CN108710543A - A kind of processing method and equipment of rendering task - Google Patents
A kind of processing method and equipment of rendering task Download PDFInfo
- Publication number
- CN108710543A CN108710543A CN201810488258.9A CN201810488258A CN108710543A CN 108710543 A CN108710543 A CN 108710543A CN 201810488258 A CN201810488258 A CN 201810488258A CN 108710543 A CN108710543 A CN 108710543A
- Authority
- CN
- China
- Prior art keywords
- rendering
- machine
- task
- itself
- main
- 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
- G06F9/54—Interprogram communication
- G06F9/545—Interprogram communication where tasks reside in different layers, e.g. user- and kernel-space
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Processing Or Creating Images (AREA)
Abstract
This application discloses a kind of processing methods of rendering task, including main rendering machine to obtain and execute number of nodes and itself idle node number needed for complete rendering task；And judge whether idle node number is less than required number of nodes；If so, the corresponding first part's rendering task of idle node number beyond main rendering machine itself is forwarded to first from the processing of rendering machine；First judges whether the idle node number of itself is less than the number of nodes executed needed for first part's rendering task from rendering machine；If so, by being forwarded to second from the processing of rendering machine, until the distribution of complete rendering task finishes from the corresponding second part rendering task of idle node number of rendering machine itself beyond first.Larger rendering task can successively be decomposed rendering by this method, have higher resource rendering capability and autgmentability, effectively realized huge multiple signals output, further improved resource utilization and its working efficiency.Disclosed herein as well is a kind of processing equipments of rendering task, it may have above-mentioned advantageous effect.
Description
Technical field
This application involves network technique field, more particularly to a kind of processing method of rendering task further relates to a kind of rendering
The processing equipment of task.
Background technology
After the completion of the structure that designs a model, it will usually be rendered to it using computer, its practical effect is shown to reach
The purpose of fruit, wherein Rendering is substantially exactly the calculating and output to graph data.With the development of video card capabilities, firmly
The raising of part speed, requirement of the related industry to graphics process are also being continuously improved, and Real-time Rendering Technology is more and more by correlation
Industry being pursued and paying close attention to.
Currently, in all kinds of rendering systems, with the continuous improvement that related industry requires graphics process, single computer
Rendering capability limit the development of Rendering and caused due to the limitation of its own hardware processing capability and display capabilities
Its rendering capability is limited, to cannot achieve the merging to multichannel input signal output, i.e., can not to larger rendering task into
Row processing, or can complete the processing task and also be unable to reach expected rendering effect, this problem resolution requirement compared with
It is particularly evident in the case that high and rendering content is more.In addition, realizing the processing of rendering task by single computer, it is not present
Thus any scalability also results in extremely low rendering efficiency.
Therefore, how the resource rendering capability and autgmentability of lifting means, to realize that huge multiple signals export, into one
It is those skilled in the art's urgent problem to be solved that step, which promotes resource utilization and its working efficiency,.
Invention content
The purpose of the application is to provide a kind of processing method of rendering task, which can be by larger rendering task
It successively decomposes and renders, there is higher resource rendering capability and autgmentability, huge multiple signals output is effectively realized, into one
Step improves resource utilization and its working efficiency；The another object of the application is to provide a kind of processing equipment of rendering task,
Also there is above-mentioned advantageous effect.
In order to solve the above technical problems, this application provides a kind of processing method of rendering task, the processing method packet
It includes：
Main rendering machine obtains the idle node number for executing number of nodes and itself needed for complete rendering task；Wherein, described
Complete rendering task is handed down to the main rendering machine by upper control node；
The main rendering machine judges whether the idle node number of itself is less than the section executed needed for the complete rendering task
Points；If so, the corresponding first part's rendering task of idle node number beyond the main rendering machine itself is forwarded to
One is handled from rendering machine；
Described first judges whether the idle node number of itself is less than from rendering machine executes first part's rendering task
Required number of nodes；If so, by being rendered from the corresponding second part of idle node number for rendering machine itself beyond described first
Task is forwarded to second and is handled from rendering machine, until all distribution finishes the complete rendering task.
Preferably, described to turn the corresponding first part's rendering task of idle node number beyond the main rendering machine itself
First is sent to be handled from rendering machine, including：
It calculates and executes the number of nodes needed for the complete rendering task and the idle node number of the main rendering machine itself
Difference obtains number of nodes difference；
The corresponding first part's rendering task of the number of nodes difference is forwarded to described first from wash with watercolours by the main rendering machine
Dye machine is handled.
Preferably, described to calculate the sky for executing number of nodes and the main rendering machine itself needed for the complete rendering task
The difference of not busy number of nodes, obtain number of nodes difference before, further include：
Obtain the total node number of the main rendering machine；Wherein, the total node number is that the idle node number is saved with occupancy
The sum of points；
The occupation proportion upper limit of the total node number is set for the main rendering machine；
Idle node can be occupied by being calculated according to the idle node number of the main rendering machine and the occupation proportion upper limit
Number obtains institute to calculate the number of nodes and the difference for occupying idle node number that execute needed for the complete rendering task
State number of nodes difference.
