CN105824704A - Method, apparatus and system for using graphic workstations - Google Patents
Method, apparatus and system for using graphic workstations Download PDFInfo
- Publication number
- CN105824704A CN105824704A CN201610206126.3A CN201610206126A CN105824704A CN 105824704 A CN105824704 A CN 105824704A CN 201610206126 A CN201610206126 A CN 201610206126A CN 105824704 A CN105824704 A CN 105824704A
- Authority
- CN
- China
- Prior art keywords
- graphics
- work station
- requirement condition
- node
- graphics workstation
- 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
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5077—Logical partitioning of resources; Management or configuration of virtualized resources
Abstract
The invention provides a method, an apparatus and a system for using graphic workstations. The method comprises: after receiving a graphic workstation request notification with specific requirements, which is sent by an external user side device, by a management node, firstly judging whether available graphic workstations exist; if the available graphic workstations exist, according to a predetermined application mirror image and any available graphic workstation, acquiring an operation interface of a virtual workstation corresponding to the specific requirements, and feeding back the operation interface to the external user side device. The management node can be utilized to carry out uniform state management and resource allocation on all the graphic workstations and provide corresponding virtual graphic workstation resources according to the specific requirements of a user, and the user can utilize the received operation interface of the virtual workstation to remotely operate the virtual workstation on a local device so as to execute work tasks, and thus, according to the scheme, the coordination capacity for the graphic workstation resources can be improved.
Description
Technical field
The present invention relates to field of computer technology, use the method for graphics workstation, Apparatus and system particularly to a kind of.
Background technology
In high-performance calculation, such as CFD (ComputationalFluidDynamics, computational fluid dynamics) field, it will usually relate to more graphics processing operation.Owing to graphics workstation configuration is high, graphics capability is strong, therefore staff usually utilizes graphics workstation to perform graphics processing operation.
At present, staff can utilize graphics workstation to perform respective task, and when supply falls short of demand for graphics workstation, coordinates mutually to use between each staff.
Owing to the working condition of each staff is possibly different from, and plan changeable, therefore existing implementation is relatively low to the coordination ability of graphics workstation resource.
Summary of the invention
The invention provides and a kind of use the method for graphics workstation, Apparatus and system, it is possible to increase the coordination ability to graphics workstation resource.
In order to achieve the above object, the present invention is achieved through the following technical solutions:
On the one hand, the invention provides a kind of method using graphics workstation, be applied to manage node, pre-build described management node and at least one external user side apparatus and the annexation with at least one external graphics work station, determine that at least one applies mirror image, also include:
S1: receive the first notification message of the request graphics workstation that external user side apparatus sends, and carry corresponding first requirement condition in described first notification message；
S2: judge whether there is available graphics workstation at least one external graphics work station described, if so, perform S3；
S3: according at least one application mirror image described, and arbitrary described available graphics workstation, it is thus achieved that the operation interface of the virtualizing graphics work station corresponding with described first requirement condition；
S4: the operation interface of described virtualizing graphics work station is sent to described external user side apparatus.
Further, described determine at least one application mirror image after, also include: be sent at least one application mirror image described outside share storage；
Described S3, including: described first requirement condition is sent to the either objective graphics workstation in described available graphics workstation；After the application mirror image corresponding with described first requirement condition utilizing described outside to share in storage at described targeted graphical work station fictionalizes respective virtual graphics workstation, receive the operation interface of the described virtualizing graphics work station sent by described targeted graphical work station.
Further, described application mirror image includes: the mirror image of graphics workstation, and/or, for building the mirror image of the software of graphics workstation；
Described first requirement condition includes: the mark of graphics workstation, and/or, for building the mark of the software of graphics workstation.
Further, described operation interface includes: telework desktop.
Further, before S1, also include: pre-build the annexation of described management node and at least one external computing nodes；
After S4, also include:
A1: utilize the operation interface of described external user side apparatus and described virtualizing graphics work station after completing corresponding task user, receive the second notification message that the task of described external user side apparatus transmission has performed, and described second notification message carries corresponding second requirement condition；
A2: judge whether to exist at least one external computing nodes described the idle calculating node corresponding with described second requirement condition, if so, perform A3, otherwise, according to the reception time of described second notification message, wait that described second notification message is ranked, and perform A2；
A3: described second requirement condition is sent to the described idle calculating node corresponding with described second requirement condition；
A4: receive the third notice message calculated by the idle task calculating node transmission that described and described second requirement condition is corresponding；
A5: described third notice message is sent to described external user side apparatus, and performs S1.
Further, described second requirement condition includes: task is required calculates any one or more in the mark of the number of node, the saving contents of task, the preservation title of task and task.
Further, before S1, also include: the management assembly of openstack is installed in advance in described management node, and unified web user login management system is set.
Further, when judging that in S2 at least one external graphics work station described does not exists available graphics workstation, also include: according to the reception time of described first notification message, wait that described first notification message is ranked, and perform S2.
On the other hand, the invention provides a kind of management node, including:
Unit is set, for setting up described management node and at least one external user side apparatus and the annexation with at least one external graphics work station, and determines that at least one applies mirror image；
Receive unit, for receiving the first notification message of the request graphics workstation that external user side apparatus sends, and described first notification message carries corresponding first requirement condition, and trigger judging unit；
Described judging unit, is used for judging whether there is available graphics workstation at least one external graphics work station described, if so, triggers processing unit；
Described processing unit, is used for according at least one application mirror image described, and arbitrary described available graphics workstation, it is thus achieved that the operation interface of the virtualizing graphics work station corresponding with described first requirement condition；And the operation interface of described virtualizing graphics work station is sent to described external user side apparatus.
Further, described unit is set, is additionally operable to be sent at least one application mirror image described outside share storage；
Described processing unit, specifically for the either objective graphics workstation described first requirement condition being sent in described available graphics workstation；And after the application mirror image corresponding with described first requirement condition utilizing described outside to share in storage at described targeted graphical work station fictionalizes respective virtual graphics workstation, receive the operation interface of the described virtualizing graphics work station sent by described targeted graphical work station.
