Abstract:
An information handling system is provided. The information handling system includes a first hypervisor running on a first host and a second hypervisor running on a second host. The first hypervisor managing a first virtual switch, and the second hypervisor managing a second virtual switch. The information handling system also includes a plurality of virtual machines (VMs), including a first VM, which is part of a first tenant, running on the first host, and a second VM, part of a second tenant, running on the second host. The first virtual switch has a mapping in memory that maps a customer-specific multicast IP address, used by the plurality of VMs to indicate a multicast group that includes VMs on the first and second tenants, to a global multicast IP address used by the first and second hosts.
Abstract:
Aspects of the present invention include an n-node link aggregation group (LAG) system comprising a set of N nodes collectively provide a logical fabric-level view that is consistent across the set of N nodes. Embodiments of the n-node system comprise a control plane mechanism to provide Layer 2 multipathing between access network devices and the core network. The n-node system provides a loop-free topology with active-active load-sharing of uplinks from access to the core.
Abstract:
An information handling system is provided. The information handling system includes a first hypervisor running on a first host and a second hypervisor running on a second host. The first hypervisor managing a first virtual switch, and the second hypervisor managing a second virtual switch. The information handling system also includes a plurality of virtual machines (VMs), including a first VM, which is part of a first tenant, running on the first host, and a second VM, part of a second tenant, running on the second host. The first virtual switch has a mapping in memory that maps a customer-specific multicast IP address, used by the plurality of VMs to indicate a multicast group that includes VMs on the first and second tenants, to a global multicast IP address used by the first and second hosts.
Abstract:
An information handling system is provided. The information handling system includes a first hypervisor running on a first host and a second hypervisor running on a second host. The first hypervisor managing a first virtual switch, and the second hypervisor managing a second virtual switch. The information handling system also includes a plurality of virtual machines (VMs), including a first VM, which is part of a first tenant, running on the first host, and a second VM, part of a second tenant, running on the second host. The first virtual switch has a mapping in memory that maps a customer-specific multicast IP address, used by the plurality of VMs to indicate a multicast group that includes VMs on the first and second tenants, to a global multicast IP address used by the first and second hosts.
Abstract:
An information handling system is provided. The information handling system includes a plurality of forwarding processors, the plurality of forwarding processors each including a memory having a forwarding host table and a forwarding route table. The information handling system also includes at least one switching device coupled to the plurality of forwarding processors, the at least one switching device including a memory having a destination module and port table, a switching device host table and a switching device route table, wherein the at least one switching device is configured to perform an extended lookup in at least one of the switching device host table and switching device route table based on values included in a fabric header of information routed to the switching device to determine a destination of the information. A method for extending lookup tables is also provided that adds lookup tables in a switching device.
Abstract:
A programmable data plane hardware load balancing system includes a plurality of server devices and an edge networking device that is located on an edge of a local area network. The edge networking device includes programmable data plane hardware that integrates each of at least one communications port, a server device table, and a load balancing engine. The at least one communications port is coupled to the plurality of server devices. The server device table includes health information about the plurality of server devices. The load balancing engine is configured to receive traffic directed to the plurality of server devices through the at least one communication port, and has been programmed to forward the traffic to one of the plurality of server devices in response to performing health based load balancing on the traffic using the health information about the plurality of server devices in the server device table.
Abstract:
A peer-to-peer wireless controller topology provisioning system includes plurality of networking devices. A plurality of peer-to-peer wireless controllers are included in each networking device and are each configured to control a portion of a single wireless link with another peer-to-peer wireless controller. A management subsystem determines relative locations for each of the networking devices, and a number of peer-to-peer wireless controllers available in each of the networking devices. The management subsystem then uses a maximum hop constraint and a minimum bandwidth constraint to generate a peer-to-peer wireless controller topology for at least some of the peer-to-peer wireless controllers that are available in the networking devices. The management subsystem then provides the peer-to-peer wireless controller topology to each of the networking devices to cause each of the networking devices to configure their available peer-to-peer wireless controllers to provide wireless links according to the peer-to-peer wireless controller topology.
Abstract:
Aspects of the present invention include increasing the number of hops that can be specifically defined in a multiprotocol label switching stack. In embodiments of the present invention, a label space can be used to represent two or more labels. In embodiments of the present invention, the label space can be used by concatenating two or more labels and redefining the multiprotocol label switching stack operations and outgoing labels.
Abstract:
Aspects of the present invention include an n-node link aggregation group (LAG) system comprising a set of N nodes collectively provide a logical fabric-level view that is consistent across the set of N nodes. Embodiments of the n-node system comprise a control plane mechanism to provide Layer 2 multipathing between access network devices and the core network. The n-node system provides a loop-free topology with active-active load-sharing of uplinks from access to the core.
Abstract:
Aspects of the present invention include extending routing capabilities to improve networking efficiencies. In embodiments, a spine-leaf network configuration may be provisioned as a single router to solve at least two issues: (1) local routing within a leaf; and (2) scalability of the hardware ARP table. In embodiments, one or more tables in the leaf node may be programmed so that data traffic coming from a first host in a first subnet or VLAN on the leaf and intended for a second host in a second subnet or VLAN on the leaf may be forwarded locally to the second host in the second subnet or VLAN without traversing to the spine layer.