Application allocation in datacenters

1. An allocation entity of a datacenter, said allocation entity comprising at least one processor and a memory coupled to the at least one processor executing program code in the memory to perform operations:

to receive an application allocation request for deploying an application or a virtual image on which the application is installed onto one of a plurality of processing units of the datacenter, the application allocation request comprising application network load information that describes the application's demands on communication traffic in the datacenter after the application or the virtual image has been deployed onto the one of the plurality of processing units;

to acquire path network load information that indicates a network load of network paths to at least a part of the plurality of processing units; and

to select a processing unit from the plurality of processing units for deploying the application or the virtual image based on the application network load information and the path network load information;

to select the processing unit by comparing the application network load information to path network load information of network paths to each one of the plurality of processing units; and

to select the processing unit based on a least network load of the network paths to each one of the plurality of processing units.

2. The allocation entity of claim 1 , wherein the network paths couple a gateway entity of the datacenter to the plurality of processing units over at least one switch level, and wherein the operations are further configured:

to acquire sub-path network load information indicating network load of network sub-paths between the gateway entity and the plurality of processing units; and

to select one of the processing units from the plurality of processing units for deploying the application based on the sub-path network load information of a sub-path coupled to the one processing unit.

3. The allocation entity of claim 2 , the operations further configured:

to determine whether a further sub-path coupled the one processing unit is available based on the application network load information; and

to select another processing unit from the plurality of processing units for deploying the application or the virtual image based on the further sub-path not being available.

4. The allocation entity of claim 1 , the operations further configured:

to receive path network reservation information; and

to select the processing unit from the plurality of processing units also based on the path network reservation information.

5. The allocation entity of claim 1 , the operations further configured to select a further processing unit from the plurality of processing units for deploying a further application or virtual image based on further application network load information and the path network load information.

6. The allocation entity of claim 1 , wherein the application network load information comprises an application network load profile.

7. The allocation entity of claim 1 , being implemented as an application running on or more processing units of the datacenter.

8. A method of operating a network entity of a datacenter, said method comprising:

receiving an application allocation request for deploying an application or a virtual image on which the application is installed onto one of a plurality of processing units of the datacenter, the application allocation request comprising application network load information that describes the application's demands on communication traffic in the datacenter after the application or the virtual image has been deployed onto the one of the plurality of processing units;

acquiring path network load information that indicates a network load of network paths to at least a part of the plurality of processing units; and

selecting a processing unit from the plurality of processing units for deploying the application or the virtual image based on the application network load information and the path network load information;

selecting the processing unit by comparing the application network load information to path network load information of network paths to each one of the plurality of processing units; and

selecting the processing unit based on a least network load of the network paths to each one of the plurality of processing units.

9. The method of claim 8 being adapted to an allocation entity comprising a processor and a memory coupled to the processor.

10. A computer program comprising a non-transitory medium having executable program code stored thereon, the executable program code loadable on at least one processor to cause the at least one processor to execute the method of claim 8 .

11. A method of operating a datacenter comprising a plurality of processing units, said method comprising:

receiving an application allocation request for deploying an application or a virtual image on which the application is installed onto one of the plurality of processing units, the application allocation request comprising application network load information that describes the application's demands on communication traffic in the datacenter after the application or the virtual image has been deployed onto the one of the plurality of processing units;

acquiring path network load information that indicates a network load of network paths to at least a part of the plurality of processing units; and

selecting a processing unit from the plurality of processing units for deploying the application or the virtual image based on the application network load information and the path network load information;

selecting the processing unit by comparing the application network load information to path network load information of network paths to each one of the plurality of processing units; and

selecting the processing unit based on a least network load of the network paths to each one of the plurality of processing units.

12. The method of operating a datacenter of claim 11 , wherein the datacenter comprises a gateway entity and at least one switch level, the plurality of processing units being coupled via network paths over the at least one switch level to said gateway entity, the method further comprising:

acquiring sub-path network load information indicating network load of network sub-paths between the gateway entity and the plurality of processing units; and

selecting one of the processing units from the plurality of processing units for deploying the application based on the sub-path network load information of a sub-path coupled to the one processing unit.

13. The method of claim 12 , further comprising:

determining whether a further sub-path coupled the one processing unit is available based on the application network load information; and

selecting another processing unit from the plurality of processing units for deploying the application based on the further sub-path not being available.

14. The method of claim 11 , further comprising:

receiving path network reservation information; and

selecting the processing unit from the plurality of processing units also based on the path network reservation information.

15. The method of claim 11 , further comprising selecting a further processing unit from the plurality of processing units for deploying a further application or virtual image based on further application network load information and the path network load information.

16. A datacenter comprising:

a plurality of processing units;

network paths to at least a part of the plurality of processing units; and

the allocation entity of claim 1 .

17. The method of claim 8 , wherein the network paths couple a gateway entity of the datacenter to the plurality of processing units over at least one switch level, the method further comprising acquiring sub-path network load information indicating network load of network sub-paths between the gateway entity and the plurality of processing units; and

selecting one of the processing units from the plurality of processing units for deploying the application based on the sub-path network load information of a sub-path coupled to the one processing unit.

18. The method of claim 17 , further comprising:

determining whether a further sub-path coupled the one processing unit is available based on the application network load information; and

selecting another processing unit from the plurality of processing units for deploying the application or the virtual image based on the further sub-path not being available.