CN104580368A - Virtual network embedded system and method for cloud rendering in optical data center network - Google Patents
Virtual network embedded system and method for cloud rendering in optical data center network Download PDFInfo
- Publication number
- CN104580368A CN104580368A CN201410748055.0A CN201410748055A CN104580368A CN 104580368 A CN104580368 A CN 104580368A CN 201410748055 A CN201410748055 A CN 201410748055A CN 104580368 A CN104580368 A CN 104580368A
- Authority
- CN
- China
- Prior art keywords
- data center
- virtual
- cloud
- network
- cloud rendering
- 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.)
- Granted
Links
- 238000009877 rendering Methods 0.000 title claims abstract description 174
- 230000003287 optical effect Effects 0.000 title claims abstract description 66
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000013507 mapping Methods 0.000 claims abstract description 16
- 230000003370 grooming effect Effects 0.000 claims abstract description 9
- 230000006835 compression Effects 0.000 claims description 27
- 238000007906 compression Methods 0.000 claims description 27
- 239000013307 optical fiber Substances 0.000 claims description 17
- 230000037361 pathway Effects 0.000 claims description 8
- 238000007596 consolidation process Methods 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 3
- 230000010354 integration Effects 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- FFBHFFJDDLITSX-UHFFFAOYSA-N benzyl N-[2-hydroxy-4-(3-oxomorpholin-4-yl)phenyl]carbamate Chemical compound OC1=C(NC(=O)OCC2=CC=CC=C2)C=CC(=C1)N1CCOCC1=O FFBHFFJDDLITSX-UHFFFAOYSA-N 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The invention provides a virtual network embedded system and method for cloud rendering in an optical data center network. The virtual network embedded system for cloud rendering in the optical data center network includes an optical backbone network, a rendering data center, an image storage data center and an image packed data center. The virtual network embedded method for cloud rendering in the optical data center network comprises the following steps: sending a virtual network embedded request to the optical data center network by a cloud rendering user; describing the virtual network embedded request in a modeling manner; sorting all cloud rendering tasks; performing virtual network embedding to current clouding rendering tasks; updating bottom layer optical data center network resource; counting consumed wave length amount and server amount; obtaining lower bound value ranges of consumed wave length amount and server amount of all cloud rendering tasks and verifying whether a current virtual network embedded operation is the optimal or not. Methods of server integration and traffic grooming are combined for node mapping and link mapping of the virtual network embedded request for real cloud rendering tasks and unreal cloud rendering tasks and the purpose of reducing rendering cost effectively is achieved.
Description
Technical field
The invention belongs to technical field of communication network, be specifically related to virtual network embedded system that in a kind of smooth data center network, facing cloud is played up and method.
Background technology
Animation rendering is the process a certain Digital Design being converted to lively animation.This technology relates to the models such as a large amount of geometry, light and material, often needs to operate on powerful calculation procedure.Along with the fast development of Accurate Model, a frame animation comprises thousands of geometrical models usually, and each geometrical model is all containing complicated optimum configurations.Therefore, animation rendering is very consuming time, and unit render time is about every frame two hours.For the animation that a duration is 160 minutes, its render time will up to 2.88 × 10
6hour.Obviously, the core objective of animation rendering shortens render time exactly.
Adopt Cluster Rendering technology, large batch of server carries out parallel computation process to same file of playing up, and shortens render time.But this parallel work-flow adds O&M cost, and the idle of a large amount of computational resource may be caused.Therefore, cloud Rendering arises at the historic moment.By cluster virtualization, the server resource of isomery is integrated into independently cloud and plays up in farm.Based on cloud computing theory, a manager plays up farm at the cloud of far-end and safeguards powerful calculation procedure, and is responsible for receiving the rendering task from user.Cloud is played up after farm performs corresponding rendering task, and rendered picture is fed back to user terminal.Without the need to buying rendering apparatus expensive limited in one's ability, user only need play up farm management person to cloud and pay short-term lease expense, just can obtain desirable rendering effect by computational resource powerful in cloud.