Preferably, the processing method of the rendering task further includes：
The upper control node is by all rendering machines in such a way that the idle node quantity of itself is successively successively decreased downwards
It is arranged.
Preferably, the processing method of the rendering task further includes：
Each part rendering knot that after the part rendering task that rendering machine has executed that itself is obtained, itself will be obtained in real time
Fruit is back to the main rendering machine by the way of successively upward；
The part rendering result is merged acquisition rendering result by the main rendering machine, and the rendering result is real
When be uploaded to the upper control node.
Preferably, the processing method of the rendering task further includes：
After each rendering machine has executed the rendering task of itself acquisition, the status information of itself is back to the upper layer and is controlled
Node processed；
The upper control node discharges the occupancy resource of the corresponding rendering machine of the status information.
Preferably, the processing method of the rendering task further includes：
The upper control node records the idle node number of all rendering machines itself.
In order to solve the above technical problems, this application provides a kind of processing equipment of rendering task, the processing equipment packet
It includes：
Upper control node, for issuing complete rendering task to main rendering machine；
The main rendering machine, for obtaining the idle node for executing number of nodes and itself needed for the complete rendering task
Number；And judge whether the idle node number of itself is less than the number of nodes executed needed for the complete rendering task；If so, will surpass
Go out the main rendering machine itself the corresponding first part's rendering task of idle node number be forwarded to first carried out from rendering machine from
Reason；
Described first executes first part's rendering from the machine of rendering, for judging whether the idle node number of itself is less than
The number of nodes of required by task；If so, by beyond described first from render machine itself the corresponding second part of idle node number
Rendering task is forwarded to second and is handled from rendering machine, until all distribution finishes the complete rendering task.
Preferably, the upper control node be additionally operable to by all rendering machines according to the idle node quantity of itself successively to
Under the mode successively decreased arranged.
Preferably, each to be additionally operable to after having executed the part rendering task that itself is obtained from rendering machine, itself is obtained in real time
The part rendering result obtained is back to the main rendering machine by the way of successively upward；
The main rendering machine is additionally operable to the part rendering result merging acquisition rendering result, and by the rendering
The upper control node is uploaded to when fructufy.
A kind of processing method of rendering task provided herein, including main rendering machine obtain and execute complete rendering task
The idle node number of required number of nodes and itself；Wherein, the complete rendering task is handed down to described by upper control node
Main rendering machine；The main rendering machine judges whether the idle node number of itself is less than the section executed needed for the complete rendering task
Points；If so, the corresponding first part's rendering task of idle node number beyond the main rendering machine itself is forwarded to
One is handled from rendering machine；Described first judges whether the idle node number of itself is less than from rendering machine executes described first
Divide the number of nodes needed for rendering task；If so, by beyond described first from the idle node number corresponding for rendering machine itself
Two part rendering tasks are forwarded to second and are handled from rendering machine, until all distribution finishes the complete rendering task.
As it can be seen that the processing method of rendering task provided herein, the complete rendering task of client input is carried out
Refinement is decomposed, and rendering machine using multilayer executes its own assigned part rendering task, it can be achieved that multichannel input signal
Merge and render output, to realize the processing to larger rendering task；Rendering task compared to single computer is handled, the application
The technical solution provided has broken the limitation of its hardware processing capability and display capabilities, and the machine of rendering is more with tree progress
Layer setting, has better scalability, effectively increases the resource rendering capability of equipment, further improving resource makes
With rate and its working efficiency.
The processing equipment of a kind of rendering task provided herein, it may have above-mentioned advantageous effect, details are not described herein.
Description of the drawings
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of application for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is a kind of flow diagram of the processing method of rendering task provided herein；
Fig. 2 is the design sketch that a kind of rendering task provided herein is successively handled；
Fig. 3 is a kind of structural schematic diagram of the processing equipment of rendering task provided herein.
Specific implementation mode
The core of the application is to provide a kind of processing method of rendering task, and this method can be by larger rendering task successively
It decomposes and renders, there is higher resource rendering capability and autgmentability, effectively realize huge multiple signals output, further carry
Resource utilization and its working efficiency are risen；Another core of the application is to provide a kind of processing equipment of rendering task, also has
There is above-mentioned advantageous effect.
To keep the purpose, technical scheme and advantage of the embodiment of the present application clearer, below in conjunction with the embodiment of the present application
In attached drawing, technical solutions in the embodiments of the present application is clearly and completely described, it is clear that described embodiment is
Some embodiments of the present application, instead of all the embodiments.Based on the embodiment in the application, those of ordinary skill in the art
The all other embodiment obtained without making creative work, shall fall in the protection scope of this application.