Further, described unit is set, is additionally operable to the annexation setting up described management node with at least one external computing nodes；
Described reception unit, it is additionally operable to utilize the operation interface of described external user side apparatus and described virtualizing graphics work station after completing corresponding task user, receive the second notification message that the task of described external user side apparatus transmission has performed, and described second notification message carries corresponding second requirement condition；
Described judging unit, it is additionally operable to judge whether at least one external computing nodes described exists the idle calculating node corresponding with described second requirement condition, if, trigger described processing unit, otherwise, according to the reception time of described second notification message, wait of ranking described second notification message, triggers described judging unit；
Described processing unit, is additionally operable to described second requirement condition is sent to the described idle calculating node corresponding with described second requirement condition；Receive the third notice message calculated by the idle task calculating node transmission that described and described second requirement condition is corresponding；And described third notice message is sent to described external user side apparatus, and trigger described reception unit.
On the other hand, the invention provides a kind of system using graphics workstation, including:
Any of the above-described described management node, at least one user side equipment and at least one graphics workstation, wherein,
Described management node maintains the connection with user side equipment each described and with graphics workstation each described respectively；
Described user side equipment, for receiving the first notification message of the request graphics workstation of input, and carries corresponding first requirement condition, and described first notification message is sent to described management node in described first notification message；Receive the operation interface of the virtualizing graphics work station corresponding with described first requirement condition that described management node sends；
Described graphics workstation, for the operation interface of the virtualizing graphics work station corresponding with described first requirement condition to the offer of described management node.
Further, this system also includes:
At least one calculates node and shares storage, wherein,
Described calculating node, for receiving the second requirement condition that described management node sends；Obtain the task corresponding with described second requirement condition in described shared storage；Calculate the described task got, generate corresponding result of calculation；According to described second requirement condition, corresponding described result of calculation is sent to described shared storage, and the third notice message calculated by corresponding task is sent to described management node；
Described shared storage, for receiving and store at least one the application mirror image sent by described management node；Receive and store the task that each described user side equipment sends；Receive and store the result of calculation that each described calculating node sends；
Described graphics workstation, specifically for receiving the first requirement condition sent by described management node；Obtain the intended application mirror image corresponding with described first requirement condition in described shared storage；Fictionalize the virtualizing graphics work station corresponding with described intended application mirror image；It is sent to the operation interface of described virtualizing graphics work station manage node；Received, by described virtualizing graphics work station, the control instruction that user inputs by operating described operation interface on described user side equipment, and by performing described control instruction, generate corresponding task；
Described user side equipment, is further used for the operation interface of the described virtualizing graphics work station received is supplied to user；And by the task of virtualizing graphics work station generation described in described operation interface display.
Further, maintain the connection described in and include: described management node, at least one user side equipment described, at least one graphics workstation described, at least one calculating node described and described shared storage are deployed in same LAN；
Described management node and described at least one calculate node deployment in same cluster, or, described management node, described at least one calculate node and at least one graphics workstation described and be deployed in same cluster.
Further, also include: the management assembly of openstack is installed in advance in described management node, calculating node each described and each described graphics workstation are respectively mounted the computation module of openstack.
The invention provides and a kind of use the method for graphics workstation, Apparatus and system, management node is after receiving the graphics workstation request notice carrying particular requirement that external user side apparatus sends, it is first determined whether there is available graphics workstation, if existing, according to the application mirror image predefined and arbitrary available graphics workstation, obtain the operation interface of the virtual workstation corresponding with described particular requirement, and this operation interface is fed back to described external user side apparatus.Owing to can utilize management node that whole graphics workstations carry out unified condition managing and resource distribution, and provide corresponding virtualizing graphics workshation resource according to the particular requirement of user for it, and user can utilize the operation interface of the virtual workstation received on the local device, operate to perform task to remote dummy work station, therefore, the present invention can improve the coordination ability to graphics workstation resource.
Accompanying drawing explanation
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, the accompanying drawing used required in embodiment or description of the prior art will be briefly described below, apparently, accompanying drawing in describing below is some embodiments of the present invention, for those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is the flow chart of a kind of method using graphics workstation that one embodiment of the invention provides；
Fig. 2 is the flow chart of the another kind of method using graphics workstation that one embodiment of the invention provides；
Fig. 3 is a kind of schematic diagram managing node that one embodiment of the invention provides；
Fig. 4 is the schematic diagram of the another kind of management node that one embodiment of the invention provides；
Fig. 5 is the schematic diagram of a kind of system using graphics workstation that one embodiment of the invention provides；
Fig. 6 is the schematic diagram of the another kind of system using graphics workstation that one embodiment of the invention provides.
Detailed description of the invention
For making the purpose of the embodiment of the present invention, technical scheme and advantage clearer; below in conjunction with the accompanying drawing in the embodiment of the present invention; technical scheme in the embodiment of the present invention is clearly and completely described; obviously; described embodiment is a part of embodiment of the present invention rather than whole embodiments, based on the embodiment in the present invention; the every other embodiment that those of ordinary skill in the art are obtained on the premise of not making creative work, broadly falls into the scope of protection of the invention.
As it is shown in figure 1, embodiments provide a kind of method using graphics workstation, it is applied to manage node, may comprise steps of:
Step 101: pre-build described management node and at least one external user side apparatus and the annexation with at least one external graphics work station, determines that at least one applies mirror image.
Step 102: receive the first notification message of the request graphics workstation that external user side apparatus sends, and carry corresponding first requirement condition in described first notification message.
Step 103: judge whether to exist at least one external graphics work station described available graphics workstation, if so, perform step 104.
Step 104: according at least one application mirror image described, and arbitrary described available graphics workstation, it is thus achieved that the operation interface of the virtualizing graphics work station corresponding with described first requirement condition.
Step 105: the operation interface of described virtualizing graphics work station is sent to described external user side apparatus.
Embodiments provide a kind of method using graphics workstation, management node is after receiving the graphics workstation request notice carrying particular requirement that external user side apparatus sends, it is first determined whether there is available graphics workstation, if existing, according to the application mirror image predefined and arbitrary available graphics workstation, obtain the operation interface of the virtual workstation corresponding with described particular requirement, and this operation interface is fed back to described external user side apparatus.Owing to can utilize management node that whole graphics workstations carry out unified condition managing and resource distribution, and provide corresponding virtualizing graphics workshation resource according to the particular requirement of user for it, and user can utilize the operation interface of the virtual workstation received on the local device, operate to perform task to remote dummy work station, therefore, the embodiment of the present invention can improve the coordination ability to graphics workstation resource.
In a kind of possible implementation, in order to the acquisition mode at the operation interface of virtualizing graphics work station is described, so, described determine at least one application mirror image after, farther include: be sent at least one application mirror image described outside share storage；
Described step 104, including: described first requirement condition is sent to the either objective graphics workstation in described available graphics workstation；After the application mirror image corresponding with described first requirement condition utilizing described outside to share in storage at described targeted graphical work station fictionalizes respective virtual graphics workstation, receive the operation interface of the described virtualizing graphics work station sent by described targeted graphical work station.