But single cluster virtualization also cannot meet the actual demand that complete cloud plays up process.A complete cloud plays up that process needs deployment to play up, the data center such as image compression and storage.Due to the restriction of geographical space, especially for trans-regional rendering task cooperation, above-mentioned different types of data center is often apart from each other.If a certain user thinks that quick obtaining is played up the rendered picture fed back on farm by cloud, then need high-speed light badly interconnected to support that the information between user and specific data center forwards.Therefore, complete cloud is played up the seamless cloud that process need realize light and data center and is merged, thus forms the light data center network supporting that cloud is played up.Be embodied in, except cluster virtualization (relating to virtual machine), also will realize optical-fiber network virtual (relating to virtual optical path).For this reason, each cloud rendering task is one in essence and plays up the virtual network embedding request of farm management person's initiation to cloud, and each virtual network is made up of virtual optical path and virtual machine.This request is manager by the condition successfully accepted can be the virtual network successful execution node of this request and link maps.Wherein, node mapping refers to the mapping of intracardiac a certain server in virtual machine to specific data, and link maps refers to that virtual optical path arrives the mapping of specifying optical fiber link shared by physical pathway.Existing virtual network embedding grammar is not all suitable for the exclusive demand and feature that cloud in light data center network plays up.
Summary of the invention
For the deficiency that prior art exists, the invention provides virtual network embedded system and method that in a kind of smooth data center network, facing cloud is played up.
Technical scheme of the present invention is:
The virtual network embedded system that in a kind of smooth data center network, facing cloud is played up, virtual network for carrying out true cloud rendering task and non-genuine cloud rendering task embeds, and this system comprises: light backbone network, rendering data center, image store data center and image compression data center;
Each node deployment optical cross-connect in light backbone network, the link between each optical cross-connect is connected by wavelength division multiplexing optical fiber;
Using the node of three in light backbone network as cloud rendering data node, dispose rendering data center respectively, image stores data center and image compression data center, light backbone network, rendering data center, image store in data center and image compression data and are formed centrally light data center network;
Rendering data center, image store intracardiacly in data center and image compression data all disposes cluster server, arranges some virtual machines, form virtual optical path between each virtual machine in cluster server.
The virtual network embedding grammar of the virtual network embedded system that facing cloud is played up in light data center network described in employing, comprises the following steps:
Step 1: cloud is played up user and sent virtual network embedding request to light data center network;
Step 2: carry out modelling description to the virtual network embedding request of true cloud rendering task and non-genuine cloud rendering task respectively, virtual network comprises virtual machine and virtual optical path;
Step 2-1: the feature that need have been walked abreast by rendering data center, image storage data center and image compression data center for true cloud rendering task, the virtual network that true cloud rendering task request embeds is characterized by with Nodes optical cross-connect a certain in light data center network as root node, storing virtual machine corresponding to virtual machine corresponding to data center and image compression data center with virtual machine corresponding to rendering data center, image is three leaf nodes, the tree being branch with the virtual optical path formed between three leaf nodes;
Step 2-2: for the feature that non-genuine cloud rendering task only needs rendering data center, image stores any class data center complete independently in the heart in data center and image compression data, the virtual network that the request of non-genuine rendering task embeds is characterized by with Nodes optical cross-connect a certain in light data center network be an end node, the corresponding virtual machine of a certain class data center virtual optical path that is other end node;
Step 3: all cloud rendering tasks are sorted;
Step 4: the virtual network of carrying out current cloud rendering task embeds, and namely carries out node mapping and link maps to the virtual network that current cloud rendering task request embeds;
Step 4-1: adopt Server Consolidation method, first of to be mapped in specific data in the heart can hold in the server of this virtual machine for the virtual machine of the virtual network current cloud rendering task request embedded, and realizes node mapping;
Step 4-2: adopt traffic grooming method, is mapped to virtual optical path on the optical fiber link of appointment shared by physical pathway, thus realizes link maps;
