Abstract: An allocation entity of a datacenter is provided, wherein said allocation entity being configured: to receive an application allocation request for allocating an application to one of a plurality of processing units of the datacenter, the application allocation request comprising application network load information; to acquire path network load information indicating 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 allocating the application based on the application network load information and the path network load information.

Abstract: A method and system of providing trustworthiness of communication among a plurality of communication nodes is described. This comprises arranging each of said communication nodes to perform a trustworthiness judging operation on received data elements for judging a received packet to be trustworthy or not, grouping said plurality of communication nodes into a plurality of distinguishable clusters, each cluster comprising at least two of said communication nodes, implementing in each respective cluster an intro-cluster trust mechanism such that trustworthiness of data elements sent by any member node of said respective cluster is judgable within said respective cluster, arranging said clusters such that each of said clusters comprises one or more multi-cluster-member nodes that belong to at least two different of said clusters, and muting inter-cluster traffic through said multi-cluster-member nodes.

Abstract: A multicast cloud controller (“MCC”) in a cloud system implements a process to manage multicast traffic in a cloud network. The MCC is coupled to at least one virtualized server for hosting one or more virtual machines (“VM”), wherein the virtualized server comprises at least one virtual switch (“VS”) that supports multiprotocol label switching (MPLS) and the virtual switch is coupled to a top of rack switch (“TORS”) that supports MPLS. MPLS is utilized to support multicast data traffic in the cloud system such that the system and method reduces state and is scalable.

Abstract: A multicast cloud controller (“MCC”) in a cloud system implements a process to manage multicast traffic in a cloud network. The MCC is coupled to at least one virtualized server for hosting one or more virtual machines (“VM”), wherein the virtualized server comprises at least one virtual switch (“VS”) that supports multiprotocol label switching (MPLS) and the virtual switch is coupled to a top of rack switch (“TORS”) that supports MPLS. MPLS is utilized to support multicast data traffic in the cloud system such that the system and method reduces state and is scalable.

Abstract: A multicast cloud controller (“MCC”) in a cloud system implements a process to manage multicast traffic in a cloud network. The MCC is coupled to at least one virtualized server for hosting one or more virtual machines (“VM”), wherein the virtualized server comprises at least one virtual switch (“VS”) that supports multiprotocol label switching (MPLS) and the virtual switch is coupled to a top of rack switch (“TORS”) that supports MPLS. MPLS is utilized to support multicast data traffic in the cloud system such that the system and method reduces state and is scalable.

Abstract: A multicast cloud controller (“MCC”) in a cloud system implements a process to manage multicast traffic in a cloud network. The MCC is coupled to at least one virtualized server for hosting one or more virtual machines (“VM”), wherein the virtualized server comprises at least one virtual switch (“VS”) that supports multiprotocol label switching (MPLS) and the virtual switch is coupled to a top of rack switch (“TORS”) that supports MPLS. MPLS is utilized to support multicast data traffic in the cloud system such that the system and method reduces state and is scalable.

Abstract: An allocation entity of a datacenter is provided, wherein said allocation entity being configured: to receive an application allocation request for allocating an application to one of a plurality of processing units of the datacenter, the application allocation request comprising application network load information; to acquire path network load information indicating 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 allocating the application based on the application network load information and the path network load information.

Abstract: Embodiments of the invention include a method performed by a cloud network manager flow entries in a cloud network. The CNM is coupled to virtualized servers for hosting virtual machines (“VM”) that each comprise a virtual switch coupled to a top of rack switch (“TORS”). The CNM receives notification messages that indicate virtual machines have been scheduled for activation on virtualized servers. The CNM determines a VM media access control (“MAC”) address associated with the VM and a virtual switch MAC address associated with that virtualized server's virtual switch. The CNM records an association between the VM MAC address and the virtual switch MAC address. The CNM further determines a label that associates the TORS with the virtual switch. The CNM sends a modifies flow entries in the virtual switch to indicate data packets matching the MPLS label and the VM MAC address should be forwarded to the VM.

Abstract: The invention relates to a method and a termination node for bundling multiple messages into a packet having a predetermined maximum size. The total size of messages received from an upper layer is checked if it is large enough to fill the maximum-sized packet. If not, the receipt of further messages is waited for a bundling delay time and if the bundling delay time has expired the messages received during said delay time are bundled into the packet. In order to ensure primarily a maximal bundling efficiency and secondarily a lowest possible delay time for the received messages before they are bundled the bundling delay time is calculated in response to an average arrival rate and an average size of the previously received messages such that the bundling delay time corresponds to an average time to fill the packet of the predetermined maximum size.