In a kind of possible implementation, in order to apply determined by illustrating to manage in node mirror image multiple may form, so,
Described application mirror image includes: the mirror image of graphics workstation, and/or, for building the mirror image of the software of graphics workstation；
Described first requirement condition includes: the mark of graphics workstation, and/or, for building the mark of the software of graphics workstation.
In a kind of possible implementation, for the ease of user can by local computing with indirect operation graphics workstation, so, described operation interface includes: telework desktop.
In a kind of possible implementation, in order to according to practical application request, illustrate that user uses the overall related realization process of graphics workstation, so,
Before step 102, farther include: pre-build the annexation of described management node and at least one external computing nodes；
After step 105, farther include:
Step 106: utilize the operation interface of described external user side apparatus and described virtualizing graphics work station after completing corresponding task user, receive the second notification message that the task of described external user side apparatus transmission has performed, and described second notification message carries corresponding second requirement condition；
Step 107: judge whether to exist at least one external computing nodes described the idle calculating node corresponding with described second requirement condition, if, perform step 108, otherwise, the reception time according to described second notification message, described second notification message is ranked wait, and perform step 107；
Step 108: described second requirement condition is sent to the described idle calculating node corresponding with described second requirement condition；
Step 109: receive the third notice message calculated by the idle task calculating node transmission that described and described second requirement condition is corresponding；
Step 110: described third notice message is sent to described external user side apparatus, and performs step 102.
In a kind of possible implementation, calculate with the task that user is submitted to according to user's request for the ease of management node, so, described second requirement condition includes: task is required calculates any one or more in the mark of the number of node, the saving contents of task, the preservation title of task and task.
In a kind of possible implementation, in order to make management node hold the management to each resource, scheduling feature, so, before step 102, farther include: the management assembly of openstack is installed in advance in described management node；
On arbitrary user side equipment, service communication can be carried out with management node for the ease of user, and check relative recording information, so, before step 102, farther include: unified web user login management system is set.
In a kind of possible implementation, in order to alleviate the graphics workstation practical situations that temporarily supply falls short of demand, so, when judging that at least one external graphics work station described does not exists available graphics workstation in step 103, farther include: according to the reception time of described first notification message, described first notification message is ranked wait, and perform step 103.
For making the object, technical solutions and advantages of the present invention clearer, below in conjunction with the accompanying drawings and specific embodiment the present invention is described in further detail.
As in figure 2 it is shown, one embodiment of the invention provides the method using graphics workstation, the method may comprise steps of:
Step 201: set up management node respectively and share storage, N1Individual local computing, N2Individual graphics workstation and N3The annexation of individual calculating node, and set up shared storage and management node, N respectively1Individual local computing, N2Individual graphics workstation and N3The annexation of individual calculating node.
In detail, N1、N2、N3It can be all the natural number not less than 1.
In the present embodiment, the local computing of the common configuration level that user side equipment can be used by user.
In the present embodiment, according to the concrete configuration situation of graphics workstation, at least one virtualizing graphics work station all can be fictionalized on each graphics workstation.Such as, the graphics workstation in the present embodiment can be to be provided with the high configuration service device that GPU (GraphicsProcessingUnit, graphic process unit) blocks, it is possible to supports to fictionalize 8 virtualization work stations.Accordingly, management node and calculating node can be all the high configuration service device not installing GPU card.
In detail, user can use the user name password of self, logs in unified management system by any one local computing, with to management node application graphics workstation, execution task etc..Management node can to graphics workstation resource, calculate node resource etc. and carry out unified resource management, dispatch and distribute.Calculate node to may be used for the task of user's submission is calculated.Shared storage may be used for storing task etc..
In the present embodiment, in order to set up above two annexation, can be by management node, N1Individual local computing, N2Individual graphics workstation, N3Individual calculating node and shared storage are deployed in same LAN.Additionally, for ease of management and application, can be by management node and N3Individual calculating node deployment is in same HPC (Highperformancecomputing, high-performance calculation) cluster.
Additionally, in a kind of possible implementation, equally will management node, N3Individual calculating node and N2Individual graphics workstation is all deployed in same cluster.In this HPC cluster, owing to graphics workstation can be the high configuration service device being provided with GPU card, therefore at virtualizing graphics work station establishment stage, management node can utilize the arbitrary available graphics workstation in this cluster to complete the foundation of virtualizing graphics work station.On the other hand, when needing the task submitting user to calculate, graphics workstation is also used as calculating node and is managed and call being managed node, thus plays the computing function identical with calculating node.
Step 202: install the management assembly of openstack in management node, be respectively mounted the computation module of openstack in each calculates node and each graphics workstation.
Specifically, openstack is a cloud computing management platform project increased income, and wherein, openstack comprises two kinds of assemblies, for management assembly and computation module.
Therefore, after completing the foundation of above two annexation, the management assembly of openstack can be installed in management node, so that management node holds the function relevant to management, and the computation module of openstack can be installed in calculating node and graphics workstation, holding so that calculating node and calculate relevant function, making graphics workstation hold can virtualized function.
Step 203: unified web user login management system is set in management node, determines N4Individual application mirror image, and by N4Individual application mirror image is sent to share storage.
In detail, with N1、N2、N3Identical, N4It it is equally the natural number not less than 1.
In detail, application mirror image can include the mirror image of graphics workstation, and/or, for building the mirror image of the software of graphics workstation.
Can be on arbitrary local computing for ease of user, according to self user name password with login system, therefore in advance unified web user login management system can be set in management node.By this mode, user can be by the concrete web-link of this system to log in.
On the other hand, corresponding virtualizing graphics work station can be provided for it according to user's actual need, therefore N can be determined in management node in advance for ease of management node4Individual application mirror image, and by N4Individual application mirror image is sent to share storage.Wherein, determined by apply mirror image can be any one in following three kinds of implementations, these three mode is respectively the mirror image of A: graphics workstation, B: for building the mirror image of software of graphics workstation, the mirror image of C: graphics workstation and being used for building the mirror image of the software of graphics workstation.
Step 204: share storage and receive and store the N that management node sends4Individual application mirror image.
Step 205: local computing 1 receives the first notification message of the request graphics workstation that user inputs by logging in web user login management system, and carries corresponding first requirement condition in the first notification message, and is sent to the first notification message manage node.
In detail, first requirement condition can include the mark of graphics workstation, and/or, for building the mark of the software of graphics workstation.