Step 5: if step 4-1 and step 4-2 all runs succeeded, bottom light data center network resource is upgraded according to mapping result, and turn to step 4 to continue the virtual network embedding operation of other cloud rendering tasks, until the virtual network of all cloud rendering tasks has embedded; Otherwise the virtual network refusing current cloud rendering task embeds request, and turns to step 4 to continue the virtual network embedding operation of other cloud rendering tasks, until the virtual network of all cloud rendering tasks has embedded;
Step 6: after completing the virtual network embedding operation of all cloud rendering tasks, add up wavelength sum that all cloud rendering tasks consume and server total;
Step 7: to minimize wavelength sum that all cloud rendering tasks consume and server adds up to target, mathematical description is carried out to the virtual network imbedding problem that facing cloud in light data center network is played up, obtains that wavelength sum floor value that all cloud rendering tasks consume is interval and the total floor value of server is interval;
Step 8: the wavelength consumed by all cloud rendering tasks of statistics is total and the total wavelength sum floor value consumed with all cloud rendering tasks respectively of server is interval and the total floor value interval of server compares analysis respectively, thus whether checking current virtual internet startup disk operates optimum: if the wavelength sum that all cloud rendering tasks of statistics consume is greater than the wavelength sum floor value interval that all cloud rendering tasks consume, or the server sum that all cloud rendering tasks consume is greater than the server sum floor value interval that all cloud rendering tasks consume, then the operation of current virtual internet startup disk is not optimum, return step 3, otherwise current virtual internet startup disk is operating as optimum.
Described preferential or non-genuine cloud rendering task is preferential with true cloud rendering task when all cloud rendering tasks are sorted.
In the traffic grooming method of described step 4-2, adopt Shortest path routing to calculate virtual optical path physical pathway to be mapped and the optical fiber link shared by this path, adopt initial hit method to be that virtual optical path distributes wavelength.
Beneficial effect:
Node mapping and link maps are carried out in the virtual network embedding request being true cloud rendering task and non-genuine cloud rendering task in conjunction with employing Server Consolidation and traffic grooming method, play up mode by mixed cloud and shorten render time simultaneously, reach the object effectively reducing and play up expense (wavelength that all rendering tasks consume is total and server is total).The present invention can build except the true rendering intent of the animation effect of similar actual scene except adopting, by introducing the non-genuine rendering intent lower to resource requirement such as ink and wash, paper-cut, reduce render time and expense, can truly play up the unique visual effects that cannot show merely for spectators present simultaneously.
Accompanying drawing explanation
Fig. 1 is the virtual network embedded system topological diagram that in the light data center network of the specific embodiment of the invention, facing cloud is played up, wherein, 1 ~ 15 is optical cross-connect, and 16 is rendering data center, 17 is image storage data center, and 18 is image compression data center;
Fig. 2 is the virtual network embedding grammar flow chart that in the light data center network of the specific embodiment of the invention, facing cloud is played up;
Fig. 3 is the cloud rendering server expense comparison diagram of the virtual network embedding grammar that in the light data center network of the specific embodiment of the invention, facing cloud is played up;
Fig. 4 is that the cloud of the virtual network embedding grammar that in the light data center network of the specific embodiment of the invention, facing cloud is played up plays up wavelength expense comparison diagram.
Embodiment
Below in conjunction with accompanying drawing, the specific embodiment of the present invention is elaborated.
The virtual network embedded system that in a kind of smooth data center network, facing cloud is played up, virtual network for carrying out true cloud rendering task and non-genuine cloud rendering task embeds, as shown in Figure 1, this system comprises: light backbone network, rendering data center, image store data center and image compression data center;
Each node deployment optical cross-connect in light backbone network, the link between each optical cross-connect is connected by wavelength division multiplexing optical fiber; Using the node of three in light backbone network as cloud rendering data node, dispose rendering data center respectively, image stores data center and image compression data center, light backbone network, rendering data center, image store in data center and image compression data and are formed centrally light data center network; Rendering data center, image store intracardiacly in data center and image compression data all disposes cluster server, arranges some virtual machines, form virtual optical path between each virtual machine in cluster server.