Referring to FIG. 1, Fig. 1 is a kind of flow diagram of the processing method of rendering task provided herein, at this
Reason method may include：
S101：Main rendering machine obtains the idle node number for executing number of nodes and itself needed for complete rendering task；Wherein,
Complete rendering task is handed down to main rendering machine by upper control node；
Specifically, client is first by pending rendering task, i.e., above-mentioned complete rendering task is sent to top level control
Node selects a rendering machine as main rendering machine by upper control node, remaining renders machine and is used as from the machine of rendering, and will connect
The complete rendering task received is issued on main rendering machine；Further, main rendering machine can appoint according to the complete rendering received
Business determines the number of nodes and the idle node number of itself executed needed for the complete rendering task, to realize to the complete rendering
The reasonable distribution of task.Wherein, acquisition modes the application of above-mentioned idle node number is not specifically limited, and can such as be calculated by correlation
Method, which calculates, to be obtained, and specifically can refer to prior art, the application no longer illustrates herein.
Wherein, main rendering machine also has both management function other than it can perform the rendering task that itself is received, can real-time monitoring
Each node behaviour in service from the machine of rendering of subordinate is conveyed request between upper control node and each machine from rendering and is answered
It answers, and is responsible for feeding back to control node etc. from the working condition of the machine of rendering by each.
S102：Main rendering machine judges whether the idle node number of itself is less than the node executed needed for complete rendering task
Number；
S103：It, will if the idle node number of main rendering machine itself is less than the number of nodes executed needed for complete rendering task
The corresponding first part's rendering task of idle node number beyond main rendering machine itself is forwarded to first and is handled from rendering machine；
S104：If the idle node number of main rendering machine itself is not less than the number of nodes executed needed for complete rendering task,
Main rendering machine carries out task processing；
Specifically, main rendering machine judges whether the idle node number of itself is less than the node executed needed for complete rendering task
Number, if it is not, it is then other any from rendering machine without occupying, directly complete render can be executed using the idle node of itself appoint
Business；If so, can be by the corresponding rendering task of idle node number beyond main rendering machine itself, i.e., above-mentioned first part, which renders, to appoint
Business, is forwarded to first and is handled from rendering machine.Wherein, this first from render machine quantity can be rendered according to above-mentioned first part
The size of task is set, since main rendering machine is attached with first from rendering machine by the idle node of itself, therefore the
One from render machine quantity without departing from itself idle node number of main rendering machine, the application does not limit this, in addition, institute
First had is uniformly connected to main rendering machine from rendering machine, after first part's rendering task to receive main rendering machine forwarding carries out
Continuous processing.
Preferably, above-mentioned to be forwarded to the corresponding first part's rendering task of idle node number beyond main rendering machine itself
First carries out processing from rendering machine may include：It calculates and executes the number of nodes needed for complete rendering task and main rendering machine itself
The difference of idle node number obtains number of nodes difference；Main rendering machine turns the corresponding first part's rendering task of number of nodes difference
First is sent to be handled from rendering machine.
Specifically, main rendering machine is when being forwarded to first by first part's rendering task and being handled from rendering machine, it can be first
The difference for executing the number of nodes needed for above-mentioned complete rendering task and the idle node number of main rendering machine itself is calculated, to be saved
Points difference, and the corresponding first part's rendering task of the number of nodes difference is forwarded to first and is handled from rendering machine.
Preferably, above-mentioned calculating executes the idle node number of number of nodes and main rendering machine itself needed for complete rendering task
Difference, can also include before obtaining number of nodes difference：Obtain the total node number of main rendering machine；Wherein, total node number is sky
The sum of not busy number of nodes and occupancy number of nodes；Based on render machine setting total node number the occupation proportion upper limit；According to main rendering machine
Idle node number and the occupation proportion upper limit, which are calculated, can occupy idle node number, be executed needed for complete rendering task with calculating
Number of nodes and the difference that can occupy idle node number, obtain number of nodes difference.
Specifically, execute the number of nodes needed for complete rendering task and the idle node number of main rendering machine itself
Before mathematic interpolation, the total node number of main rendering machine can also be further obtained, i.e. the occupancy number of nodes of its own is saved with idle
The summation of points；It is simultaneously the occupation proportion upper limit of its total node number of setting themselves, further, you can according in the occupation proportion
Limit and above-mentioned idle node number, which be calculated, can occupy idle node number, calculated and executed needed for complete rendering task with this
Number of nodes can occupy the difference of idle node number with this, to obtain above-mentioned number of nodes difference.Wherein, the above-mentioned occupation proportion upper limit
Setting easily facilitates realization system redundancy, to increase the reliability of system, for example, the total node number of main rendering machine is 1000,
The occupation proportion upper limit is set as 70%, i.e., wherein 700 nodes are that can occupy number of nodes, remaining 300 are then used as standby node,
Task processing can not be carried out to solve the problems, such as that nodes break down can be occupied individually.
S105：First judges whether the idle node number of itself is less than from rendering machine executes needed for first part's rendering task
Number of nodes；
S106：If first is less than the node executed needed for first part's rendering task from the idle node number for rendering machine itself
Number, then will be forwarded to second from rendering beyond first from the corresponding second part rendering task of idle node number of rendering machine itself
Machine is handled, until all distribution finishes complete rendering task；
S107：If first is not less than the section executed needed for first part's rendering task from the idle node number for rendering machine itself
It counts, then main rendering machine and first carries out task processing from the machine of rendering.