In the present embodiment, requirement condition entrained in the notification message of request graphics workstation, mirror image can be applied corresponding determined by management node.Such as, for implementation A of above-mentioned application mirror image, this requirement condition can be the mark of graphics workstation needed for user.Therefore the operation interface of the virtualizing graphics work station corresponding with this mark can be fed back to local computing 1 by management node.Wherein, operation interface can be a web-link.Specifically, can be in the interface of web user login management system, the mark of whole graphics workstation mirror images in display management node, so that user can click according to self-demand.
Same, for above-mentioned implementation B, this requirement condition can be the mark of the software needed for user in graphics workstation, thus can in above-mentioned interface the mark of whole Software Mirrorings in display management node, in order to user clicks.For above-mentioned implementation C, this requirement condition can be the mark of graphics workstation needed for user, or the mark of corresponding software, therefore can all apply the marks of mirror images in above-mentioned interface in display management node, in order to user clicks.
For example, in high-performance calculation, such as CFD (ComputationalFluidDynamics, computational fluid dynamics) application, the operation of complete set submits to flow process can be pre-treatment, calculating, post processing these three stage.Wherein, pretreatment stage can be that user builds model, and calculation stages can be to calculate the model that system of users builds to verify, post-processing stages is user according to modelling verification result, processes the model built and perfect.Owing to pre-treatment and post-processing stages are required to use substantial amounts of graphics calculations resource, therefore need to operate in graphics workstation.
First, user can be on local computing 1, according to own user name password to log in the web user login management system that management node provides.The most according to the actual requirements, the requirement condition of the required graphics workstation of input, as being provided with flash software, and after input completes, click on " application graphics workstation " button in this system page.Then, local computing 1, when receiving the notification message of request graphics workstation of user's input, can send it to manage node.
Step 206: management node is when receiving the first notification message that local computing 1 sends, it is judged that N2Whether individual graphics workstation exists available graphics workstation, if so, performs step 207, otherwise, according to the reception time of the first notification message, wait that the first notification message is ranked, and perform step 206.
Owing to a graphics workstation can fictionalize multiple virtualizing graphics work station, therefore in the present embodiment, available graphics workstation can include idle graphics workstation and the graphics workstation being not yet fully used.Such as, if a graphics workstation can fictionalize 8 virtualizing graphics work stations, then, when it has fictionalized 7 virtualizing graphics work stations, it is similarly available graphics workstation.
Management node, when receiving the notification message of the request graphics workstation that local computing 1 sends, needs to judge whether to exist in all graphics workstations available graphics workstation.If existing, then can perform follow-up virtualization operations.Accordingly, if not existing, figure workshation resource is described, and supply falls short of demand, temporary shortages, therefore local computing 1 can be utilized to send the time of request according to the reception time of notification message, i.e. user, ranks this notification message etc. pending.After other users complete the use of virtualizing graphics work station, management node can reclaim this virtualizing graphics workshation resource automatically, waits in line the user of state to be supplied to be in.
Such as, when last available graphics workstation resource is occupied, according to the reception time of each notification message, all notification messages received afterwards can be ranked.If the notification message of local computing 1 comes the 10th, then, after needing the most front 9 notification messages to be all successfully processed, again judge whether available graphics workstation.
In a kind of possible implementation, user, after logging in web user login management system, can check the details corresponding with self, if the application submitted to is either with or without processed, the queuing situation etc. of application.
In the present embodiment, management node judges to exist available graphics workstation, respectively graphics workstation 1 and graphics workstation 3.
Step 207: first requirement condition is sent to graphics workstation 3 by management node.
When there is available graphics workstation, the requirement condition carried in notification message can be sent to arbitrary available graphics workstation, such as graphics workstation 3 by management node.
Step 208: graphics workstation 3 obtains the application mirror image corresponding with first requirement condition shared in storage, and fictionalize corresponding virtualizing graphics work station according to this application mirror image, and it is sent to the telework desktop of this virtualizing graphics work station manage node.
Such as, first requirement condition is the mark of flash software, therefore graphics workstation 3 can obtain the mirror image of the corresponding flash software shared in storage.Then, utilize this mirror image, fictionalize the virtualizing graphics work station being provided with flash software, then interface will be operated accordingly, such as telework desktop, feed back to management node.
Step 209: management node receives the telework desktop of the virtualizing graphics work station that graphics workstation 3 sends, and sends it to local computing 1.
In the present embodiment, the telework desktop received is fed back to corresponding local computing 1 by management node, so that the telework desktop that user can be provided by operation local computing 1, with indirect utilization corresponding virtualizing graphics work station.Owing to a graphics workstation can fictionalize multiple virtualizing graphics work station, and user on local computing, can utilize the telework desktop of virtualizing graphics work station to perform task.Therefore, this implementation can reduce the configuration quantity of graphics workstation, reduces cost and puts into.
Performance and the configuration of virtualizing graphics work station are more or less the same with former graphics workstation, it is possible to meet user's practical application request, but its performance is considerably beyond local computing.In the present embodiment, user on local computing, can utilize the telework desktop of the virtualizing graphics work station of management node transmission to perform task, equally reaches to perform, with on height configuration graphics workstation, the effect that task is identical.This implementation not only solves local computing configuration can not meet the problem of demand, and adds the convenience utilizing local computing to perform task.
Step 210: local computing 1 receives the telework desktop that management node sends, so that user performs task, and after task has been performed, receive the second notification message that the task of user's input has performed, and the second notification message carries corresponding second requirement condition.
In detail, any one or more during the second requirement condition may include that the required mark calculating the number of node, the saving contents of task, the preservation title of task and task of task.
In detail, this telework desktop, when receiving the telework desktop of management node feeding back, can be supplied to user by local computing 1.User, on local computing 1, by operating the telework desktop of virtualizing graphics work station, can input a series of control instruction, and this remote virtualization graphics workstation can receive and perform this control instruction, to generate corresponding task.Meanwhile, local computing 1 can show this task by telework desktop.
After task has performed, task is submitted to by user, it is possible to according to the actual requirements, and input calculates the requirement condition needed for this task, such as the number of required calculating node.After input completes, then click on " operation submission " button in the web user login management system page, to send, to local computing 1, the notification message that task has performed.
Such as, the task that user submits to comprises three subfiles, and tasks name is task 1, and subfile title is respectively subfile 1, subfile 2 and subfile 3, these three subfiles are both placed in sharing at the position X of storage, and each subfile is both needed to a calculating node and calculates.Accordingly, the requirement condition that the second notification message carries can comprise: task 1, subfile 1, subfile 2, subfile 3, position X and calculating node demand number are 3.
