculation (High performance computing, abbreviation HPC) refers to (to be made usually using many processors
For a part for individual machine) or a certain cluster in organize several computers (as single computing resource operate) calculating
System and environment, it is that one kind incorporates software and hardware resources, forms the advantage of scale that magnanimity calculates, and by network offering services,
The system-level super large model for calculating analysis is supported to calculate.High performance computing system can carry out multi-disciplinary analysis and calculate, and can
Carry out the work such as system-level large artificial.Calculate node in high performance computing system generally include Linux, Windows and
Unix operating systems, the unified management and scheduling to high performance computing system are realized by job scheduling software.Due in high property
It can need to realize the computation schema of multi-host parallel in computing system, therefore the calculation procedure in high performance computing system needs to use
Operating system user identity is run, and the user of job scheduling system needs to be mapped with operating system user to ensure to count
Calculate the normal operation of program.
, it is necessary to carry out the establishment to operating system user using script after user's establishment of job scheduling system.When high property
The operating system of calculate node is polytype in energy computing system, and such as a part of calculate node uses Linix, and some calculates
, it is necessary to carry out unification to user management in the operating system of these isomeries in the case that node uses Windows operating system.
Because the user management mode of different types of operating system is different, the mode of generally use is manually different types of at present
Corresponding user is created in operating system, it is less efficient.
Therefore the problem of requiring consideration for how simultaneously to be managed collectively different types of operating system user, i.e., right
While user management in job scheduling system, unified pipe can be carried out to the user in different types of operating system simultaneously
Reason, realizes the simplification to user management mode, it is desirable to provide a kind of user's Explore of Unified Management Ideas of high performance computing system.
The content of the invention
It is an object of the invention to provide a kind of user's Explore of Unified Management Ideas of high performance computing system, while to inhomogeneity
User in the operating system of type is managed collectively, to realize to the simplification in a manner of user management.
To reach above-mentioned purpose, the present invention uses following technical proposals：
A kind of user's Explore of Unified Management Ideas of high performance computing system, user's Explore of Unified Management Ideas operate for Linux
System and the unified management of Windows operating system user, comprise the following steps：
S1：NIS servers are built in (SuSE) Linux OS, the unified management for (SuSE) Linux OS user；
S2：AD domain servers are built in Windows operating system, the unified pipe for Windows operating system user
Reason；
S3：WebService services are disposed in job scheduling system server, NIS servers and AD domain servers, are used
In the information transmission of high performance computing system；
S4：Job scheduling system is managed to user, generates user profile；
S5：Serviced by WebService and send user profile to NIS servers and AD domain servers；
S6：NIS servers and AD domain servers are managed based on user's information.
Preferably, this method also includes：
S7：Relational database is disposed in job scheduling system server, for storing user profile.
It is further preferred that user is managed including creating user, modification user and deleting user
Preferably, step S4 includes：
S41：Job scheduling system creates user；
S42：Judge that user whether there is in relational database；
S43：If so, then terminate to create；If it is not, then generate the user profile of user and be stored in relational database.
Preferably, step S4 includes：
S41：Job scheduling system changes user；
S42：Judge that user whether there is in relational database；
S43：If it is not, then terminate to change；If so, then change the user profile of user and be stored in relational database.
Preferably, step S4 includes：
S41：Job scheduling system deletes user；
S42：Judge that user whether there is in relational database；
S43：If it is not, then terminate to delete；If so, the user profile of user is then deleted from relational database.
Preferably, NIS servers and AD domain servers are managed by script and/or program to user.
Beneficial effects of the present invention are as follows：
A kind of user's Explore of Unified Management Ideas of high-performance computer system of the present invention, made by WebService services
The transmission of user profile is carried out between industry scheduling system server, NIS servers and AD domain servers, to different types of operation
System user is managed collectively, can be simultaneously to different types of while the user management in job scheduling system
User in operating system is managed collectively, and is realized the user in high performance computing system and is managed collectively, solves high property
The user management problem of the calculate node of different type operating system, realizes the simplification to user management mode in energy system.
Brief description of the drawings
The embodiment of the present invention is described in further detail below in conjunction with the accompanying drawings.