On the basis of step S102 to step S104, first judges whether the idle node number of itself is less than from rendering machine
The number of nodes needed for first part's rendering task is executed, if it is not, it is then other from rendering machine without occupying again, it can directly utilize certainly
The idle node of body executes first part's rendering task；If so, can by beyond first from render machine itself idle node
The corresponding rendering task of number, i.e., above-mentioned second part rendering task are forwarded to second and are handled from rendering machine.Wherein, this
Two can be set from the quantity of the machine of rendering according to the size of above-mentioned second part rendering task, be passed through from rendering machine due to first
The idle node of itself is attached with second from rendering machine, thus second from the quantity of the machine of rendering without departing from first from the machine of rendering
Itself idle node number, the application do not limit this, in addition, all second from rendering machine be uniformly connected to first from
Rendering machine carries out subsequent processing to receive the second part rendering task that first forwards from the machine of rendering.
And so on, until the complete rendering task that main rendering machine receives all is assigned, i.e., whole rendering machines
It can complete the processing of the complete rendering task.Wherein, first from the quantity of the machine of rendering, second from the quantity of the machine of rendering and follow-up
The quantity of each layer from the machine of rendering is independent, can be configured according to actual conditions, whether quantity is identical, not influence this Shen
Please technical solution implementation.
It should be noted that above-mentioned main rendering machine is by the corresponding first part of idle node number beyond main rendering machine itself
Rendering task be forwarded to first from the mode of the machine of rendering be only a kind of embodiment provided herein, not uniquely, for example,
The complete rendering task can be also divided into more one's share of expenses for a joint undertaking rendering tasks by main rendering machine, and will be one-to-one per one's share of expenses for a joint undertaking rendering task
Each first is distributed to from rendering machine, i.e., each first receives an one's share of expenses for a joint undertaking rendering task from rendering machine is handled.
Specifically, for example, the idle node number of main rendering machine itself is N, i.e., maximum connecting node number is N, and it is held
The number of nodes needed for complete rendering task that row receives is more than N, therefore the complete rendering task can be divided into the first son of M parts and rendered
Task, wherein M is the integer less than or equal to N, and every part of first sub- rendering task size can not wait；At this point, top level control section
Point can available free node from rendering machine list based on render machine select most matched M from rendering machine as first from
Rendering machine, i.e., the first layer of main rendering machine is from the machine of rendering.Further, each first judges to execute first received from rendering machine
Sub- rendering task whether exceed itself idle node number, if exceeding, by the first sub- rendering task be separated into P1, P2 ... .PM
Part, and supported to control node acquisition request by the main rendering machine in upper layer, upper control node is in available free node from rendering
In machine list for first selected from rendering machine it is most matched from rendering machine as second from rendering machine, i.e., the second of main rendering machine
Layer from render machine.And so on, the step is repeated until the bottom, such as T layers so that whole rendering machines can be completed just
Complete rendering task, wherein be each all higher than from the processing capacity of the machine of rendering or be equal to input signal all the way, in order to a wash with watercolours
Dye machine can complete the processing to input signal all the way.It will be apparent that the number of plies from the machine of rendering is fewer, i.e., above-mentioned T values are smaller, then
Illustrate that its rendering task is smaller, rendering speed is faster.
The complete rendering task that client inputs is carried out refinement point by the processing method of rendering task provided herein
Solution renders machine using multilayer and executes its own assigned part rendering task, it can be achieved that merging wash with watercolours to multichannel input signal
Dye output, to realize the processing to larger rendering task；Rendering task compared to single computer is handled, provided herein
Technical solution broken the limitations of its hardware processing capability and display capabilities, rendering machine is carried out multilayer with tree to be set
It sets, has better scalability, effectively increase the resource rendering capability of equipment, further improve resource utilization
And its working efficiency.
On the basis of the above embodiments：
As a kind of preferred embodiment, the processing method of above-mentioned rendering task can also include：Upper control node is by institute
There is rendering machine to be arranged in such a way that the idle node quantity of itself is successively successively decreased downwards.
Specifically, to reduce the use of rendering machine, resource utilization is further increased, upper control node can be by each wash with watercolours
Dye machine is arranged according to itself descending sequence of idle node number, and the most rendering machine of idle node quantity is made
Based on render machine, be arranged in order, using the rendering machine of idle node minimum number as the bottom from render machine.
