Red Hat, Inc.

Markus Thoemmes

Francesco Guardiani

Roland Ludwig Huss

A method includes monitoring, during a first time interval, traffic associated with one or more applications executed by a cluster of compute nodes and determining, in view of the traffic associated with the one or more applications during the first time interval, that the traffic is predicted to exceed a capacity threshold of the cluster of compute nodes at an end of a second time interval. The method further includes initiating startup of an additional compute node to be added to the cluster of compute nodes for executing replicas of the one or more applications.

Trajectory-based hierarchical autoscaling for serverless applications

Systems and methods for determining if a controller that can service a custom resource (CR) exists are disclosed. The method includes evaluating a custom resource (CR) and determining whether a controller among a plurality of controllers can service the CR based on filter metadata associated with each of the plurality of controllers.

Metadata driven static determination of controller availability

Systems and methods for extending a container orchestration engine API in-process are disclosed. The method includes compiling each of one or more custom resource definition (CRD) controllers that are created in a cluster at run-time into a respective isolation module to generate one or more isolation modules, wherein the one or more isolation modules are all hosted in a service. The method also includes, in response to detecting an API event serviced by a CRD controller of the one or more CRD controllers, executing a respective isolation module of the CRD controller.

Extending the Kubernetes API in-process

Systems and methods for compiling functions into a single entity are disclosed. An execution graph representing dependencies between a plurality of functions is analyzed to identify portions of the graph that include two or more of the plurality of functions used as a combined entity and have a single entry point. The plurality of functions corresponds to a decomposed application. The processing device combines the functions of each of the identified one or more portions of the graph into a composition. For each portion of the graph that includes two or more of the plurality of functions used as a combined entity and has multiple entry points, the processing device determines whether to combine each function providing a subsequent entry point to the portion into a composition with other functions of the portion based at least in part on a network latency and a scale efficiency of the application.

Compiling monoglot function compositions into a single entity

Systems and methods for extending a container orchestration engine API in-process are disclosed. A processing device may compile each of one or more custom resource definition (CRD) controllers that are created in a cluster at run-time into a respective isolation module to generate one or more isolation modules, wherein the one or more isolation modules are all hosted in a service. The processing device may monitor for application program interface (API) events serviced by each of the one or more CRD controllers and in response to detecting an API event serviced by a CRD controller of the one or more CRD controllers, executing a respective isolation module of the CRD controller.

Extending the kubernetes API in-process

A method includes receiving a first request to execute a first instance of a function and generating a runtime container to execute the first instance of the function. The runtime instance includes a set of resources allocated to execute the function. The method further includes determining a consumption of the set of resources allocated to the runtime container in view of execution of the first instance of the function. The method further includes, in response to receiving a second request to execute a second instance of the function, initiating the second instance of the function within the runtime container in view of the consumption of the set of resources.

Function parallelism in a runtime container of a function-as-a-service (FAAS) system

Systems and methods for compiling functions into a single entity are disclosed. An execution graph representing dependencies between a plurality of functions is analyzed to identify portions of the graph that include two or more of the plurality of functions used as a combined entity and have a single entry point. The plurality of functions corresponds to a decomposed application. The processing device compiles the functions of each of the identified one or more portions of the graph into a composition. For each portion of the graph that includes two or more of the plurality of functions used as a combined entity and has multiple entry points, the processing device determines whether to compile each function providing a subsequent entry point to the portion into a composition with other functions of the portion based at least in part on a network latency and a scale efficiency of the application.

Systems and methods for determining if a controller that can service a custom resource (CR) exists are disclosed. A processing device annotates a corresponding deployment of each of a plurality of controllers with filter metadata obtained from the controller. The filter metadata of a controller comprises at least an object type that the controller is to service. In response to generating a CR, the processing device may compare the definitions of the CR with the filter metadata from each of the plurality of controllers, wherein the definitions of the CR comprise at least an object type of the CR. In response to determining that none of the plurality of controllers have filter metadata matching the definitions of the CR, the processing device may provide to a user a no-match alert indicating that there is no controller among the plurality of controllers that can service the CR.

Growing community of inventors

Grasbrunn, Germany

Markus Thoemmes