US 12,236,269 B2
Distribution of workloads in cluster environment using server warranty information
Vaideeswaran Ganesan, Bangalore (IN); Praveen Lalgoudar, Bangalore (IN); Rushyendra Velamuri, Bangalore (IN); Winston X. Fernandes, Sirsi (IN); Pandiyan Varadharajan, Bangalore (IN); Rekha M S, Karnataka (IN); and Abhishek Gupta, Rohini (IN)
Assigned to Dell Products, L.P., Round Rock, TX (US)
Filed by Dell Products, L.P., Round Rock, TX (US)
Filed on Aug. 24, 2021, as Appl. No. 17/409,930.
Claims priority of application No. 202111033260 (IN), filed on Jul. 23, 2021.
Prior Publication US 2023/0044503 A1, Feb. 9, 2023
Int. Cl. G06F 9/50 (2006.01); G06F 11/20 (2006.01); G06N 20/00 (2019.01)
CPC G06F 9/5005 (2013.01) [G06F 11/2051 (2013.01); G06N 20/00 (2019.01); G06F 2201/87 (2013.01)] 11 Claims
OG exemplary drawing
 
1. A method for managing workloads assigned to nodes in a data center, comprising:
assigning a primary node among a cluster of nodes, wherein each of the nodes are Information Handling Systems (IHS) hosting a plurality of virtual machines (VM), and wherein each workload is assigned to run on a selected VM in the cluster, the primary node monitoring a status of the nodes in the cluster;
identifying, by the primary node, nodes within the cluster that are in a quarantine mode, wherein a node is placed in quarantine mode when a potential node failure is detected but the node is still operating;
determining a remaining warranty time for the nodes in quarantine status by retrieving warranty information for each component in the node from a database within a vendor warranty system;
determining an expected duration for a selected workload;
determining a criticality of the selected workload by retrieving a workload tag assigned to the workload by a user of the cluster;
assigning, by the primary node, the selected workload to a node in the cluster based upon the remaining warranty time, the expected workload duration, and the workload criticality to ensure that the selected workload is completed within an assigned node's expected lifetime; and
executing the selected workload on the assigned node in the cluster.