Referring to FIG. 2, Fig. 2 is the design sketch that a kind of rendering task provided herein is successively handled.For example, upper layer is controlled
After node processed arranges rendering machine, the rendering machine list of acquisition is：X[n]={ X0, X1, X2 ... ..Xi ... ..Xn }, i.e., always
Shared n+1 platforms render machine, and the corresponding maximum idle node number of every rendering machine is：{ Y0, Y1, Y2 ... .Yi ... Yn }, and Y0>Y1>
Y2….>Yi…>Yn。
When needing to merge output (m> to the roads m signal;1) when, machine is rendered based on the optional X0 of control node, judges X0
Whether complete rendering task, i.e. m< can be completed alone;Whether Y0 is true, if set up, flow terminates, and illustrates that current X0 can be only
It is vertical to complete the complete rendering task；If invalid, first node that X1 is suspended to X0 is first of first layer from wash with watercolours
Dye machine.
Further, the reserved node of X1 comes out the main rendering machine as spare next layer, judges this two renderings again
Whether machine can complete complete rendering task, that is, judge m<Whether Y1-1+Y0-1 is true, wherein Y0-1 represents this rendering of X0
Machine remaining number of nodes other than meeting X1 all the way, Y1-1 represent X1 and have reserved number of nodes outside a node, if set up, flow
Terminate, illustrates that current this can assist main rendering machine to complete rendering task and X2 is suspended to X0 if invalid from rendering machine
Second node as second of first layer from the machine of rendering.
Further, the reserved node of X2 comes out the main rendering machine as spare next layer, judges this three renderings again
Whether machine can complete complete rendering task, that is, judge m<Whether Y2-1+Y1-1+Y0-2 true, wherein Y0-2 represent X0 this
Rendering machine remaining number of nodes in addition to meeting two-way X1 and X2, if set up, flow terminates, illustrate current two can be with from rendering machine
Main rendering machine is assisted to complete rendering task, if invalid, and so on, until i-th Xi is suspended to the last one section of X0
Point as first layer the last one from rendering machine.
Further, whether judge remaining idle node sum more than i, if no more than if by remaining rendering task again from
X1 starts to be assigned to every from the idle node that rendering machine is reserved in advance, on the contrary then represent this i platform from wash with watercolours until be assigned
Dye machine cannot still assist main rendering machine to complete whole rendering tasks, at this point, the first of first layer terminates from the distribution of rendering machine,
I.e.：X1,X2…Xi.
Based on the above results, then using X1 as the second layer from the main rendering machine of the machine of rendering, remaining rendering machine to be allocated is
{ X (i+1), X (i+2) ... Xn }, with above-mentioned steps be X1, X2 ... Xi select the second layer from the machine of rendering, if rendering task is still
Do not complete, same method choice third layer from the machine of rendering, and so on, until complete rendering task is fully completed.
The method that the present embodiment is provided by each rendering machine according to itself descending sequence of idle node number into
Row arrangement, and using the most rendering machine of idle node quantity as main rendering machine, can effectively reduce the use of rendering machine, go forward side by side one
Step improves the utilization rate of resource.
As a kind of preferred embodiment, the processing method of above-mentioned rendering task can also include：It is each to be executed from the machine of rendering
After the complete part rendering task itself obtained, the part rendering result that itself is obtained is returned by the way of successively upward in real time
To main rendering machine；Part rendering result is merged acquisition rendering result by main rendering machine, and by the rendering result it is real-time on
Reach the upper control node.
Specifically, after rendering task is assigned, the part rendering task that each layer receives itself from rendering machine carries out
Processing, after being disposed, you can the various pieces rendering result respectively obtained is back to master in a manner of successively upward in real time
Sub- rendering result is fed back to layer second from the bottom from rendering machine by rendering machine, the i.e. bottom from rendering machine, second from the bottom layer by layer from wash with watercolours
Dye machine merge complete render after again Real-time Feedback to layer third from the bottom from render machine, and so on successively passback various pieces render
As a result, until various pieces rendering result all returns to main rendering machine, finally again by main rendering machine to the various pieces that receive
Rendering result, which merges, obtains final rendering result, and it is uploaded to control node in real time, and task terminates.
It should be noted that the real-time Transmission mode of above-mentioned rendering result and part rendering result is only that the application is carried
A kind of embodiment supplied uniquely can also not set uplink time interval, i.e. timing realizes that rendering result and part render
As a result transmission, the application do not limit this.
Further, it can also show that equipment is established with upper control node by setting to connect, the wash with watercolours after with display merging
Dye is as a result, efficiently solve the problems, such as that single computer hardware display capabilities are limited.
As a kind of preferred embodiment, the processing method of above-mentioned rendering task can also include：Each rendering machine has executed
After the rendering task that itself is obtained, the status information of itself is back to upper control node；Upper control node release conditions
The occupancy resource of the corresponding rendering machine of information.
Specifically, after each rendering machine has executed the rendering task that itself is obtained, you can return to the status information of itself
To upper control node, to inform that the rendering task that upper control node itself is distributed is completed, further, top level control section
The point i.e. occupancy resource of the corresponding rendering machine of the releasable status information discharges its node space, becomes idle node,
In order to which next complete rendering task can be continued to execute.Wherein, the quantity of the corresponding rendering machine of the status information is not unique, can
Can be separate unit or more, i.e. the status information likely corresponds to separate unit from rendering machine or main rendering machine, it is also possible to correspond to more
From the machine of rendering.