Step 211: the second notification message received is sent to manage node by local computing 1, and according to the second requirement condition, be sent to task share storage.
Due to user the most according to demand, with information such as the preservation title of input service task, save locations, therefore local computing 1 is after receiving the requirement condition of user's input, on the one hand can be according to this requirement condition, it is sent to the task that user submits to share storage to preserve, on the other hand it is sent to manage node by the notification message that user inputs, so that management node is according to the requirement condition carried in notification message, the corresponding node that calculates of distribution is to calculate the task submitted to.
Step 212: share storage and receive and store the task that local computing 1 sends.
Step 213: management node is when receiving the second notification message that local computing 1 sends, it is judged that N3Whether individual calculating node exists the idle calculating node corresponding with the second requirement condition, if so, performs step 213, otherwise, according to the reception time of the second notification message, wait that the second notification message is ranked, and perform step 212.
It is similar to scheduling of resource and the way to manage of graphics workstation resource with step 206 manages node, management node is when receiving the second notification message that local computing 1 sends, also according to the requirement condition carried in notification message, judge whether to calculate accordingly node resource, if existing, then perform subsequent operation, otherwise, process that the second notification message is ranked.Additionally, after arbitrary occupied calculating node completes the calculating of task, management node can reclaim this calculating node resource automatically.
In this embodiment, in the web user login management system page of each user, title and save location, the disposition of task, the queuing situation etc. of task of the task that user submits to can be shown.
Such as, management node judges to exist 10 idle calculating nodes, then can be the most optional 3, as calculated node 1, calculating node 2 and calculate node 3, as target computing nodes.And management node can distribute calculating node 1 for performing subfile 1, calculating node 2 for performing subfile 2, calculating node 3 for performing subfile 3.
Step 214: the second requirement condition is sent to the idle target computing nodes corresponding with this second requirement condition by management node.
When managing node and judging the calculating node that there is the condition of meeting the requirements, this requirement condition can be sent to calculate accordingly node, so that calculating node is according to this requirement condition, go to obtain and share the relevant work task preserved in storage.
Step 215: target computing nodes receives the second requirement condition that management node sends, and obtains the task corresponding with the second requirement condition shared in storage.
Such as, calculate node 1, calculate node 2 and calculate the node 3 task distribution according to management node, obtain subfile 1, subfile 2 and the subfile 3 at the position X shared in storage respectively.
Step 216: the task that target computing nodes is got by calculating, generate corresponding result of calculation, and according to the second requirement condition, result of calculation is sent to share storage, and the third notice message calculated by corresponding task is sent to manage node.
This task, after getting corresponding task, is calculated, and generates corresponding result of calculation by each calculating node.Such as, when task be user set up when designing a model, calculate whether node exists exception by calculating verifying that this designs a model, and generate the result of calculation comprising abnormal information.After calculating completes, on the one hand can be sent to the result of calculation of generation share storage, send for representing the notification message calculated on the other hand can to management node.
Step 217: share storage and receive and store the result of calculation that target computing nodes sends.
Step 218: management node receives the third notice message that the task sent by target computing nodes has calculated, and third notice message is sent to local computing 1, and performs step 205.
Management node receive calculate node send after representing the notification message calculated, this notification message can be transmitted to corresponding local computing, so that user can view corresponding information after login system, thus reach to notify the effect of user.In addition, management node can also predefine the contact method of each user, such as mailbox, after management node gets the notification message that any operative task computation completes, may determine that the user corresponding with this task, and by corresponding contact method directly to notify user.
CFD application as above, user can apply for graphics workstation in pretreatment stage, and utilize the virtualizing graphics work station of management node feeding back to perform task.Management node combines and calculates node, can calculate the task that user submits in calculation stages.In post-processing stages, task can be carried out perfect according to the result of calculation generated in calculation stages by user, and implement step and be similarly user and apply for graphics workstation, so that management node feeding back corresponding virtualizing graphics work station.Owing to the specific implementation of post-processing stages is identical with pretreatment stage, therefore this is not described in detail by this law embodiment.It is apparent that after post-processing stages completes, it is the same that user equally continues same pretreatment stage, again submit amended task to, again to perform above-mentioned calculation stages, so circulate, until task is successfully completed.
In the present embodiment, graphics workstation resource and calculating node resource can be managed collectively and dispatch by management node, and complete to carry out afterwards the most automatically reclaiming in the utilization of resources, this process, without artificially coordinating to participate in, facilitates user's Appropriate application to each resource.Meanwhile, corresponding resource and service can be provided to it according to the particular requirement condition of each user, therefore the practicality using graphics workstation can be improved.
As it is shown on figure 3, embodiments provide a kind of management node 30, including:
Unit 301 is set, for setting up described management node 30 and at least one external user side apparatus and the annexation with at least one external graphics work station, and determines that at least one applies mirror image；
Receive unit 302, for receiving the first notification message of the request graphics workstation that external user side apparatus sends, and described first notification message carries corresponding first requirement condition, and trigger judging unit 303；
Described judging unit 303, is used for judging whether to exist at least one external graphics work station described available graphics workstation, if so, triggers processing unit 304；
Described processing unit 304, is used for according at least one application mirror image described, and arbitrary described available graphics workstation, it is thus achieved that the operation interface of the virtualizing graphics work station corresponding with described first requirement condition；And the operation interface of described virtualizing graphics work station is sent to described external user side apparatus.
In an embodiment of the invention, described unit 301 is set, is further used for being sent at least one application mirror image described outside sharing storage；
Described processing unit 304, specifically for the either objective graphics workstation described first requirement condition being sent in described available graphics workstation；And after the application mirror image corresponding with described first requirement condition utilizing described outside to share in storage at described targeted graphical work station fictionalizes respective virtual graphics workstation, receive the operation interface of the described virtualizing graphics work station sent by described targeted graphical work station.
In an embodiment of the invention, refer to Fig. 4, in described management node 30,
Described unit 301 is set, is further used for the annexation setting up described management node 30 with at least one external computing nodes；
Described reception unit 302, it is further used for utilizing the operation interface of described external user side apparatus and described virtualizing graphics work station after completing corresponding task user, receive the second notification message that the task of described external user side apparatus transmission has performed, and described second notification message carries corresponding second requirement condition；
Described judging unit 303, it is further used for judging whether at least one external computing nodes described exists the idle calculating node corresponding with described second requirement condition, if, trigger described processing unit 304, otherwise, according to the reception time of described second notification message, wait of ranking described second notification message, triggers described judging unit 303；
Described processing unit 304, is further used for described second requirement condition is sent to the described idle calculating node corresponding with described second requirement condition；Receive the third notice message calculated by the idle task calculating node transmission that described and described second requirement condition is corresponding；And described third notice message is sent to described external user side apparatus, and trigger described reception unit 302.