Light data center network comprises optical cross-connect set N, optical fiber link set E and data center set D.
15 nodes are had in the light backbone network of present embodiment, | N|=15, | E|=22, | D|=3.The initial capacity of each server is equal, unit service device capacity SC=16, and the initial capacity of each wavelength is also equal, per wavelength capacity WC=16.
The virtual network embedding grammar of the virtual network embedded system that facing cloud is played up in light data center network described in employing, as shown in Figure 2, comprises the following steps:
Step 1: cloud is played up user and sent virtual network embedding request to light data center network;
Step 2: carry out modelling description to the virtual network embedding request of true cloud rendering task and non-genuine cloud rendering task respectively, virtual network comprises virtual machine and virtual optical path;
Step 2-1: the feature that need have been walked abreast by rendering data center, image storage data center and image compression data center for true cloud rendering task, the virtual network that true cloud rendering task request embeds is characterized by with Nodes optical cross-connect a certain in light data center network as root node, storing virtual machine corresponding to virtual machine corresponding to data center and image compression data center with virtual machine corresponding to rendering data center, image is three leaf nodes, the tree being branch with the virtual optical path formed between three leaf nodes;
In the present embodiment, above-mentioned virtual network can be characterized by the tree < s of four-tuple, D, t
2, r
2>;
S locates optical cross-connect node for user, and s ∈ N;
T
2for the multidimensional virtual optical path demand of the virtual network that true rendering task request embeds, and
for from optical cross-connect node s to the virtual optical path weight at rendering data center, and
for from user's optical cross-connect node s to the virtual optical path weight at image compression data center, and
for storing the virtual optical path weight of data center from user's optical cross-connect node s to image, and
R
2for the multidimensional virtual machine demand of the virtual network that true cloud rendering task request embeds, and
R
2=t
2=4,4,4},
for the corresponding virtual machine weight in rendering data center, and
for the corresponding virtual machine weight in image compression data center, and
for image stores the corresponding virtual machine weight of data center, and
Step 2-2: for the feature that non-genuine cloud rendering task only needs rendering data center, image stores any class data center complete independently in the heart in data center and image compression data, the virtual network that the request of non-genuine cloud rendering task embeds is characterized by with Nodes optical cross-connect a certain in light data center network be an end node, the corresponding virtual machine of a certain class data center virtual optical path that is other end node;
In the present embodiment, above-mentioned virtual network can be characterized by the virtual optical path < s of tlv triple, t
1, r
1>, wherein, t
1for the virtual optical path demand of the virtual network that the request of non-genuine rendering task embeds, r
1for the virtual machine demand of the virtual network that the request of non-genuine rendering task embeds, r
1=t
1={ 2,2,2}.
Step 3: preferential or non-genuine cloud rendering task preferential with true cloud rendering task, sorts to all cloud rendering tasks;
In the present embodiment, adopt two kinds of sortords, one truly plays up preferential, then corresponding method can called after MRF; Another kind is that non-genuine is played up preferentially, then corresponding method can called after SVRF.
Step 4: the virtual network of carrying out current cloud rendering task embeds, namely carries out node mapping and link maps to the virtual network (tree or light path) that current cloud rendering task request embeds;
Step 4-1: adopt Server Consolidation method, first of to be mapped in specific data in the heart can hold in the server of this virtual machine for the virtual machine of the virtual network current cloud rendering task request embedded, and realizes node mapping;
First that to be mapped in specific data in the heart by virtual machine in tree or light path can be held in the server of this virtual machine, thus realize node mapping.In the present embodiment, the corresponding virtual machine in rendering data center can only be mapped to rendering data central interior, the corresponding virtual machine in image compression data center can only be mapped to image compression data central interior, and the corresponding virtual machine of image storage data center can only be mapped to image storage data center inside.Adopt the reason of Server Consolidation method to be just that it can make abundant virtual machine be mapped in same server, thus reduce server usage quantity, reduce and play up expense.