Preferably, the processing method of the rendering task can also include：Upper control node records all rendering machines itself
Idle node number.
Specifically, for ease of executing next complete rendering task, upper control node can also further record each rendering
The idle node number of machine when in order to continue complete rendering task upon receipt, can carry out the complete rendering task rational
Resource allocation improves resource utilization to reduce the waste of idle node.
To solve the above problems, referring to FIG. 3, Fig. 3 is a kind of processing equipment of rendering task provided herein
Schematic diagram, the processing equipment may include：
Upper control node 10, for issuing complete rendering task to main rendering machine 20；
Main rendering machine 20, for obtaining the idle node number for executing number of nodes and itself needed for complete rendering task；And
Judge whether the idle node number of itself is less than the number of nodes executed needed for complete rendering task；If so, main rendering will be exceeded
The corresponding first part's rendering task of idle node number of machine itself is forwarded to first and is handled from rendering machine 30；
First executes rendering task institute of first part from machine 30 is rendered, for judging whether the idle node number of itself is less than
The number of nodes needed；If so, by turning from the corresponding second part rendering task of idle node number for rendering machine itself beyond first
It is sent to second to be handled from rendering machine 40, until all distribution finishes complete rendering task.
As a kind of preferred embodiment, upper control node 10 can also be used to save all rendering machines according to the idle of itself
The mode that point quantity is successively successively decreased downwards is arranged.
It is each to be additionally operable to after having executed the part rendering task that itself is obtained from rendering machine as a kind of preferred embodiment,
The part rendering result that itself is obtained is back to main rendering machine 20 by the way of successively upward in real time；
Main rendering machine 20 is additionally operable to part rendering result merging acquisition rendering result, and by rendering result it is real-time on
Reach upper control node 10.
As a kind of preferred embodiment, after each rendering machine is additionally operable to the rendering task for having executed itself acquisition, by itself
Status information be back to upper control node 10；
Upper control node 10 is additionally operable to the occupancy resource of the corresponding rendering machine of release conditions information.
As a kind of preferred embodiment, upper control node 10 is additionally operable to record the idle node of all rendering machines itself
Number.
Above method embodiment is please referred to for the introduction of equipment provided by the present application, this will not be repeated here by the application.
Each embodiment is described by the way of progressive in specification, the highlights of each of the examples are with other realities
Apply the difference of example, just to refer each other for identical similar portion between each embodiment.For device disclosed in embodiment
Speech, since it is corresponded to the methods disclosed in the examples, so description is fairly simple, related place is referring to method part illustration
.
Professional further appreciates that, unit described in conjunction with the examples disclosed in the embodiments of the present disclosure
And algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is implemented in hardware or software actually, depends on the specific application and design constraint of technical solution.Profession
Technical staff can use different methods to achieve the described function each specific application, but this realization is not answered
Think to exceed scope of the present application.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
The processing method and equipment of rendering task provided herein are described in detail above.It is used herein
The principle and implementation of this application are described for specific case, and the explanation of above example is only intended to help to understand
The present processes and its core concept.It should be pointed out that for those skilled in the art, not departing from this
Apply principle under the premise of, can also to the application, some improvement and modification can also be carried out, these improvement and modification also fall into the application
Scope of the claims element.
Claims (10)
1. a kind of processing method of rendering task, which is characterized in that including：
Main rendering machine obtains the idle node number for executing number of nodes and itself needed for complete rendering task；Wherein, described complete
Rendering task is handed down to the main rendering machine by upper control node；
The main rendering machine judges whether the idle node number of itself is less than the number of nodes executed needed for the complete rendering task；
If so, the corresponding first part's rendering task of idle node number beyond the main rendering machine itself is forwarded to first from wash with watercolours
Dye machine is handled；
Described first judges whether the idle node number of itself is less than from rendering machine executes needed for first part's rendering task
Number of nodes；If so, by beyond described first from render machine itself the corresponding second part rendering task of idle node number
It is forwarded to second to be handled from rendering machine, until all distribution finishes the complete rendering task.
2. processing method as described in claim 1, which is characterized in that described to save the free time beyond the main rendering machine itself
Corresponding first part's rendering task of counting is forwarded to first and is handled from rendering machine, including：
The difference for executing the number of nodes needed for the complete rendering task and the idle node number of the main rendering machine itself is calculated,
Obtain number of nodes difference；
The corresponding first part's rendering task of the number of nodes difference is forwarded to described first from rendering machine by the main rendering machine
It is handled.
3. processing method as claimed in claim 2, which is characterized in that the calculating executes needed for the complete rendering task
The difference of the idle node number of number of nodes and the main rendering machine itself further includes before obtaining number of nodes difference：
Obtain the total node number of the main rendering machine；Wherein, the total node number is the idle node number and occupancy number of nodes
The sum of；
The occupation proportion upper limit of the total node number is set for the main rendering machine；
Idle node number can be occupied by being calculated according to the idle node number of the main rendering machine and the occupation proportion upper limit, with
Number of nodes and the difference for occupying idle node number executed needed for the complete rendering task is calculated, the node is obtained
Number difference.