As it is shown in figure 5, embodiments provide a kind of system using graphics workstation, including:
Any of the above-described described management node 30, at least one user side equipment 40 and at least one graphics workstation 50, wherein,
Described management node 30 maintains the connection with user side equipment each described 40 and with graphics workstation each described 50 respectively；
Described user side equipment 40, for receiving the first notification message of the request graphics workstation of input, and carries corresponding first requirement condition, and described first notification message is sent to described management node 30 in described first notification message；Receive the operation interface of the virtualizing graphics work station corresponding with described first requirement condition that described management node 30 sends；
Described graphics workstation 50, for the operation interface of the virtualizing graphics work station corresponding with described first requirement condition to the offer of described management node 30.
In an embodiment of the invention, refer to Fig. 6, this system can also include:
At least one calculating node 60 and shared storage 70, wherein,
Described calculating node 60, for receiving the second requirement condition that described management node 30 sends；Obtain the task corresponding with described second requirement condition in described shared storage 70；Calculate the described task got, generate corresponding result of calculation；According to described second requirement condition, corresponding described result of calculation is sent to described shared storage 70, and the third notice message calculated by corresponding task is sent to described management node 30；
Described shared storage 70, for receiving and store at least one the application mirror image sent by described management node 30；Receive and store the task that each described user side equipment 40 sends；Receive and store the result of calculation that each described calculating node 60 sends；
Described graphics workstation 50, the first requirement condition sent by described management node 30 specifically for reception；Obtain the intended application mirror image corresponding with described first requirement condition in described shared storage 70；Fictionalize the virtualizing graphics work station corresponding with described intended application mirror image；It is sent to the operation interface of described virtualizing graphics work station manage node 30；Received, by described virtualizing graphics work station, the control instruction that user inputs by operating described operation interface on described user side equipment 40, and by performing described control instruction, generate corresponding task；
Described user side equipment 40, is further used for the operation interface of the described virtualizing graphics work station received is supplied to user；And by the task of virtualizing graphics work station generation described in described operation interface display.
In detail, described shared storage 70 maintains the connection with described management node 30, each described user side equipment 40, each described graphics workstation 50, each described calculating node 60 respectively；Described management node 30 maintains the connection with described shared storage 70, each described user side equipment 40, each described graphics workstation 50, each described calculating node 60 respectively.
By this implementation, user side equipment, such as the local computing of common configuration, after receiving the operation interface of virtualizing graphics work station of management node feeding back, this operation interface can be supplied to user.User, by using the operation interface of virtualizing graphics work station on local computing, equally reaches the effect being similar to directly use graphics workstation, and this mode can facilitate user to perform daily task, and less demanding to local computing configuration degree.
In an embodiment of the invention, maintain the connection described in may include that by described management node 30, at least one user side equipment 40 described, at least one graphics workstation 50 described, described at least one calculate node 60 and described shared storage 70 and be deployed in same LAN；
Described management node 30 and at least one calculating node 60 described are deployed in same cluster, or, described management node 30, at least one calculating node 60 described and at least one graphics workstation 50 described are deployed in same cluster.
In the present embodiment, for the deployment of HPCC, graphics workstation can be deployed in outside cluster, it is also possible to be deployed in cluster.And when graphical Work stands in cluster, graphics workstation on the one hand can be as the graphics workstation being physically present, to fictionalize corresponding multiple virtualizing graphics work station, on the other hand, can also serve as calculating node to use, the task identical with calculating node can be played and calculate effect.This clustered deploy(ment) mode can according to practical application request unrestricted choice.
In an embodiment of the invention, the management assembly of openstack can be installed in described management node 30 in advance, calculating node each described 60 and each described graphics workstation 50 are respectively mounted the computation module of openstack.
The contents such as the information between each unit in said apparatus is mutual, execution process, due to the inventive method embodiment based on same design, particular content can be found in the narration in the inventive method embodiment, and here is omitted.
In sum, each embodiment of the present invention at least has the advantages that
1, in the embodiment of the present invention, management node is after receiving the graphics workstation request notice carrying particular requirement that external user side apparatus sends, it is first determined whether there is available graphics workstation, if existing, according to the application mirror image predefined and arbitrary available graphics workstation, obtain the operation interface of the virtual workstation corresponding with described particular requirement, and this operation interface is fed back to described external user side apparatus.Owing to can utilize management node that whole graphics workstations carry out unified condition managing and resource distribution, and provide corresponding virtualizing graphics workshation resource according to the particular requirement of user for it, and user can utilize the operation interface of the virtual workstation received on the local device, operate to perform task to remote dummy work station, therefore, the embodiment of the present invention can improve the coordination ability to graphics workstation resource.
2, in the embodiment of the present invention, owing to a graphics workstation can fictionalize multiple virtualizing graphics work station, and user on local computing, can utilize the telework desktop of virtualizing graphics work station to perform task.Therefore, this implementation can reduce the configuration quantity of graphics workstation, reduces cost and puts into.
3, in the embodiment of the present invention, performance and the configuration of virtualizing graphics work station are more or less the same with former graphics workstation, it is possible to meet user's practical application request, but its performance is considerably beyond local computing.In the present embodiment, user on local computing, can utilize the telework desktop of the virtualizing graphics work station of management node transmission to perform task, equally reaches to perform, with on height configuration graphics workstation, the effect that task is identical.This implementation not only solves local computing configuration can not meet the problem of demand, and adds the convenience utilizing local computing to perform task.
4, in the embodiment of the present invention, graphics workstation resource and calculating node resource can be managed collectively and dispatch by management node, and complete to carry out afterwards the most automatically reclaiming in the utilization of resources, this process, without artificially coordinating to participate in, facilitates user's Appropriate application to each resource.Meanwhile, corresponding resource and service can be provided to it according to the particular requirement condition of each user, therefore the practicality using graphics workstation can be improved.
