International Business Machines Corporation

Kai Hong Du

Shashi Pratap Singh

Lu Zhang

Zi Jin Tao

Ying Lin Xu

Xiao Li Xu

Xiao Jian Nie

Heng Guo Ge

Lin Wei Wu

Lei Zhang

Gang Tang

Chen Bo Li

Guo Chun Bian

Le Od Zhang

Da Shen

ShiMing Qu

Umesh Chandra Sahoo

Fu Hui Li

Jiu Xing Nie

A computer-implemented method, according to one approach, includes: determining whether a destination for a domain name system (DNS) query corresponds to an existing source network address translation (SNAT) port in response to receiving the DNS query. In response to determining that the destination for the DNS query corresponds to an existing SNAT port, the DNS query is modified to incorporate the existing SNAT port. A map entry corresponding to the existing SNAT port is also updated, and the modified DNS query is satisfied. Other systems, methods, and computer program products are described in additional approaches.

Using a source network address translation (SNAT) port to facilitate multiple queries

Methods, computer program products, and systems are presented. The method computer program products, and systems can include, for instance: receiving, by a Network Virtualization Edge (NVE) of a first virtual network (VN), a multicast packet from a virtual machine (VM) of the first virtual network (VN); determining, by the NVE, whether the NVE currently supports forwarding the multicast packet to a second NVE of a second virtual network (VN), the second virtual network (VN) being logically separated from the first virtual network (VN); responsively to determining that the NVE currently supports forwarding the multicast packet to a second NVE of a second virtual network (VN) forwarding, by the NVE, the multicast packet to the second NVE.

Multicast traffic across virtual networks (VNs)

Methods, computer program products, and systems are presented. The methods include, for instance: a gateway stack that includes a master, a backup, and at least one slave amongst nodes of the GS based on an election pursuant to a gateway stack protocol. The gateway stack provides gateway services for a Network Virtualization over Layer 3 (NVO3) network in a fail-safe manner by utilizing all of the nodes in the gateway stack. A data interface between the gateway stack and a switch is aggregated to evenly distribute inbound packets amongst the nodes of the gateway stack.

Distributed virtual gateway appliance

A method, system, and computer program product for enabling communication between different overlay solutions. The method includes one or more processor obtaining an address resolution request from a first resource in a first domain for a second resource in a second domain. Both domains are in a shared computing environment and the second domain includes a control plane engine. The one or more processor obtains source information from the request, registers the information to the engine, and utilizes the request to query the engine to locate the second resource in the domain. The second resource matches the request. The processor receives a response from the engine identifying a first host and the first host is in the second domain and is the host of the second resource. The processor sends a notification request to the first host requesting that the first host reply to the request.

Domain intercommunication in shared computing environments

Methods, computer program products, and systems are presented. The methods include, for instance: providing a distributed virtual gateway for Network Virtualization over Layer 3 (NVO3) network. A gateway stack having three or more nodes is implemented as a distributed virtual gateway, providing Layer 2 or Layer 3 gateway services in a fail-safe manner. Nodes of the gateway stack are configured to autonomously process and forward inbound NVO3 data packets with known destination addresses without engaging a master of the gateway stack.

In an approach to routing and forwarding a data packet with a forwarding RBridge and a corresponding RBridge, a computer, in response to receiving a TRILL data packet, performs the following steps: parsing a TRILL header of the TRILL data packet, and acquiring an egress RBridge nickname of the TRILL data packet; obtaining routing information corresponding to the egress RBridge nickname, wherein the information indicates whether the forwarding RBridge is the penultimate hop RBridge on the forwarding path to which the egress RBridge nickname is corresponding; and in response to the indication in the information that the forwarding RBridge is the penultimate hop RBridge on the forwarding path to which the egress RBridge nickname is corresponding, stripping the TRILL header of the TRILL data packet off to obtain an original data packet, and forwarding the original data packet to the egress RBridge from an out port specified by the information.

RBridge for routing and forwarding a data packet

A mechanism for multicast (M-VIM) is used to resolve address looking up problems for different types of VXLAN related overlay networks for multicast and broadcast traffics. A Multicast Overlay Ferry Proxy (MOFP) is introduced between a VXLAN domain and the CNI domain for ferrying control plane and data panel information. The M-VIM mechanism comprises three parts: a VXLAN Module; a CNI Module; and a proxy mechanism.

Growing community of inventors

Shanghai, China

Kai Hong Du