4. the processing method as described in claims 1 to 3 any one, which is characterized in that further include：
The upper control node carries out all rendering machines in such a way that the idle node quantity of itself is successively successively decreased downwards
Arrangement.
5. processing method as claimed in claim 4, which is characterized in that further include：
It is each from rendering machine executed itself obtain part rendering task after, in real time by itself obtain part rendering result adopt
It is back to the main rendering machine with successively upward mode；
The part rendering result is merged acquisition rendering result by the main rendering machine, and by the rendering result it is real-time on
Reach the upper control node.
6. processing method as claimed in claim 5, which is characterized in that further include：
After each rendering machine has executed the rendering task of itself acquisition, the status information of itself is back to the top level control section
Point；
The upper control node discharges the occupancy resource of the corresponding rendering machine of the status information.
7. processing method as claimed in claim 6, which is characterized in that further include：
The upper control node records the idle node number of all rendering machines itself.
8. a kind of processing equipment of rendering task, which is characterized in that including：
Upper control node, for issuing complete rendering task to main rendering machine；
The main rendering machine, for obtaining the idle node number for executing number of nodes and itself needed for the complete rendering task；
And judge whether the idle node number of itself is less than the number of nodes executed needed for the complete rendering task；If so, will exceed
The corresponding first part's rendering task of idle node number of the main rendering machine itself is forwarded to first and is handled from rendering machine；
Described first executes first part's rendering task from the machine of rendering, for judging whether the idle node number of itself is less than
Required number of nodes；If so, by being rendered from the corresponding second part of idle node number for rendering machine itself beyond described first
Task is forwarded to second and is handled from rendering machine, until all distribution finishes the complete rendering task.
9. processing equipment as claimed in claim 8, which is characterized in that the upper control node is additionally operable to all rendering machines
It is arranged in such a way that the idle node quantity of itself is successively successively decreased downwards.
10. processing equipment as claimed in claim 9, which is characterized in that each to be additionally operable to execute itself acquisition from rendering machine
Part rendering task after, in real time by itself obtain part rendering result the main wash with watercolours is back to by the way of successively upward
Dye machine；
The main rendering machine is additionally operable to the part rendering result merging acquisition rendering result, and by the rendering result
It is uploaded to the upper control node in real time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810488258.9A CN108710543A (en) | 2018-05-21 | 2018-05-21 | A kind of processing method and equipment of rendering task |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810488258.9A CN108710543A (en) | 2018-05-21 | 2018-05-21 | A kind of processing method and equipment of rendering task |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108710543A true CN108710543A (en) | 2018-10-26 |
Family
ID=63869351
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810488258.9A Pending CN108710543A (en) | 2018-05-21 | 2018-05-21 | A kind of processing method and equipment of rendering task |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108710543A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111506407A (en) * | 2020-04-14 | 2020-08-07 | 中山大学 | Resource management and job scheduling method, system and medium combining Pull mode and Push mode |
| CN113157415A (en) * | 2021-04-21 | 2021-07-23 | 中国电影科学技术研究所 | Farm rendering method and device, electronic equipment and storage medium |
| WO2022222403A1 (en) * | 2021-04-21 | 2022-10-27 | 上海商汤科技开发有限公司 | Task distribution system, method, and apparatus, computer device, and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1268701A (en) * | 1999-03-29 | 2000-10-04 | 国际商业机器公司 | Method and device for co-operation agency system for target intensifying effect distribution arrangement |
| US6192388B1 (en) * | 1996-06-20 | 2001-02-20 | Avid Technology, Inc. | Detecting available computers to participate in computationally complex distributed processing problem |
| EP2463824A1 (en) * | 2010-12-13 | 2012-06-13 | Thomson Licensing | Method and input-output device for rendering at least one of audio, video and computer graphics content and servicing device for delivering at least one of pre-rendered audio, pre-rendered video and pre-rendered computer graphics content |
| CN103530083A (en) * | 2013-10-31 | 2014-01-22 | 广东粤铁瀚阳科技有限公司 | Cloud computing based super display platform for mass information |
| CN103765376A (en) * | 2011-06-16 | 2014-04-30 | 柯斯提克绘图公司 | Graphics processor with non-blocking concurrent architecture |
| CN204066193U (en) * | 2014-06-09 | 2014-12-31 | 中航远景（北京）科技股份有限公司 | A kind of virtual scene driving device |
