Dell Products Inc.

Edward Guohua Ding

Jim LeWei Ji

Muzhar S Khokhar

Tianming Zhang

Lewei Ji

Drake Yuan Qiu

An information handling system may include at least one processor and a memory. The information handling system may be configured to: add a node to a cluster of information handling systems; while the node is a member of the cluster, cause a plurality of components of the node to be changed according to a first set of changes; for each change in the first set of changes, store a record of the change; and in response to an indication that the node is to be removed from the cluster, cause the plurality of components of the node to be changed according to a second set of changes, wherein the changes in the second set of changes are configured to undo the changes in the first set of changes.

Scaling cluster without re-imaging nodes

An information handling system may receive a request from a particular remote cluster regarding a cluster scaling event; receive first information from a plurality of other remote clusters indicative of a success or a failure of a corresponding cluster expansion event that was performed at such other remote clusters; receive second information from the plurality of other remote clusters indicative of scores for such other remote clusters in a plurality of metrics; determine, based on the first and second information, a ranking of the metrics based on their criticality to the cluster scaling event; receive third information from the particular remote cluster indicative of scores for the particular remote cluster in the plurality of metrics; and determine a likelihood of success for the cluster scaling event based on the determined ranking of the metrics and the scores for the particular remote cluster in the plurality of metrics.

Knowledge base for predicting success of cluster scaling

An information handling system may include at least one processor and a non-transitory memory coupled to the at least one processor. The information handling system may be configured to: transmit a request to a remote information handling system regarding an upcoming cluster scaling event for an information handling system cluster; receive information from the remote information handling system indicative of a plurality of known-good cluster configurations; determine a closest one of the plurality of known-good cluster configurations; make one or more configuration changes to the information handling system cluster to cause a configuration of the information handling system cluster to match the closest one; and initiate the cluster scaling event.

Knowledge base for correcting baseline for cluster scaling

A Hyper-Converged Infrastructure (HCl) system that includes a plurality of HCl log generating components and an HCl storage system that provides at least a portion of a log database. The HCl system receives a request from a management system to store a first log bundle of the plurality of HCl log generating components and determines the at least one second log bundle that is stored in the log database is at least a size threshold. The HCl system performs a log database clean operation on the at least one second log bundle and determines that the log database clean operation on the at least one second log bundle has provided an available storage capacity in the log database that is sufficient to store the first log bundle. The HCl system then stores the first log bundle in the log database.

Hyper-converged infrastructure (HCI) log system

A Hyper-Converged Infrastructure (HCI) system that includes a plurality of HCI log generating components and an HCI storage system that provides at least a portion of a log database. The HCI system receives a request from a management system to store a first log bundle of the plurality of HCI log generating components and determines the at least one second log bundle that is stored in the log database is at least a size threshold. The HCI system performs a log database clean operation on the at least one second log bundle and determines that the log database clean operation on the at least one second log bundle has provided an available storage capacity in the log database that is sufficient to store the first log bundle. The HCI system then stores the first log bundle in the log database.

Growing community of inventors

Shanghai, China

Edward Guohua Ding