It should be noted that, in this article, the relational terms of such as first and second etc is used merely to separate an entity or operation with another entity or operating space, and not necessarily requires or imply the relation or sequentially that there is any this reality between these entities or operation.And, term " includes ", " comprising " or its any other variant are intended to comprising of nonexcludability, so that include that the process of a series of key element, method, article or equipment not only include those key elements, but also include other key elements being not expressly set out, or also include the key element intrinsic for this process, method, article or equipment.In the case of there is no more restriction, statement the key element " including a 〃 " and limiting, it is not excluded that there is also other same factor in including the process of described key element, method, article or equipment.
One of ordinary skill in the art will appreciate that: all or part of step realizing said method embodiment can be completed by the hardware that programmed instruction is relevant, aforesaid program can be stored in the storage medium of embodied on computer readable, this program upon execution, performs to include the step of said method embodiment；And aforesaid storage medium includes: in the various media that can store program code such as ROM, RAM, magnetic disc or CD.
Last it should be understood that the foregoing is only presently preferred embodiments of the present invention, it is merely to illustrate technical scheme, is not intended to limit protection scope of the present invention.All any modification, equivalent substitution and improvement etc. done within the spirit and principles in the present invention, are all contained in protection scope of the present invention.
Claims (10)
1. the method using graphics workstation, it is characterized in that, be applied to manage node, pre-build described management node and at least one external user side apparatus and the annexation with at least one external graphics work station, determine that at least one applies mirror image, also include:
S1: receive the first notification message of the request graphics workstation that external user side apparatus sends, and carry corresponding first requirement condition in described first notification message；
S2: judge whether there is available graphics workstation at least one external graphics work station described, if so, perform S3；
S3: according at least one application mirror image described, and arbitrary described available graphics workstation, it is thus achieved that the operation interface of the virtualizing graphics work station corresponding with described first requirement condition；
S4: the operation interface of described virtualizing graphics work station is sent to described external user side apparatus.
Method the most according to claim 1, it is characterised in that
Described determine at least one application mirror image after, farther include: be sent at least one application mirror image described outside share storage；
Described S3, including:
Described first requirement condition is sent to the either objective graphics workstation in described available graphics workstation；
After the application mirror image corresponding with described first requirement condition utilizing described outside to share in storage at described targeted graphical work station fictionalizes respective virtual graphics workstation, receive the operation interface of the described virtualizing graphics work station sent by described targeted graphical work station；
And/or,
Described application mirror image includes: the mirror image of graphics workstation, and/or, for building the mirror image of the software of graphics workstation；
Described first requirement condition includes: the mark of graphics workstation, and/or, for building the mark of the software of graphics workstation；
And/or,
Described operation interface includes: telework desktop.
Method the most according to claim 1, it is characterised in that
Before S1, farther include: pre-build the annexation of described management node and at least one external computing nodes；
After S4, farther include:
A1: utilize the operation interface of described external user side apparatus and described virtualizing graphics work station after completing corresponding task user, receive the second notification message that the task of described external user side apparatus transmission has performed, and described second notification message carries corresponding second requirement condition；
A2: judge whether to exist at least one external computing nodes described the idle calculating node corresponding with described second requirement condition, if so, perform A3, otherwise, according to the reception time of described second notification message, wait that described second notification message is ranked, and perform A2；
A3: described second requirement condition is sent to the described idle calculating node corresponding with described second requirement condition；
A4: receive the third notice message calculated by the idle task calculating node transmission that described and described second requirement condition is corresponding；
A5: described third notice message is sent to described external user side apparatus, and performs S1.
Method the most according to claim 3, it is characterised in that described second requirement condition includes: task is required calculates any one or more in the mark of the number of node, the saving contents of task, the preservation title of task and task.
5. according to described method arbitrary in Claims 1-4, it is characterised in that
Before S1, farther include: the management assembly of openstack is installed in advance in described management node, and unified web user login management system is set；
And/or,
When judging that in S2 at least one external graphics work station described does not exists available graphics workstation, farther include: according to the reception time of described first notification message, wait that described first notification message is ranked, and perform S2.
6. a management node, it is characterised in that including:
Unit is set, for setting up described management node and at least one external user side apparatus and the annexation with at least one external graphics work station, and determines that at least one applies mirror image；
Receive unit, for receiving the first notification message of the request graphics workstation that external user side apparatus sends, and described first notification message carries corresponding first requirement condition, and trigger judging unit；
Described judging unit, is used for judging whether there is available graphics workstation at least one external graphics work station described, if so, triggers processing unit；
Described processing unit, is used for according at least one application mirror image described, and arbitrary described available graphics workstation, it is thus achieved that the operation interface of the virtualizing graphics work station corresponding with described first requirement condition；And the operation interface of described virtualizing graphics work station is sent to described external user side apparatus.
Management node the most according to claim 6, it is characterised in that
Described unit is set, is further used for being sent at least one application mirror image described outside sharing storage；
Described processing unit, specifically for the either objective graphics workstation described first requirement condition being sent in described available graphics workstation；And after the application mirror image corresponding with described first requirement condition utilizing described outside to share in storage at described targeted graphical work station fictionalizes respective virtual graphics workstation, receive the operation interface of the described virtualizing graphics work station sent by described targeted graphical work station；
And/or,
Described unit is set, is further used for the annexation setting up described management node with at least one external computing nodes；
Described reception unit, it is further used for utilizing the operation interface of described external user side apparatus and described virtualizing graphics work station after completing corresponding task user, receive the second notification message that the task of described external user side apparatus transmission has performed, and described second notification message carries corresponding second requirement condition；
Described judging unit, it is further used for judging whether at least one external computing nodes described exists the idle calculating node corresponding with described second requirement condition, if, trigger described processing unit, otherwise, according to the reception time of described second notification message, wait of ranking described second notification message, triggers described judging unit；
Described processing unit, is further used for described second requirement condition is sent to the described idle calculating node corresponding with described second requirement condition；Receive the third notice message calculated by the idle task calculating node transmission that described and described second requirement condition is corresponding；And described third notice message is sent to described external user side apparatus, and trigger described reception unit.
8. the system using graphics workstation, it is characterised in that including:
Management node, at least one user side equipment and at least one graphics workstation as described in arbitrary in claim 6 and 7, wherein,
Described management node maintains the connection with user side equipment each described and with graphics workstation each described respectively；
Described user side equipment, for receiving the first notification message of the request graphics workstation of input, and carries corresponding first requirement condition, and described first notification message is sent to described management node in described first notification message；Receive the operation interface of the virtualizing graphics work station corresponding with described first requirement condition that described management node sends；
Described graphics workstation, for the operation interface of the virtualizing graphics work station corresponding with described first requirement condition to the offer of described management node.