Abstract: Embodiments of the invention include a method performed by a cloud network manager flow entries in a cloud network. The CNM is coupled to virtualized servers for hosting virtual machines (“VM”) that each comprise a virtual switch coupled to a top of rack switch (“TORS”). The CNM receives notification messages that indicate virtual machines have been scheduled for activation on virtualized servers. The CNM determines a VM media access control (“MAC”) address associated with the VM and a virtual switch MAC address associated with that virtualized server's virtual switch. The CNM records an association between the VM MAC address and the virtual switch MAC address. The CNM further determines a label that associates the TORS with the virtual switch. The CNM sends a modifies flow entries in the virtual switch to indicate data packets matching the MPLS label and the VM MAC address should be forwarded to the VM.

Abstract: The invention relates to a method and a termination node for bundling multiple messages into a packet having a predetermined maximum size. The total size of messages received from an upper layer is checked if it is large enough to fill the maximum-sized packet. If not, the receipt of further messages is waited for a bundling delay time and if the bundling delay time has expired the messages received during said delay time are bundled into the packet. In order to ensure primarily a maximal bundling efficiency and secondarily a lowest possible delay time for the received messages before they are bundled the bundling delay time is calculated in response to an average arrival rate and an average size of the previously received messages such that the bundling delay time corresponds to an average time to fill the packet of the predetermined maximum size.

Abstract: A system and method for transporting data units for a communication network. In a communication between two peers a proxy and a network entity for controlling a partial communication are provided. One or both of the proxy and the network entity is arranged to modify one or more of the data units, in order to provide communication control related information that is different from communication control related information exchanged between the two peers The respective other or both of the proxy and the network entity is arranged to perform data unit transmission control based on the different communication control related information.

Abstract: A system operator may connect to an infrastructure resource and request a slice of spectrum resources with desired properties and a desired capacity. A slice manager of the infrastructure resource checks whether the request can be fulfilled from the available free capacity and optionally considers additional policies. If the outcome is positive, the slice manager offers a slice with a specific capacity, which may or may not be identical to the system operator' s requested capacity. The offer may contain other information such as cost and billing details. In case the offer is acceptable to the system operator, he sends an accept message to the slice, manager. Optionally, there may be additional negotiation steps.

Abstract: A method and system of providing trustworthiness of communication among a plurality of communication nodes is described. This comprises arranging each of said communication nodes to perform a trustworthiness judging operation on received data elements for judging a received packet to be trustworthy or not, grouping said plurality of communication nodes into a plurality of distinguishable clusters, each cluster comprising at least two of said communication nodes, implementing in each respective cluster an intro-cluster trust mechanism such that trustworthiness of data elements sent by any member node of said respective cluster is judgable within said respective cluster, arranging said clusters such that each of said clusters comprises one or more multi-cluster-member nodes that belong to at least two different of said clusters, and muting inter-cluster traffic through said multi-cluster-member nodes.

Abstract: A system and method for transporting data units for a communication network. In a communication between two peers a proxy and a network entity for controlling a partial communication are provided. One or both of the proxy and the network entity is arranged to modify one or more of the data units, in order to provide communication control related information that is different from communication control related information exchanged between the two peers The respective other or both of the proxy and the network entity is arranged to perform data unit transmission control based on the different communication control related information.

Abstract: Method and a device are disclosed for a fast performance of network operations via a network with high delay times by means of a module for processing system calls of an application layer and for initiating network operations of a network layer. In the module a differentiation between a blocking and non-blocking implementation mode is made. A non-blocking execution mode means that the considered system call returns a logical value as a result to the application, which signals whether the system call was successfully executed. In this case it is provided by the method and device to directly send a logical value to the application when a non-blocking system call is called, without having waited for the actual result of the operation executed in the communicating partner instance and corresponding to the system call. The handling of the results of the actually executed operations takes place at a later time.

Abstract: The invention relates to a method and a device for a fast performance of network operations via a network with high delay times by means of a module for processing system calls of an application layer and for initiating network operations of a network layer. In said module a differentiation between a blocking and a non-blocking implementation mode is made. A non-blocking execution mode means that the considered system call returns a logical value as result to the application, which signalizes whether the system call was successfully executed. In this case it is provided by the invention to directly send a logical value to the application when a non-blocking system call is called, without having waited for the actual result of the operation executed in the communicating partner instance and corresponding to the system call. The handling of the results of the actually executed operations takes place at a later time.