International Business Machines Corporation

Shaival Jagdishbhai Chokshi

Omar Cardona

Xiaohan Qin

Patrick Tam Vo

Adekunle Bello

Rakesh Sharma

Francisco Jesus Alanis

Kavitha Vittal Murthy Baratakke

Vishal Ramachandra Mansur

Van Hoa Lee

Nikhil Hegde

David R Willoughby

David Lee Randall

Ashwini Kulkarni

Thi Ngoc Tran

A system, and computer program product for leader device selection in a control cluster using a shared virtual local area network (VLAN) are provided in the illustrative embodiments. A broadcast Hello packet is received from a sender virtual device over the VLAN at a first virtual device. When the first virtual device satisfies a partnership criterion specified in the Hello packet, a first ranking of the first virtual device is evaluated to be higher than a ranking of a sender virtual device. A first Live packet is unicast over the VLAN to the sender virtual device and a leadership role in the control cluster is taken by the first virtual device. The sender virtual device is caused to stop a broadcasting operation at the sender virtual device. A second Hello packet is broadcasted from the first virtual device over the VLAN.

Leader device selection in control clusters using shared VLAN

At a first virtual device executing in a data processing system, a determination is made whether a partner list associated with the first virtual device includes a threshold number of partner virtual devices, the first virtual device participating in the control cluster. A first Live packet is unicast over the VLAN to each partner virtual device identified in the partner list. A unicast of a second Live packet from a second virtual device to the first virtual device is received over the VLAN, the second virtual device being identified in the partner list. A Hello packet is broadcast over the VLAN to cause the Hello packet to reach a third virtual device, wherein the Hello packet is configured to discover a new partner virtual device to add to the partner list, discovering the new partner enlarging the control cluster.

Partner discovery in control clusters using shared VLAN

Mechanisms are provided for performing traffic load balancing on ingress traffic directed to a Link Aggregation Group (LAG). The mechanisms monitor a ingress traffic load across a plurality of links of the Link Aggregation Group (LAG). The mechanisms determine if the ingress traffic load across the plurality of links is unbalanced. Moreover, the mechanisms, in response to determining that the ingress traffic load across the plurality of links is unbalanced, send a message to a switch associated with the LAG requesting the switch to modify routing of ingress traffic to the LAG to perform ingress traffic load balancing.

Autonomic traffic load balancing in link aggregation groups by modification of switch ingress traffic routing

A method and technique for data flow affinity for heterogeneous virtual machines includes: connecting first and second logical partitions to a shared virtual adapter, the first logical partition having a maximum receive unit (MRU) different than a MRU of the second logical partition; responsive to receiving data by the shared virtual adapter destined for the first or second logical partitions, performing transmission control protocol (TCP) segmentation offload processing of the data by the shared virtual adapter utilizing the MRU of the first logical partition as the transmission unit for data destined for the first logical partition and utilizing the MRU of the second logical partition as the transmission unit for data destined for the second logical partition; and disabling large receive offload (LRO) processing of data by a physical adapter if either of the first or second logical partitions does not support LRO.

Data flow affinity for heterogenous virtual machines

Autonomic ingress traffic load balancing in link aggregation groups by modification of switch routing

A method for leader device selection in a control cluster using a shared virtual local area network (VLAN) is provided in the illustrative embodiments. A broadcast Hello packet is received from a sender virtual device over the VLAN at a first virtual device. When the first virtual device satisfies a partnership criterion specified in the Hello packet, a first ranking of the first virtual device is evaluated to be higher than a ranking of a sender virtual device. A first Live packet is unicast over the VLAN to the sender virtual device and a leadership role in the control cluster is taken by the first virtual device. The sender virtual device is caused to stop a broadcasting operation at the sender virtual device. A second Hello packet is broadcasted from the first virtual device over the VLAN.

Embodiments of the present invention provide methods, systems, and computer program products that enable secure network communications with logical partitions. A gateway between a physical network adapter and at least one virtual network trunk adapter receives a packet. The gateway tags the packet with an indication of an origin of the packet. The gateway delivers the tagged packet to an intrusion prevention system for intrusion analysis. When the gateway receives the tagged packet from the intrusion prevention system, the gateway forwards the tagged packet according to the indication of origin of the tagged packet.

Securing network communications with logical partitions

Provided are techniques for active SEA learning about a client LPAR MAC addresses via address resolution protocol (ARP) packets received on a virtual interface (of the active SEA). Any new client MAC addresses learned on the active SEA are sent to the inactive SEA via a control channel. When SEA failover happens, as the previously inactive SEA is about to become active, it will first send out RARP (reverse ARP) packets with the client MAC addresses as the source MAC addresses respectively. This effectively informs the switch connected to the previously inactive SEA that these client MAC addresses are to be routed through this switch port; the client MAC addresses saved on the switch connected to the previously active SEA are cleared as a result.

SEA failover mechanism with minimized packet losses

Autonomic traffic load balancing in link aggregation groups

Byte utilization is improved in Remote Direct Memory Access (RDMA) communications by detecting a plurality of concurrent messages on a plurality of application sockets which are destined for the same application, client or computer, intercepting those messages and consolidating their payloads into larger payloads, and then transmitting those consolidated messages to the destination, thereby increasing the payload-to-overhead byte utilization of the RDMA transmissions. At the receiving end, multiplexing information is used to unpack the consolidated messages, and to put the original payloads into a plurality of messages which are then fed into the receiving sockets to the destination application, client or computer, thereby making the consolidation process transparent between the initiator and the target.

Remote direct memory access socket aggregation

According to one aspect of the present disclosure, a method and technique for data flow affinity for heterogeneous virtual machines is disclosed. The method includes: connecting first and second logical partitions to a shared virtual adapter, the first logical partition having a maximum receive unit (MRU) different than a MRU of the second logical partition; and responsive to receiving data by the shared virtual adapter destined for the first or second logical partitions, performing transmission control protocol (TCP) segmentation offload processing of the data by the shared virtual adapter utilizing the MRU of the first logical partition as the transmission unit for data destined for the first logical partition and utilizing the MRU of the second logical partition as the transmission unit for data destined for the second logical partition.

According to one aspect of the present disclosure a system and technique for data flow affinity for heterogeneous virtual machines includes a hypervisor operable to manage a virtualized environment comprising first and second logical partitions, the first logical partition having a maximum receive unit (MRU) different than a MRU of the second logical partition. The system also includes a virtual input/output server (VIOS) having a shared virtual adapter connected to the first and second logical partitions. The shared virtual adapter is configured to, responsive to receiving data destined for the first or second logical partitions, perform transmission control protocol (TCP) segmentation offload processing of the data utilizing the MRU of the first logical partition as the transmission unit for data destined for the first logical partition and utilizing the MRU of the second logical partition as the transmission unit for data destined for the second logical partition.

Growing community of inventors

Round Rock, TX, United States of America

Shaival Jagdishbhai Chokshi