The system of use graphics workstation the most according to claim 8, it is characterised in that
Also include: at least one calculates node and shares storage, wherein,
Described calculating node, for receiving the second requirement condition that described management node sends；Obtain the task corresponding with described second requirement condition in described shared storage；Calculate the described task got, generate corresponding result of calculation；According to described second requirement condition, corresponding described result of calculation is sent to described shared storage, and the third notice message calculated by corresponding task is sent to described management node；
Described shared storage, for receiving and store at least one the application mirror image sent by described management node；Receive and store the task that each described user side equipment sends；Receive and store the result of calculation that each described calculating node sends；
Described graphics workstation, specifically for receiving the first requirement condition sent by described management node；Obtain the intended application mirror image corresponding with described first requirement condition in described shared storage；Fictionalize the virtualizing graphics work station corresponding with described intended application mirror image；It is sent to the operation interface of described virtualizing graphics work station manage node；Received, by described virtualizing graphics work station, the control instruction that user inputs by operating described operation interface on described user side equipment, and by performing described control instruction, generate corresponding task；
Described user side equipment, is further used for the operation interface of the described virtualizing graphics work station received is supplied to user；And by the task of virtualizing graphics work station generation described in described operation interface display.
The system of use graphics workstation the most according to claim 9, it is characterised in that
Described maintaining the connection includes: described management node, at least one user side equipment described, at least one graphics workstation described, at least one calculating node described and described shared storage are deployed in same LAN；
Described management node and described at least one calculate node deployment in same cluster, or, described management node, described at least one calculate node and at least one graphics workstation described and be deployed in same cluster；
And/or,
Farther include: the management assembly of openstack is installed in advance in described management node, calculating node each described and each described graphics workstation are respectively mounted the computation module of openstack.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610206126.3A CN105824704A (en) | 2016-04-05 | 2016-04-05 | Method, apparatus and system for using graphic workstations |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610206126.3A CN105824704A (en) | 2016-04-05 | 2016-04-05 | Method, apparatus and system for using graphic workstations |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105824704A true CN105824704A (en) | 2016-08-03 |
Family
ID=56526592
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610206126.3A Pending CN105824704A (en) | 2016-04-05 | 2016-04-05 | Method, apparatus and system for using graphic workstations |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105824704A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107515760A (en) * | 2016-06-16 | 2017-12-26 | 上海宽带技术及应用工程研究中心 | A kind of OpenStack multinodes automation installation method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103491188A (en) * | 2013-09-30 | 2014-01-01 | 上海沃帆信息科技有限公司 | Implementation method for enabling multiple users to share graphic workstation by using virtual desktop and GPU pass-through |
| CN104065716A (en) * | 2014-06-18 | 2014-09-24 | 江苏物联网研究发展中心 | OpenStack based Hadoop service providing method |
| CN104333607A (en) * | 2014-11-25 | 2015-02-04 | 武汉理工数字传播工程有限公司 | Device and method for realizing mass transmission of virtual cloud graphic workstation |
| CN105242957A (en) * | 2015-09-28 | 2016-01-13 | 广州云晫信息科技有限公司 | Method and system for cloud computing system to allocate GPU resources to virtual machine |
-
2016
- 2016-04-05 CN CN201610206126.3A patent/CN105824704A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103491188A (en) * | 2013-09-30 | 2014-01-01 | 上海沃帆信息科技有限公司 | Implementation method for enabling multiple users to share graphic workstation by using virtual desktop and GPU pass-through |
| CN104065716A (en) * | 2014-06-18 | 2014-09-24 | 江苏物联网研究发展中心 | OpenStack based Hadoop service providing method |
| CN104333607A (en) * | 2014-11-25 | 2015-02-04 | 武汉理工数字传播工程有限公司 | Device and method for realizing mass transmission of virtual cloud graphic workstation |
| CN105242957A (en) * | 2015-09-28 | 2016-01-13 | 广州云晫信息科技有限公司 | Method and system for cloud computing system to allocate GPU resources to virtual machine |
Non-Patent Citations (1)
| Title |
|---|
| 刘袆阳等: "远程可视化平台研究与实现", 《高性能计算发展与应用》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107515760A (en) * | 2016-06-16 | 2017-12-26 | 上海宽带技术及应用工程研究中心 | A kind of OpenStack multinodes automation installation method and system |
| CN107515760B (en) * | 2016-06-16 | 2020-10-16 | 上海宽带技术及应用工程研究中心 | OpenStack multi-node automatic installation method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8656387B2 (en) | Method and system for workload distributing and processing across a network of replicated virtual machines | |
| CN112346859B (en) | Resource scheduling method and device, electronic equipment and storage medium | |
| CN104601664B (en) | A kind of control system of cloud computing platform resource management and scheduling virtual machine | |
| CN105979007A (en) | Acceleration resource processing method and device and network function virtualization system | |
| CN103795804A (en) | Storage resource scheduling method and storage calculation system | |
| CN103986786A (en) | Remote cloud desktop operation system | |
| CN105308553B (en) | Dynamic provides storage | |
| CN112104723A (en) | Multi-cluster data processing system and method | |
| CN104615487A (en) | System and method for optimizing parallel tasks | |
| CN112084015B (en) | Cloud computing-based simulation cloud platform building system and method | |
| CN104410699A (en) | Resource management method and system of open type cloud computing | |
| CN110806928A (en) | Job submitting method and system | |
| CN105933154A (en) | Management method of cloud calculation resources | |
| CN110532060A (en) | A kind of hybrid network environmental data collecting method and system | |
| CN114489989A (en) | Method and system for parallel scheduling based on proxy client | |
| CN104735134B (en) | A kind of method and apparatus serviced for providing calculating | |
| CN105824704A (en) | Method, apparatus and system for using graphic workstations | |
| US11755379B2 (en) | Liaison system and method for cloud computing environment | |
| CN115361382B (en) | Data processing method, device, equipment and storage medium based on data group | |
| CN112445602A (en) | Resource scheduling method, device and system and electronic equipment | |
| CN105933136A (en) | Resource scheduling method and system | |
| CN116204307A (en) | Federal learning method and federal learning system compatible with different computing frameworks | |
| CN112181403B (en) | Development operation and maintenance integrated implementation method, device, equipment and readable storage medium | |
| CN103631648A (en) | Task processing method and system | |
| CN106055410A (en) | Cloud computing memory resource allocation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160803 |
|
| RJ01 | Rejection of invention patent application after publication |