Abstract: This patent application relates to an agile network architecture that can be employed in data centers, among others. One implementation provides a virtual layer-2 network connecting machines of a layer-3 infrastructure.

Abstract: Among other things, one or more techniques and/or system are provided for routing a data packet. Because the data packet may correspond to a source device having an address on a first network that may be similar to an address of a second device on a second network, the data packet may be translated to create a transformed data packet that may be distinguishable from data packets of the second device. For example, the data packet may be translated from a first address format, such as IPv4, to a second address format, such as IPv6, to create a transformed data packet. The transformed data packet may comprise a prefix that may distinctly identify the source device during routing. In this way, the transformed data packet may be identifiable as corresponding to the source device during routing through a network to a destination device.

Abstract: Among other things, one or more techniques and/or system are provided for routing a data packet. Because the data packet may correspond to a source device having an address on a first network that may be similar to an address of a second device on a second network, the data packet may be translated to create a transformed data packet that may be distinguishable from data packets of the second device. For example, the data packet may be translated from a first address format, such as IPv4, to a second address format, such as IPv6, to create a transformed data packet. The transformed data packet may comprise a prefix that may distinctly identify the source device during routing. In this way, the transformed data packet may be identifiable as corresponding to the source device during routing through a network to a destination device.

Abstract: A source device obtains a data packet that includes both a destination address and a payload. The source device selects an exit point address of multiple exit point addresses corresponding to the destination address based on one or more policies. The source device encapsulates the data packet with a header that includes the selected exit point address, and the encapsulated data packet is provided to the backbone network. The encapsulated data packet is routed through the backbone network based on the exit point address, and an edge router of the backbone network identifies an interface of the edge router that corresponds to the exit point address. The header is removed from the encapsulated data packet, and the data packet is added to a buffer of the interface for routing to one or more other devices outside of the backbone network.

Abstract: Among other things, one or more techniques and/or system are provided for routing a data packet. Because the data packet may correspond to a source device having an address on a first network that may be similar to an address of a second device on a second network, the data packet may be translated to create a transformed data packet that may be distinguishable from data packets of the second device. For example, the data packet may be translated from a first address format, such as IPv4, to a second address format, such as IPv6, to create a transformed data packet. The transformed data packet may comprise a prefix that may distinctly identify the source device during routing. In this way, the transformed data packet may be identifiable as corresponding to the source device during routing through a network to a destination device.

Abstract: A method of networking a plurality of servers together within a data center is disclosed. The method includes the step of addressing a data packet for delivery to a destination server by providing the destination server address as a flat address. The method further includes the steps of obtaining routing information required to route the packet to the destination server. This routing information may be obtained from a directory service servicing the plurality of servers. Once the routing information is obtained, the data packet may be routed to the destination server according to the flat address of the destination server and routing information obtained from the directory service.

Abstract: Computing unit enclosures are often configured to connect units (e.g., server racks or trays) with a wired network. Because the network type may vary (e.g., Ethernet, InfiniBand, and Fibre Channel), such enclosures often provide network resources connecting each unit with each supported network type. However, such architectures may present inefficiencies such as unused network resources, and may constrain network support for the units to a small set of supported network types. Presented herein are enclosure architectures enabling flexible and efficient network support by including a backplane comprising a backplane bus that exchanges data between the units and a network adapter using an expansion bus protocol, such as PCI-Express.

Abstract: Among other things, one or more techniques and/or system are provided for routing a data packet. Because the data packet may correspond to a source device having an address on a first network that may be similar to an address of a second device on a second network, the data packet may be translated to create a transformed data packet that may be distinguishable from data packets of the second device. For example, the data packet may be translated from a first address format, such as IPv4, to a second address format, such as IPv6, to create a transformed data packet. The transformed data packet may comprise a prefix that may distinctly identify the source device during routing. In this way, the transformed data packet may be identifiable as corresponding to the source device during routing through a network to a destination device.

Abstract: An intelligent lookup service for a network is provided for clients of a network requesting services of the network that intelligently determines, based on a service requirement of the requested service, optimal service endpoint(s) for providing the requested service. The intelligent lookup service can incorporate predetermined mapping policy and traffic measurements into the determination. In addition, a feedback loop is provided from clients and/or service endpoints to the lookup service concerning measurements about prior connections in the network. The lookup service can include a set of beacons distributed in the network and against which measurements about the network are recorded. A client receives, from the lookup service in response to a request for a network address, a set of candidate service endpoints that pertain to the requested network address and the client connects to one of the candidate service endpoints based on policy or context.

Abstract: Exemplary methods, systems, and computer program products describe selecting a gateway based on health and performance information of a plurality of gateways. The techniques describe gateways advertising health and performance information, computing devices creating a table of this health and performance information, and selecting a gateway using the table. In response to changes in the health and performance information, the computing device may select a different gateway. The process allows network traffic load to be distributed across a plurality of gateways. This process further provides resilience by allowing a plurality of active gateways to substitute for a non-functioning gateway.