Step 4-2: adopt traffic grooming method, is mapped to virtual optical path on the optical fiber link of appointment shared by physical pathway, thus realizes link maps;
Adopt traffic grooming method, virtual optical path is mapped on the optical fiber link of appointment shared by physical pathway, thus realizes link maps.In the present embodiment, adopt Shortest path routing to calculate virtual optical path physical pathway to be mapped and the optical fiber link shared by this path, adopt initial hit method to be that virtual optical path distributes wavelength.Adopt the reason of traffic grooming method to be just that it can make abundant virtual optical path be mapped on same wavelength, thus reduce wavelength usage quantity, reduce and play up expense.
Step 5: if step 4-1 and step 4-2 all runs succeeded, bottom light data center network resource is upgraded according to mapping result, and turn to step 4 to continue the virtual network embedding operation of other cloud rendering tasks, until the virtual network of all cloud rendering tasks has embedded; Otherwise the virtual network refusing current cloud rendering task embeds request, and turns to step 4 to continue the virtual network embedding operation of other cloud rendering tasks, until the virtual network of all cloud rendering tasks has embedded;
Step 6: after completing the virtual network embedding operation of all cloud rendering tasks, add up wavelength sum that all cloud rendering tasks consume and server total;
Step 7: to minimize wavelength sum that all cloud rendering tasks consume and server adds up to target, mathematical description is carried out to the virtual network imbedding problem that facing cloud in light data center network is played up, obtains that wavelength sum floor value that all cloud rendering tasks consume is interval and the total floor value of server is interval;
Step 7-1: in light data center network the virtual network imbedding problem that facing cloud is played up target for the wavelength sum that minimizes all cloud rendering tasks and consume and server total, mathematical description is as follows:
Minimize MC (1)
Wherein,
MC≥MC
1·MC
2(2)
Adopt maximum wavelength call number MC shared in formula (3) statistics cloud render process
1,
represent once a certain wavelength is taken by virtual optical path, then MC
1(initial value is 0) adds 1 certainly; Similarly, maximum service device call number MC shared in formula (4) statistics cloud render process is adopted
2; Obviously, optimization aim is the wavelength sum that minimizes all rendering tasks and consume and server sum M C;
W: wavelength call number;
boolean variable, if P
1bar non-genuine rendering task, virtual optical path from s to j occupies the wavelength w of optical fiber link (m, n), then this Boolean variable is 1, otherwise is 0;
boolean variable, if P
2the true rendering task of bar, virtual optical path from s to j occupies the wavelength w of optical fiber link (m, n), then this Boolean variable is 1, otherwise is 0;
boolean variable, if by user's optical cross-connect node s initiate, V
1the virtual machine of individual non-genuine rendering task is mapped to the server f in data center j
jin, then this Boolean variable is 1, otherwise is 0;
boolean variable, if by user's optical cross-connect node s initiate, V
2the virtual machine of individual true rendering task is mapped to the server f in data center j
jin, then this Boolean variable is 1, otherwise is 0;
F
j: the server call number in data center j;
J: for completing the data center of rendering task, and j ∈ D;
The initial sum of mn: certain optical fiber link stops end points, and mn ∈ E;
P
1: the virtual optical path call number in the virtual network that the request of non-genuine rendering task embeds;
P
2: the virtual optical path call number in the virtual network that true rendering task request embeds;
V
1: the virtual machine call number in the virtual network that the request of non-genuine rendering task embeds;
V
2: the virtual machine call number in the virtual network that true rendering task request embeds.