| CN104767776A (en) * | 2014-01-07 | 2015-07-08 | 腾讯科技（深圳）有限公司 | Real-time network routing adjustment method, device and system |
| CN105282561A (en) * | 2015-10-21 | 2016-01-27 | 北京中科大洋科技发展股份有限公司 | Stereoscopic television signal editing system based on cluster rendering |
| US20160140689A1 (en) * | 2014-11-13 | 2016-05-19 | Nvidia Corporation | Supersampling for spatially distributed and disjoined large-scale data |
-
2018
- 2018-05-21 CN CN201810488258.9A patent/CN108710543A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6192388B1 (en) * | 1996-06-20 | 2001-02-20 | Avid Technology, Inc. | Detecting available computers to participate in computationally complex distributed processing problem |
| CN1268701A (en) * | 1999-03-29 | 2000-10-04 | 国际商业机器公司 | Method and device for co-operation agency system for target intensifying effect distribution arrangement |
| EP2463824A1 (en) * | 2010-12-13 | 2012-06-13 | Thomson Licensing | Method and input-output device for rendering at least one of audio, video and computer graphics content and servicing device for delivering at least one of pre-rendered audio, pre-rendered video and pre-rendered computer graphics content |
| CN103765376A (en) * | 2011-06-16 | 2014-04-30 | 柯斯提克绘图公司 | Graphics processor with non-blocking concurrent architecture |
| CN103530083A (en) * | 2013-10-31 | 2014-01-22 | 广东粤铁瀚阳科技有限公司 | Cloud computing based super display platform for mass information |
| CN104767776A (en) * | 2014-01-07 | 2015-07-08 | 腾讯科技（深圳）有限公司 | Real-time network routing adjustment method, device and system |
| CN204066193U (en) * | 2014-06-09 | 2014-12-31 | 中航远景（北京）科技股份有限公司 | A kind of virtual scene driving device |
| US20160140689A1 (en) * | 2014-11-13 | 2016-05-19 | Nvidia Corporation | Supersampling for spatially distributed and disjoined large-scale data |
| CN105282561A (en) * | 2015-10-21 | 2016-01-27 | 北京中科大洋科技发展股份有限公司 | Stereoscopic television signal editing system based on cluster rendering |
Non-Patent Citations (2)
| Title |
|---|
| XIN GAO ET AL.: "《Study on the hybrid software architecture of distributed parallel rendering system for virtual design》", 《2009 7TH IEEE INTERNATIONAL CONFERENCE ON INDUSTRIAL INFORMATICS》 * |
| 仝伯兵 等: "《集群渲染***构建及优化》", 《软件导刊》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111506407A (en) * | 2020-04-14 | 2020-08-07 | 中山大学 | Resource management and job scheduling method, system and medium combining Pull mode and Push mode |
| WO2021208240A1 (en) * | 2020-04-14 | 2021-10-21 | 中山大学 | Pull mode and push mode combined resource management and job scheduling method and system, and medium |
| CN113157415A (en) * | 2021-04-21 | 2021-07-23 | 中国电影科学技术研究所 | Farm rendering method and device, electronic equipment and storage medium |
| WO2022222403A1 (en) * | 2021-04-21 | 2022-10-27 | 上海商汤科技开发有限公司 | Task distribution system, method, and apparatus, computer device, and storage medium |
| CN113157415B (en) * | 2021-04-21 | 2023-10-13 | 中国电影科学技术研究所 | Farm rendering method, device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108710543A (en) | A kind of processing method and equipment of rendering task | |
| CN106209682A (en) | Business scheduling method, device and system | |
| CN102929834A (en) | Many-core processor and inter-core communication method thereof and main core and auxiliary core | |
| CN104281489B (en) | Multithreading requesting method and system under SOA framework | |
| CN103631873B (en) | A kind of data compression method and storage system | |
| CN112134964B (en) | Controller distribution method, computer equipment, storage medium and network service system | |
| CN106713375A (en) | Method and device for allocating cloud resources | |
| CN106959895A (en) | The resource regulating method and system of quick release thread | |
| CN108282526A (en) | Server dynamic allocation method and system between double clusters | |
| CN104866460B (en) | A kind of fault-tolerant adaptive reconfigurable System and method for based on SoC | |
| CN114968601A (en) | Scheduling method and scheduling system for AI training jobs with resources reserved according to proportion | |
| CN100464541C (en) | Method and system for realizing time-division multiplexing bearing resource share | |
| CN105045727B (en) | A kind of method and apparatus for accessing shared drive | |
| CN107396445A (en) | A kind of information reporting dispatching method, system, host device and slave devices | |
| CN104184685B (en) | Data center resource distribution method, apparatus and system | |
| CN109039694A (en) | A kind of the global network resource allocation methods and device of service-oriented | |
| CN113067833A (en) | Collaborative configuration service method and related components | |
| CN104253807B (en) | A kind of media resource control method and device | |
| CN113658351A (en) | Product production method and device, electronic equipment and storage medium | |
| CN103299298A (en) | Service processing method and system | |
| CN111209098A (en) | Intelligent rendering scheduling method, server, management node and storage medium | |
| CN104657204B (en) | short task processing method, device and operating system | |
| CN106447755A (en) | Animation rendering system | |
| CN107273196A (en) | Bioinformatics high-performance calculation job scheduling and system administration external member | |
| CN103677994B (en) | Distributed data processing system, device and method |
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: 20181026 |
|
| RJ01 | Rejection of invention patent application after publication |