Fig. 1 shows to build running environment schematic diagram.
Fig. 2 shows to create user's schematic flow sheet.
Fig. 3 shows to change user's schematic flow sheet.
Fig. 4 shows to delete user's schematic flow sheet.
Embodiment
In order to illustrate more clearly of the present invention, the present invention is done further with reference to preferred embodiments and drawings
It is bright.Similar part is indicated with identical reference in accompanying drawing.It will be appreciated by those skilled in the art that institute is specific below
The content of description is illustrative and be not restrictive, and should not be limited the scope of the invention with this.
A kind of user's Explore of Unified Management Ideas of high performance computing system, user's Explore of Unified Management Ideas operate for Linux
System and the unified management of Windows operating system user, including：
The first step builds running environment
1) NIS servers, are built, are responsible for the unified management of (SuSE) Linux OS user.
2) AD domain servers, are built, are responsible for the unified management of Windows operating system user.
3) WebService services, are disposed in job scheduling system server, NIS servers, AD domain servers, are responsible for
Information transmission in whole high performance computing system.
4) relational database, is disposed in job scheduling system server, is responsible for the storage to user profile.
Second step user management mode
1), job scheduling system is created to user, changed, deleted.
2), user profile is stored to relational database.
3), user management relevant information is sent to NIS servers and AD domain servers by WebService.
4) after, NIS servers and AD domain servers obtain user management relevant information, realized by script or program corresponding
Management operation.
Calculate node in high performance computing system is usually Linux, Windows, Unix operating system, and the present invention is directed to
Computer node is the unified pipe of a kind of user of high performance computing system proposition of (SuSE) Linux OS and Windows operating system
Reason method.In the high performance computing system, unified management and the tune of high performance computing system are realized by job scheduling software
Degree, specifically, high performance computing system includes job scheduling system in the present invention, for realizing multiprocessing parallel calculation.
Job scheduling system needs the user identity using (SuSE) Linux OS and Windows operating system to be run,
I.e. job scheduling system needs to be mapped with the user of (SuSE) Linux OS and Windows operating system to ensure to calculate journey
The normal operation of sequence.
NIS (Network Information Service, the network information service) purposes is exactly to provide letter to client
Breath inquiry, the account number of such as user, password, default directory, meanwhile, in high performance computing system, all nodes need
Identical user profile, come centralized user information management it is a kind of efficient wide sharp mode based on NIS servers, the addition of user
Deleting only needs to operate in main controlled node, reduces the complexity of user management.
In the present invention, the deployment of running environment is carried out first.Specifically, as shown in figure 1, being taken in (SuSE) Linux OS
NIS servers are built, AD domain servers are built in Windows operating system, meanwhile, taken in job scheduling system server, NIS
It is engaged in disposing WebService services in device and AD domain servers, completes the deployment of running environment.
In above-mentioned running environment, job scheduling system is managed to user identity, including creates user, modification user
With delete user, wherein
(1) user is created
It is as shown in Figure 2 to create user's flow：
Job scheduling system, which is sent, to be started to create user instruction, is created user and is generated user profile；
User profile in call relation database；
Whether the user profile for judging to create is present in the relational database；
If so, then job scheduling system terminates to create user's flow；
If it is not, then sent user profile to NIS servers and AD domain servers by WebService, meanwhile, it will use
Family information is added in database and stored；
NIS servers create Linux user by script, and are synchronized to calculate node；
AD domain servers create Windows domain users by script；
Terminate to create user's flow.
(2) user is changed
It is as shown in Figure 3 to change user's flow：
Job scheduling system, which is sent, to be started to change user instruction, is changed user and is generated user profile；
User profile in call relation database；
Judge whether the user profile of modification is present in the relational database；
If it is not, then job scheduling system terminates to change user's flow；
If so, then sent user profile to NIS servers and AD domain servers by WebService, meanwhile, it will use
Family information is changed to be stored into database；
NIS servers change Linux user by script, and are synchronized to calculate node；
AD domain servers change Windows domain users by script；
Terminate modification user's flow.
