Abstract: The current document is directed to methods and systems that carry out live migration of multi-node applications in distributed-computing systems, including cloud-computing applications in cloud-computing facilities. The disclosed methods and systems employ an installer/agent and a data-transformation component to install new versions of a multi-node application within a distributed computing system and to then warm start each application-provided service within the newly installed new-version application nodes. Following a data sync operation, each warm-started application service is brought online to service subsequently received service requests as older-version corresponding services of current application nodes complete request services and terminate, providing live migration from an older version multi-node application to a new-version multi-node application without service-provision interruption.

Abstract: The present disclosure provides an approach for training a machine learning model by first training the model on a generic dataset and then iteratively training the model on “easy” domain specific training data before moving on to “difficult” domain specific training data. Inputs of a domain-specific dataset are run on the generically-trained model to determine which inputs generate an accuracy score above a threshold. The inputs with an accuracy score above a threshold are used to retrain the model, along with the corresponding outputs. The retraining continues until all domain specific dataset has been used to train the model, or until no remaining inputs of the domain specific dataset generate an accuracy score, when run on the model, that is above a threshold.

Abstract: The present disclosure provides an approach for training a machine learning model by first training the model on a generic dataset and then iteratively training the model on “easy” domain specific training data before moving on to “difficult” domain specific training data. Inputs of a domain-specific dataset are run on the generically-trained model to determine which inputs generate an accuracy score above a threshold. The inputs with an accuracy score above a threshold are used to retrain the model, along with the corresponding outputs. The retraining continues until all domain specific dataset has been used to train the model, or until no remaining inputs of the domain specific dataset generate an accuracy score, when run on the model, that is above a threshold.

Abstract: The current document is directed to methods and systems that carry out live migration of multi-node applications in distributed-computing systems, including cloud-computing applications in cloud-computing facilities. The disclosed methods and systems employ an installer/agent and a data-transformation component to install new versions of a multi-node application within a distributed computing system and to then warm start each application-provided service within the newly installed new-version application nodes. Following a data sync operation, each warm-started application service is brought online to service subsequently received service requests as older-version corresponding services of current application nodes complete request services and terminate, providing live migration from an older version multi-node application to a new-version multi-node application without service-provision interruption.