Step 7-2: the wavelength sum floor value that theory deduction all clouds rendering task consumes is interval and the total floor value of server is interval;
For the data center of certain type, the total number interval of occupied minimum service device is,
Wherein,
θ with
be the positive integer being greater than 1.
SR
1: the set of non-genuine rendering task, and
SR
2: true rendering task set, and
For every bar optical fiber link, the total number interval of occupied minimum wavelength is,
Wherein,
α and β is the positive integer being greater than 1; n
s,jthe total number of paths of link disjoint between recording light cross-connector node s and data center j, such as, n
s, RDCthe total number of paths of link disjoint between recording light cross-connector node s and rendering data center RDC.
the non-genuine rendering task set initiated by user's optical cross-connect node s;
the true rendering task set initiated by user's optical cross-connect node s;
Step 8: the wavelength consumed by all cloud rendering tasks of statistics is total and the total wavelength sum floor value consumed with all cloud rendering tasks respectively of server is interval and the total floor value interval of server compares analysis respectively, thus whether checking current virtual internet startup disk operates optimum: if the wavelength sum that all cloud rendering tasks of statistics consume is greater than the wavelength sum floor value interval that all cloud rendering tasks consume, or the server sum that all cloud rendering tasks consume is greater than the server sum floor value interval that all cloud rendering tasks consume, then the operation of current virtual internet startup disk is not optimum, return step 3, otherwise current virtual internet startup disk is operating as optimum.
The theoretical dividing value interval of the present invention and derivation is compared: as shown in Figure 3, when non-genuine cloud rendering task and true cloud rendering task number all rise to 800 situations from 100, the MRF method no matter employing proposes or SVRF method, the server sum of all cloud rendering tasks consumption all drops in the theory interval of corresponding floor value, thus demonstrates the validity of present system and method.
As shown in Figure 4, when non-genuine cloud rendering task and true cloud rendering task number all rise to 800 from 100, the MRF method no matter employing proposes or SVRF method, the wavelength sum of all cloud rendering tasks consumption all drops in the theory interval of corresponding floor value, thus demonstrates the validity of present system and method.
Claims (4)
1. the virtual network embedded system that in a light data center network, facing cloud is played up, virtual network for carrying out true cloud rendering task and non-genuine cloud rendering task embeds, it is characterized in that, comprising: light backbone network, rendering data center, image store data center and image compression data center;
Each node deployment optical cross-connect in light backbone network, the link between each optical cross-connect is connected by wavelength division multiplexing optical fiber;
Using the node of three in light backbone network as cloud rendering data node, dispose rendering data center respectively, image stores data center and image compression data center, light backbone network, rendering data center, image store in data center and image compression data and are formed centrally light data center network;
Rendering data center, image store intracardiacly in data center and image compression data all disposes cluster server, arranges some virtual machines, form virtual optical path between each virtual machine in cluster server.