Abstract: A source device obtains a data packet that includes both a destination address and a payload. The source device selects an exit point address of multiple exit point addresses corresponding to the destination address based on one or more policies. The source device encapsulates the data packet with a header that includes the selected exit point address, and the encapsulated data packet is provided to the backbone network. The encapsulated data packet is routed through the backbone network based on the exit point address, and an edge router of the backbone network identifies an interface of the edge router that corresponds to the exit point address. The header is removed from the encapsulated data packet, and the data packet is added to a buffer of the interface for routing to one or more other devices outside of the backbone network.

Abstract: A system for commoditizing data center networking is disclosed. The system includes an interconnection topology for a data center having a plurality of servers and a plurality of nodes of a network in the data center through which data packets may be routed. The system uses a routing scheme where the routing is oblivious to the traffic pattern between nodes in the network, and wherein the interconnection topology contains a plurality of paths between one or more servers. The multipath routing may be Valiant load balancing. It disaggregates the function of load balancing into a group of regular servers, with the result that load balancing server hardware can be distributed amongst racks in the data center leading to greater agility and less fragmentation. The architecture creates a huge, flexible switching domain, supporting any server/any service, full mesh agility, and unregimented server capacity at low cost.

Abstract: A Totally Stubby Edge (TSE) participates in a cloud under the condition that the TSE may select paths for frames that the TSE introduces to the cloud, but may not transit frames between nodes in the cloud. The edge submits, to an administrator of the cloud, a request to join the cloud. If the administrator allows the request, then the edge is given access to the address tables that define the structure of the cloud, and may insert itself into the structure. The edge may use the address tables to select paths for frames that the edge introduces to the cloud. Normally path decisions are made by devices that the administrator trusts and controls. However, since the TSE selects paths for its own frames but does not transit frames between other nodes in the cloud, the TSE may select paths even if it is untrusted by the administrator.

Abstract: An enterprise namespace may be extended into a cloud of networked resources. A portion of the cloud may be dynamically partitioned, and the extension of the enterprise namespace established within the portion. Cloud resources thus remain as easily accessible to enterprise users as those which are physically located on the enterprise network. Thus, components such as applications, virtual machine instantiations, application states, server states, etc., may be easily migrated between the enterprise network and the cloud.

Abstract: This patent application relates to an agile network architecture that can be employed in data centers, among others. One implementation provides a virtual layer-2 network connecting machines of a layer-3 infrastructure.

Abstract: Exemplary methods, systems, and computer program products describe selecting a gateway based on health and performance information of a plurality of gateways. The techniques describe gateways advertising health and performance information, computing devices creating a table of this health and performance information, and selecting a gateway using the table. In response to changes in the health and performance information, the computing device may select a different gateway. The process allows network traffic load to be distributed across a plurality of gateways. This process further provides resilience by allowing a plurality of active gateways to substitute for a non-functioning gateway.

Abstract: Systems and methods that distribute load balancing functionalities in a data center. A network of demultiplexers and load balancer servers enable a calculated scaling and growth operation, wherein capacity of load balancing operation can be adjusted by changing the number of load balancer servers. Accordingly, load balancing functionality/design can be disaggregated to increase resilience and flexibility for both the load balancing and switching mechanisms of the data center.

Abstract: A Totally Stubby Edge (TSE) participates in a cloud under the condition that the TSE may select paths for frames that the TSE introduces to the cloud, but may not transit frames between nodes in the cloud. The edge submits, to an administrator of the cloud, a request to join the cloud. If the administrator allows the request, then the edge is given access to the address tables that define the structure of the cloud, and may insert itself into the structure. The edge may use the address tables to select paths for frames that the edge introduces to the cloud. Normally path decisions are made by devices that the administrator trusts and controls. However, since the TSE selects paths for its own frames but does not transit frames between other nodes in the cloud, the TSE may select paths even if it is untrusted by the administrator.

Abstract: A system for commoditizing data center networking is disclosed. The system includes an interconnection topology for a data center having a plurality of servers and a plurality of nodes of a network in the data center through which data packets may be routed. The system uses a routing scheme where the routing is oblivious to the traffic pattern between nodes in the network, and wherein the interconnection topology contains a plurality of paths between one or more servers. The multipath routing may be Valiant load balancing. It disaggregates the function of load balancing into a group of regular servers, with the result that load balancing server hardware can be distributed amongst racks in the data center leading to greater agility and less fragmentation. The architecture creates a huge, flexible switching domain, supporting any server/any service, full mesh agility, and unregimented server capacity at low cost.

Abstract: A method of networking a plurality of servers together within a data center is disclosed. The method includes the step of addressing a data packet for delivery to a destination server by providing the destination server address as a flat address. The method further includes the steps of obtaining routing information required to route the packet to the destination server. This routing information may be obtained from a directory service servicing the plurality of servers. Once the routing information is obtained, the data packet may be routed to the destination server according to the flat address of the destination server and routing information obtained from the directory service.