(3) user is deleted
It is as shown in Figure 4 to delete user's flow：
Job scheduling system, which is sent, to be started to delete user instruction, and user profile is deleted in generation；
User profile in call relation database；
Whether the user profile for judging to delete is present in the relational database；
If it is not, then job scheduling system terminates to delete user's flow；
If so, then sent user profile to NIS servers and AD domain servers by WebService, meanwhile, it will use
Family information is deleted from database；
NIS servers delete Linux user by script, and are synchronized to calculate node；
AD domain servers delete Windows domain users by script；
Terminate to create user's flow.
It should be understood that after NIS servers and AD domain servers obtain user management relevant information, pass through script or program
Corresponding management operation is realized, wherein establishment, modification and deletion action of the management operation corresponding to job scheduling system.
Obviously, the above embodiment of the present invention is only intended to clearly illustrate example of the present invention, and is not pair
The restriction of embodiments of the present invention, for those of ordinary skill in the field, may be used also on the basis of the above description
To make other changes in different forms, all embodiments can not be exhaustive here, it is every to belong to this hair
Row of the obvious changes or variations that bright technical scheme is extended out still in protection scope of the present invention.
Claims (7)
1. user's Explore of Unified Management Ideas of a kind of high performance computing system, it is characterised in that user's Explore of Unified Management Ideas is used
In the unified management of (SuSE) Linux OS and Windows operating system user, comprise the following steps：
S1：NIS servers are built in (SuSE) Linux OS, the unified management for (SuSE) Linux OS user；
S2：AD domain servers are built in Windows operating system, the unified management for Windows operating system user；
S3：WebService clothes are disposed in job scheduling system server, the NIS servers and the AD domain servers
Business, the information transmission for high performance computing system；
S4：Job scheduling system is managed to user, generates user profile；
S5：Serviced by WebService and send the user profile to the NIS servers and the AD domain servers；
S6：The NIS servers and the AD domain servers are managed based on the user's information.
2. user's Explore of Unified Management Ideas according to claim 1, it is characterised in that also include：
S7：Relational database is disposed in job scheduling system server, for storing the user profile.
3. user's Explore of Unified Management Ideas according to claim 2, it is characterised in that be managed to user and used including creating
Family, modification user and deletion user.
4. user's Explore of Unified Management Ideas according to claim 3, it is characterised in that the step S4 includes：
S41：Job scheduling system creates user；
S42：Judge that the user whether there is in the relational database；
S43：If so, then terminate to create；If it is not, then generate the user profile of the user and be stored in the relational database
In.
5. user's Explore of Unified Management Ideas according to claim 3, it is characterised in that the step S4 includes：
S41：Job scheduling system changes user；
S42：Judge that the user whether there is in the relational database；
S43：If it is not, then terminate to change；If so, then change the user profile of the user and be stored in the relational database
In.
6. user's Explore of Unified Management Ideas according to claim 3, it is characterised in that the step S4 includes：
S41：Job scheduling system deletes user；
S42：Judge that the user whether there is in the relational database；
S43：If it is not, then terminate to delete；If so, the user profile of the user is then deleted from the relational database.