2. adopt the virtual network embedding grammar of the virtual network embedded system that facing cloud is played up in smooth data center network according to claim 1, it is characterized in that, comprise the following steps:
Step 1: cloud is played up user and sent virtual network embedding request to light data center network;
Step 2: carry out modelling description to the virtual network embedding request of true cloud rendering task and non-genuine cloud rendering task respectively, virtual network comprises virtual machine and virtual optical path;
Step 2-1: the feature that need have been walked abreast by rendering data center, image storage data center and image compression data center for true cloud rendering task, the virtual network that true cloud rendering task request embeds is characterized by with Nodes optical cross-connect a certain in light data center network as root node, storing virtual machine corresponding to virtual machine corresponding to data center and image compression data center with virtual machine corresponding to rendering data center, image is three leaf nodes, the tree being branch with the virtual optical path formed between three leaf nodes;
Step 2-2: for the feature that non-genuine cloud rendering task only needs rendering data center, image stores any class data center complete independently in the heart in data center and image compression data, the virtual network that the request of non-genuine cloud rendering task embeds is characterized by with Nodes optical cross-connect a certain in light data center network be an end node, the corresponding virtual machine of a certain class data center virtual optical path that is other end node;
Step 3: all cloud rendering tasks are sorted;
Step 4: the virtual network of carrying out current cloud rendering task embeds, and namely carries out node mapping and link maps to the virtual network that current cloud rendering task request embeds;
Step 4-1: adopt Server Consolidation method, first of to be mapped in specific data in the heart can hold in the server of this virtual machine for the virtual machine of the virtual network current cloud rendering task request embedded, and realizes node mapping;
Step 4-2: adopt traffic grooming method, is mapped to virtual optical path on the optical fiber link of appointment shared by physical pathway, thus realizes link maps;
Step 5: if step 4-1 and step 4-2 all runs succeeded, bottom light data center network resource is upgraded according to mapping result, and turn to step 4 to continue the virtual network embedding operation of other cloud rendering tasks, until the virtual network of all cloud rendering tasks has embedded; Otherwise the virtual network refusing current cloud rendering task embeds request, and turns to step 4 to continue the virtual network embedding operation of other cloud rendering tasks, until the virtual network of all cloud rendering tasks has embedded;
Step 6: after completing the virtual network embedding operation of all cloud rendering tasks, add up wavelength sum that all cloud rendering tasks consume and server total;
Step 7: to minimize wavelength sum that all cloud rendering tasks consume and server adds up to target, mathematical description is carried out to the virtual network imbedding problem that facing cloud in light data center network is played up, obtains that wavelength sum floor value that all cloud rendering tasks consume is interval and the total floor value of server is interval;
Step 8: the wavelength consumed by all cloud rendering tasks of statistics is total and the total wavelength sum floor value consumed with all cloud rendering tasks respectively of server is interval and the total floor value interval of server compares analysis respectively, thus whether checking current virtual internet startup disk operates optimum: if the wavelength sum that all cloud rendering tasks of statistics consume is greater than the wavelength sum floor value interval that all cloud rendering tasks consume, or the server sum that all cloud rendering tasks consume is greater than the server sum floor value interval that all cloud rendering tasks consume, then the operation of current virtual internet startup disk is not optimum, return step 3, otherwise current virtual internet startup disk is operating as optimum.
3. the virtual network embedding grammar that in smooth data center network according to claim 2, facing cloud is played up, is characterized in that, described preferential or non-genuine cloud rendering task is preferential with true cloud rendering task when sorting to all cloud rendering tasks.
4. the virtual network embedding grammar that in smooth data center network according to claim 2, facing cloud is played up, it is characterized in that, in the traffic grooming method of described step 4-2, adopt Shortest path routing to calculate virtual optical path physical pathway to be mapped and the optical fiber link shared by this path, adopt initial hit method to be that virtual optical path distributes wavelength.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410748055.0A CN104580368B (en) | 2014-12-09 | 2014-12-09 | Facing cloud is rendered in light data central site network virtual network embedded system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410748055.0A CN104580368B (en) | 2014-12-09 | 2014-12-09 | Facing cloud is rendered in light data central site network virtual network embedded system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104580368A true CN104580368A (en) | 2015-04-29 |
