US 11,929,897 B2
Highly-scalable, software-defined, in-network multicasting of load statistics data
Muhammad Shahbaz, Santa Clara, CA (US); Mukesh Hira, Palo Alto, CA (US); and Lalith Suresh, Mountain View, CA (US)
Assigned to VMware, Inc., Palo Alto, CA (US)
Filed by VMware LLC, Palo Alto, CA (US)
Filed on Oct. 25, 2021, as Appl. No. 17/510,053.
Application 17/510,053 is a continuation of application No. 16/432,477, filed on Jun. 5, 2019, granted, now 11,184,259.
Prior Publication US 2022/0045923 A1, Feb. 10, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. H04L 43/062 (2022.01); H04L 45/16 (2022.01); H04L 47/125 (2022.01)
CPC H04L 43/062 (2013.01) [H04L 45/16 (2013.01); H04L 47/125 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for highly-scalable, in-network multicasting of load statistics data, the method comprising:
receiving, at a physical switch, a packet comprising statistics data for virtual machines executing on a host, the statistics data being associated with a multicast (“MC”) group;
performing multicast distribution of packets, the performing comprising:
determining that the packet comprises a special header corresponding to the MC group;
determining that the special header includes an identifier of the physical switch;
identifying a bitmap associated with the identifier of the physical switch; and
using the bitmap, determining one or more ports of the physical switch to replicate the packet on; and
providing the packet to an interface controller for transmitting the packet to the port of the physical switch; and
balancing a service load across a set of servers using a plurality of load balancers by sending multicast packets to the plurality of load balancers via the physical switch by including the special header, the multicast packets including load statistics data of the set of servers.