Abstract: The present disclosure provides for matching users into user groups based on user attributes. A neural network is trained using training data including user data of users in an observed grouping and various outcomes or optimization parameters associated with the observed grouping. The trained neural network is used to process a set of queried players determine a match quality for the queried players with respect to certain outcomes or optimization parameters based at least in part on the user data associated with the users. The match quality is a measure of how optimized a group containing the queried users is with respect to the optimization parameters and can be compared to a threshold. The queried users may be matched into a group if the match quality meets or exceeds the threshold.

Abstract: Techniques for singleton coordination in an actor-based system are described herein. In some examples, one or more actors within an actor based system may be designated as singleton actors. Each singleton actor is permitted to have only a single associated instance in the actor-based system. In some cases, when multiple executing instances (i.e., duplicates) of a singleton actor are detected, the duplicates may be resolved, for example, such that no more than a single instance of the singleton actor is retained. In some examples, one or more singleton coordinators may be responsible for selecting one or more hubs on which to generate new or re-generated singletons, monitoring existing singletons to ensure that they continue to execute, and re-generating existing singletons after termination.

Abstract: Remote computing resource service providers, including online retailer, provide externally facing computer systems that allow users to interact with the service provider. Furthermore, the service provider may maintain computer systems and service inside an isolated network not exposed to users. Occasionally, service providers may test these externally facing computer systems using one or more external hosts operating on a public network. A coordinator may archive aggregated testing resources located on the isolated network and deploy the testing resources to the one or more external hosts. A database may be used to track the deployment state of the one or more external hosts in order to allow the coordinator to manage the testing process.

Abstract: Techniques for debugging of a collection of instructions executed by a computing system including a set of actors are described herein. A first sub-set of actors from the set of actors to be monitored for debugging and one or more error condition associated with the collection of instructions may be identified. A set of indications including indications of messages processed by the actors, indications of associated actor states prior to processing of the messages, and indications of associated actor states subsequent to processing of the messages may be received from the sub-set of actors. It may then be determined that a first actor of the sub-set of actors has satisfied the error condition. Based at least in part on the set of indications, one or more messages associated with satisfaction of the error condition may then be identified, for example for re-processing and debugging.

Abstract: Techniques for singleton coordination in an actor-based system are described herein. In some examples, one or more actors within an actor based system may be designated as singleton actors. Each singleton actor is permitted to have only a single associated instance in the actor-based system. In some cases, when multiple executing instances (i.e., duplicates) of a singleton actor are detected, the duplicates may be resolved, for example, such that no more than a single instance of the singleton actor is retained. In some examples, one or more singleton coordinators may be responsible for selecting one or more hubs on which to generate new or re-generated singletons, monitoring existing singletons to ensure that they continue to execute, and re-generating existing singletons after termination.

Abstract: Remote computing resource service providers, including online retailer, provide externally facing computer systems that allow users to interact with the service provider. Furthermore, the service provider may maintain computer systems and service inside an isolated network not exposed to users. Occasionally, service providers may test these externally facing computer systems using one or more external hosts operating on a public network. A coordinator may archive aggregated testing resources located on the isolated network and deploy the testing resources to the one or more external hosts. A database may be used to track the deployment state of the one or more external hosts in order to allow the coordinator to manage the testing process.

Abstract: Remote computing resource service providers, including online retailer, provide externally facing computer systems that allow users to interact with the service provider. Furthermore, the service provider may maintain computer systems and service inside an isolated network not exposed to users. Occasionally, service providers may test these externally facing computer systems using one or more external hosts operating on a public network. A coordinator may archive aggregated testing resources located on the isolated network and deploy the testing resources to the one or more external hosts. A database may be used to track the deployment state of the one or more external hosts in order to allow the coordinator to manage the testing process.

Abstract: Techniques for tracking, profiling and recognizing attributes associated with transactions associated with computer system services. A bitmap representation of the unique identifiers created from name-value pairs of attributes is created and one or more operations are performed on the representation to process the bitmap representation. The one or more operations include operations to, for example, compress the bitmap representation, sub-sample the bitmap representation, locate patterns of attributes within the bitmap representation and other such processing operations. The processed bitmaps may then be further queried to analyze the event data.

Abstract: Techniques for debugging of a collection of instructions executed by a computing system including a set of actors are described herein. A first sub-set of actors from the set of actors to be monitored for debugging and one or more error condition associated with the collection of instructions may be identified. A set of indications including indications of messages processed by the actors, indications of associated actor states prior to processing of the messages, and indications of associated actor states subsequent to processing of the messages may be received from the sub-set of actors. It may then be determined that a first actor of the sub-set of actors has satisfied the error condition. Based at least in part on the set of indications, one or more messages associated with satisfaction of the error condition may then be identified, for example for re-processing and debugging.

Abstract: Remote computing resource service providers, including online retailers, provide externally facing computer systems that allow users to interact with the service provider. Furthermore, the service provider may maintain computer systems and services inside an isolated network not exposed to users. Occasionally, service providers may test these externally facing computer systems using one or more external hosts operating on a public network. As a result side effects may occur within the isolated network that is not accessible from the public network. The side effects may be recorded in a queue which may be accessed from inside the isolated network by one or more services. The one or more services may then reverse the side effects based at least in part on information contained in the queue.

Abstract: Techniques for singleton coordination in an actor-based system are described herein. In some examples, one or more actors within an actor based system may be designated as singleton actors. Each singleton actor is permitted to have only a single associated instance in the actor-based system. In some cases, when multiple executing instances (i.e., duplicates) of a singleton actor are detected, the duplicates may be resolved, for example, such that no more than a single instance of the singleton actor is retained. In some examples, one or more singleton coordinators may be responsible for selecting one or more hubs on which to generate new or re-generated singletons, monitoring existing singletons to ensure that they continue to execute, and re-generating existing singletons after termination.