| CN104580368B CN104580368B (en) | 2017-10-17 |
Family
ID=53095526
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410748055.0A Expired - Fee Related CN104580368B (en) | 2014-12-09 | 2014-12-09 | Facing cloud is rendered in light data central site network virtual network embedded system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104580368B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107479964A (en) * | 2016-06-08 | 2017-12-15 | 成都赫尔墨斯科技股份有限公司 | A kind of cloud rendering system |
| CN111901635A (en) * | 2020-06-17 | 2020-11-06 | 北京视博云信息技术有限公司 | Video processing method, device, storage medium and equipment |
| CN112132945A (en) * | 2020-11-24 | 2020-12-25 | 江苏普旭软件信息技术有限公司 | High-definition cloud rendering method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102904794A (en) * | 2012-09-27 | 2013-01-30 | 北京邮电大学 | Method and device for mapping virtual network |
| CN103457752A (en) * | 2012-05-30 | 2013-12-18 | 中国科学院声学研究所 | Virtual network mapping method |
| CN106038400A (en) * | 2016-06-28 | 2016-10-26 | 龙梅奎 | Hair-care shampoo |
-
2014
- 2014-12-09 CN CN201410748055.0A patent/CN104580368B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103457752A (en) * | 2012-05-30 | 2013-12-18 | 中国科学院声学研究所 | Virtual network mapping method |
| CN102904794A (en) * | 2012-09-27 | 2013-01-30 | 北京邮电大学 | Method and device for mapping virtual network |
| CN106038400A (en) * | 2016-06-28 | 2016-10-26 | 龙梅奎 | Hair-care shampoo |
Non-Patent Citations (3)
| Title |
|---|
| ANNA TZANAKAKI: ""Energy Efficiency in integrated IT and Optical"", 《COMPUTER COMMUNICATION WORKSHOPS》 * |
| GEORGAKILAS K N: ""Converged Optical Network and Data Center Virtual Infrastructure Planning"", 《JOURNAL OF OPTICAL COMMUNICATIONS AND NETWORKING》 * |
| SHUPING PENG: ""An Impairment-aware Virtual Optical Network Composition Mechanism for Future Internet"", 《OPTICS EXPRESS》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107479964A (en) * | 2016-06-08 | 2017-12-15 | 成都赫尔墨斯科技股份有限公司 | A kind of cloud rendering system |
| CN111901635A (en) * | 2020-06-17 | 2020-11-06 | 北京视博云信息技术有限公司 | Video processing method, device, storage medium and equipment |
| CN112132945A (en) * | 2020-11-24 | 2020-12-25 | 江苏普旭软件信息技术有限公司 | High-definition cloud rendering method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104580368B (en) | 2017-10-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10841241B2 (en) | Intelligent placement within a data center | |
| US11706645B2 (en) | Systems and methods for communications node upgrade and selection | |
| US10057179B2 (en) | Policy based scaling of network resources | |
| CN106385329B (en) | Processing method, device and the equipment of resource pool | |
| CN103561055B (en) | Web application automatic elastic extended method under conversation-based cloud computing environment | |
| CN106453457A (en) | Multi-priority service instance distribution in cloud computing platform | |
| CN104052803A (en) | Decentralized distributed rendering method and system | |
| CN112015536A (en) | Kubernetes cluster container group scheduling method, device and medium | |
| EP3268905A1 (en) | Hybrid cloud management | |
| CN107133083A (en) | A kind of virtual data center resource provider method based on virtualization technology | |
| CN110213345A (en) | A kind of AI algorithm management system | |
| CN106815254A (en) | A kind of data processing method and device | |
| CN104580368A (en) | Virtual network embedded system and method for cloud rendering in optical data center network | |
| CN106250202B (en) | The device and method of virtual machine is opened in a kind of cloud platform automation | |
| CN106909462A (en) | A kind of cloud resource regulating method and device | |
| CN105260232A (en) | Method for realizing optimal allocation of internal resources of cloud computing virtual machine and system | |
| CN104572298B (en) | The resource regulating method and device of video cloud platform | |
| CN108833592A (en) | Cloud host schedules device optimization method, device, equipment and storage medium | |
| CN109960579A (en) | A kind of method and device of adjustment business container | |
| CN105045906B (en) | The predictor method and device of impression information clicking rate | |
| CN109714652A (en) | Date storage method, system, storage medium and the equipment of ODN management system | |
| FR3093842A1 (en) | Method and system for optimizing clusters of virtual machines in a cloud computing platform | |
| Ke et al. | DisaggRec: Architecting Disaggregated Systems for Large-Scale Personalized Recommendation | |
| CN108076029A (en) | The building method and device of Workforce Management system | |
| CN104579896B (en) | A kind of partitioning method and device of virtual network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20171017 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |