发明名称 Methods and apparatus for efficient use of link aggregation groups
摘要 An apparatus includes a first edge device configured to receive a data unit destined to a peripheral processing device that is operatively coupled to a network interconnect via a LAG associated with a second edge device and a third edge device. The first edge device is configured to select an edge device set that includes the third edge device and excludes the second edge device, from a group of edge device sets. Each edge device set from the group of edge device sets is directly coupled to the peripheral processing device. The first edge device is configured to send an instance of the data unit to each edge device from the edge device set such that the third edge device sends an instance of the data unit to the peripheral processing device based on a selection method that omits ports on the second edge device as potential selections.
申请公布号 US9154371(B1) 申请公布日期 2015.10.06
申请号 US201313908310 申请日期 2013.06.03
申请人 Juniper Networks, Inc. 发明人 Srinivasan Shriram
分类号 H04L12/24;H04L12/803 主分类号 H04L12/24
代理机构 Cooley LLP 代理人 Cooley LLP
主权项 1. An apparatus, comprising: a first edge device configured to be included in a plurality of edge devices and configured to be operatively coupled to the remaining edge devices from the plurality of edge devices via a network interconnect, the first edge device configured to receive a data unit associated with a multicast group and to be sent to a peripheral processing device from a plurality of peripheral processing devices, the peripheral processing device being operatively coupled to the network interconnect via a link aggregation group (LAG) associated with a set of ports on a second edge device from the plurality of edge devices and a set of ports on a third edge device from the plurality of edge devices, the first edge device configured to select, in response to receiving the data unit, an edge device set (1) including the third edge device, (2) not including the second edge device, and (3) from a plurality of edge device sets associated with the multicast group, each edge device set from the plurality of edge device sets collectively being directly coupled to each peripheral processing device from the plurality of peripheral processing devices, each edge device set from the plurality of edge device sets being different than the remaining edge device sets from the plurality of edge device sets, the first edge device configured to send an instance of the data unit to each edge device associated with the edge device set such that the third edge device sends an instance of the data unit to the peripheral processing device via a port from the set of ports on the third edge device based on a selection method that omits the ports from the set of ports on the second edge device as potential selections.
地址 Sunnyvale CA US