About Me
I'm a Staff Software Engineer at Kargo.
Patents
Managing operator pattern service connections to an environment
A method includes receiving a request from an operator pattern service to perform an operation on a computing environment and determining whether the operator pattern service has permission to perform the operation on the computing environment. The method further includes in response to determining that the operator pattern service has permission to perform the operation, providing the request to the computing environment for the operation to be performed on the computing environment.
Migrating the runtime state of a container between two nodes
Embodiments of the present disclosure relate to migrating the in-memory state of a containerized application to a destination node. A processing device may transmit an availability request to each of one or more nodes and identify a destination node to which a container is to be migrated based on a response from each of the one or more nodes indicating resources available to provision a replica of the container. The processing device may determine whether the destination node comprises a replica of each base layer of a set of base layers of the container and if not, transmit a replica of each base layer the destination node is missing to the destination node. A snapshot of the container may be acquired and the set of base layers may be removed from the snapshot to generate a replica of the in-memory layer, which may be transmitted to the destination node.
Detecting multiple operator pattern services making conflicting resource modifications
Systems and methods for detecting multiple operator pattern services making conflicting resource modifications. An example method comprises receiving from a first computing process running on a computing platform, a resource modification request to modify a resource of the computing platform. The method then determines, using a resource modification data store, whether a second computing process is regulating the resource associated with the resource modification request. Responsive to determining that the second computing process is regulating the resource associated with the resource modification request, the method notifies a managing process of the computing platform.
Dynamically adjusting reconciliation time-delays
Time delays used in a reconciliation process can be dynamically adjusted. For example, a system can receive a request from a client for a time delay value. The time delay value can be a timespan in which to wait between a first execution and a second execution of reconciliation software. The request can indicate a result of the first execution. In response to receiving the request, the system can select an algorithm from among a group of algorithms based on the result of the first execution. The system can then determine the time delay value by executing the algorithm. The system can transmit the time delay value to the client, which can wait for the timespan prior to initiating the second execution of the reconciliation software.