7. user's Explore of Unified Management Ideas according to claim 1, it is characterised in that the NIS servers and the AD domains
Server is managed by script and/or program to user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710849994.8A CN107450976A (en) | 2017-09-20 | 2017-09-20 | A kind of user Explore of Unified Management Ideas of high performance computing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710849994.8A CN107450976A (en) | 2017-09-20 | 2017-09-20 | A kind of user Explore of Unified Management Ideas of high performance computing system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107450976A true CN107450976A (en) | 2017-12-08 |
Family
ID=60496991
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710849994.8A Pending CN107450976A (en) | 2017-09-20 | 2017-09-20 | A kind of user Explore of Unified Management Ideas of high performance computing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107450976A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112631774A (en) * | 2020-12-23 | 2021-04-09 | 广东省大湾区集成电路与***应用研究院 | Task processing method, device, equipment and medium based on grid computing |
| WO2021073415A1 (en) * | 2019-10-15 | 2021-04-22 | 平安科技（深圳）有限公司 | User management method and device for server out-of-band management system |
| CN113497814A (en) * | 2020-03-19 | 2021-10-12 | 中科星图股份有限公司 | Satellite image processing algorithm hybrid scheduling system and method |
| CN113778790A (en) * | 2021-08-19 | 2021-12-10 | 北京仿真中心 | Method and system for monitoring state of computing system based on Zabbix |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103812865A (en) * | 2014-01-28 | 2014-05-21 | 北京仿真中心 | Method of realizing transparent user login under cloud resource platform |
| CN105429999A (en) * | 2015-12-17 | 2016-03-23 | 北京荣之联科技股份有限公司 | Unified identity authentication system based on cloud platform |
| CN105577656A (en) * | 2015-12-17 | 2016-05-11 | 北京荣之联科技股份有限公司 | Unified identity authentication method based on cloud platform |
-
2017
- 2017-09-20 CN CN201710849994.8A patent/CN107450976A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103812865A (en) * | 2014-01-28 | 2014-05-21 | 北京仿真中心 | Method of realizing transparent user login under cloud resource platform |
| CN105429999A (en) * | 2015-12-17 | 2016-03-23 | 北京荣之联科技股份有限公司 | Unified identity authentication system based on cloud platform |
| CN105577656A (en) * | 2015-12-17 | 2016-05-11 | 北京荣之联科技股份有限公司 | Unified identity authentication method based on cloud platform |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021073415A1 (en) * | 2019-10-15 | 2021-04-22 | 平安科技（深圳）有限公司 | User management method and device for server out-of-band management system |
| CN113497814A (en) * | 2020-03-19 | 2021-10-12 | 中科星图股份有限公司 | Satellite image processing algorithm hybrid scheduling system and method |
| CN112631774A (en) * | 2020-12-23 | 2021-04-09 | 广东省大湾区集成电路与***应用研究院 | Task processing method, device, equipment and medium based on grid computing |
| CN113778790A (en) * | 2021-08-19 | 2021-12-10 | 北京仿真中心 | Method and system for monitoring state of computing system based on Zabbix |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11128465B2 (en) | Zero-knowledge identity verification in a distributed computing system | |
| CN104573115B (en) | Support the realization method and system of the integrated interface of multi-type database operation | |
| Lazouski et al. | Usage control in cloud systems | |
| CN107450976A (en) | A kind of user Explore of Unified Management Ideas of high performance computing system | |
| JP2019534496A (en) | Managed query service | |
| Ju et al. | iGraph: an incremental data processing system for dynamic graph | |
| CN105138679B (en) | A kind of data processing system and processing method based on distributed caching | |
| CN105468720A (en) | Method for integrating distributed data processing systems, corresponding systems and data processing method | |
| US20200250587A1 (en) | Framework for multi-tenant data science experiments at-scale | |
| US11656902B2 (en) | Distributed container image construction scheduling system and method | |
| WO2021013056A1 (en) | Microservice-based data processing method and apparatus, and device and readable storage medium | |
| CN109726004A (en) | A kind of data processing method and device | |
| US20230040635A1 (en) | Graph-based impact analysis of misconfigured or compromised cloud resources | |
| Xu et al. | Test suite reduction using weighted set covering techniques | |
| CN112685499A (en) | Method, device and equipment for synchronizing process data of work service flow | |
| US7752225B2 (en) | Replication and mapping mechanism for recreating memory durations | |
| Convolbo et al. | DRASH: A data replication-aware scheduler in geo-distributed data centers | |
| CN108153859A (en) | A kind of effectiveness order based on Hadoop and Spark determines method parallel | |
| CN117873719A (en) | Cluster dynamic updating method based on DevOps | |
| US20230093868A1 (en) | Mechanism for real-time identity resolution in a distributed system | |
| US10853131B2 (en) | Dataflow life cycles | |
| KR101935249B1 (en) | Method for processing based on flow using stored procedure, system and apparatus | |
| Kang et al. | A comparative analysis of iterative MapReduce systems | |
| Lee et al. | Benchmarking large-scale object storage servers | |
| CN110019415A (en) | Cloud computing big data digging system based on mobile terminal |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171208 |
|
| RJ01 | Rejection of invention patent application after publication |