US 12,271,749 B2
Containerized workload scheduling
Aditi Ghag, Palo Alto, CA (US); Pranshu Jain, Palo Alto, CA (US); Yaniv Ben-Itzhak, Herzliya (IL); Sujata Banerjee, Palo Alto, CA (US); and Yongzhe Fan, Beijing (CN)
Assigned to VMWare LLC, Palo Alto, CA (US)
Filed by VMware LLC, Palo Alto, CA (US)
Filed on Apr. 25, 2019, as Appl. No. 16/394,663.
Prior Publication US 2020/0341789 A1, Oct. 29, 2020
Int. Cl. G06F 9/50 (2006.01); G06F 9/455 (2018.01); H04L 12/26 (2006.01); H04L 29/08 (2006.01); H04L 43/0876 (2022.01)
CPC G06F 9/45558 (2013.01) [H04L 43/0876 (2013.01); G06F 2009/45595 (2013.01)] 26 Claims
OG exemplary drawing
 
1. A method for containerized workload scheduling, comprising:
monitoring network traffic between a first containerized workload deployed on a node in a virtual computing environment (VCE) and other containerized workloads in the VCE to determine affinities between the first containerized workload and the other containerized workloads in the VCE, wherein the affinities are based on information derived from the monitoring of the network traffic between the first containerized workload and the other containerized workloads that change over time, such that the affinities characterize dynamic communication patterns of the network traffic between the first containerized workload and the other containerized workloads in the VCE to assess inter-containerized workload interactions in real time; and
scheduling, based, at least in part, on the determined affinities characterizing the dynamic communication patterns of the network traffic between the first containerized workload and the other containerized workloads, subsequent execution of a second containerized workload on the node on which the first containerized workload is deployed;
wherein the network traffic comprises flows characterized by flow sizes, the flows being classified flow types based on the flow size as mouse flows or elephant flows, the affinities being associated with the flow types.