Abstract: The present embodiments relate to updating a dataflow interactive cluster with zero downtime. A request to update a first dataflow cluster can be received, and a second dataflow cluster can be generated as a replacement cluster to execute received queries. Generating the second dataflow cluster can include identifying a second series of executor nodes that are configured to execute queries from the gateway node via a second driver node. A first update to a configuration of a host configuration node can be performed to register the second dataflow cluster as an active endpoint and identify the first dataflow cluster as an inactive endpoint. When no active queries exist, a second update to the configuration can be provided to remove the first dataflow cluster from the configuration to direct subsequent queries from the gateway node to the second dataflow cluster.

Abstract: A distributed computing system is described that leverages a nearline storage layer to minimize the downtime required for bootstrapping a new computing cluster in the distributed computing system. The system executes a computing cluster comprising a set of computing nodes and determines a set of one or more data segments to be written to a nearline storage system. The system writes the data segments to the nearline storage system. In certain examples, the system receives a request to create a second computing cluster and responsive to the request, bootstraps the second computing cluster using the set of data segments stored on the nearline storage system. The system additionally leverages the nearline storage layer to accelerate query processing by the computing nodes of a computing cluster.

Abstract: A distributed computing system is described that leverages a nearline storage layer to minimize the downtime required for bootstrapping a new computing cluster in the distributed computing system. The system executes a computing cluster comprising a set of computing nodes and determines a set of one or more data segments to be written to a nearline storage system. The system writes the data segments to the nearline storage system. In certain examples, the system receives a request to create a second computing cluster and responsive to the request, bootstraps the second computing cluster using the set of data segments stored on the nearline storage system. The system additionally leverages the nearline storage layer to accelerate query processing by the computing nodes of a computing cluster.

Abstract: A distributed computing system is described that leverages a nearline storage layer to minimize the downtime required for bootstrapping a new computing cluster in the distributed computing system. The system executes a computing cluster comprising a set of computing nodes and determines a set of one or more data segments to be written to a nearline storage system. The system writes the data segments to the nearline storage system. In certain examples, the system receives a request to create a second computing cluster and responsive to the request, bootstraps the second computing cluster using the set of data segments stored on the nearline storage system. The system additionally leverages the nearline storage layer to accelerate query processing by the computing nodes of a computing cluster.

Abstract: Techniques for providing improved distributed caching are disclosed. A distributed computing system can be implemented with a cluster including a plurality of worker nodes configured to host one or more executors for processing data related to a query. The worker nodes can host a cache accessible to the executors. The data can be processed as a plurality of data segments. The worker nodes can be uniformly assigned a plurality of token bounds defining a range of integer token values. A hashing algorithm can be used to compute a token for each data segment associated with the query. Tasks can be launched on the executors preferentially, such that the task for processing a data segment having a token within the token bounds associated with the preferred executor. Executors can be instructed to review the associated cache to identify outlier data segments and inform other nodes in the cluster.

Abstract: The present embodiments relate to updating a dataflow interactive cluster with zero downtime. A request to update a first dataflow cluster can be received, and a second dataflow cluster can be generated as a replacement cluster to execute received queries. Generating the second dataflow cluster can include identifying a second series of executor nodes that are configured to execute queries from the gateway node via a second driver node. A first update to a configuration of a host configuration node can be performed to register the second dataflow cluster as an active endpoint and identify the first dataflow cluster as an inactive endpoint. When no active queries exist, a second update to the configuration can be provided to remove the first dataflow cluster from the configuration to direct subsequent queries from the gateway node to the second dataflow cluster.

Abstract: The present embodiments relate to updating a dataflow interactive cluster with zero downtime. A request to update a first dataflow cluster can be received, and a second dataflow cluster can be generated as a replacement cluster to execute received queries. Generating the second dataflow cluster can include identifying a second series of executor nodes that are configured to execute queries from the gateway node via a second driver node. A first update to a configuration of a host configuration node can be performed to register the second dataflow cluster as an active endpoint and identify the first dataflow cluster as an inactive endpoint. When no active queries exist, a second update to the configuration can be provided to remove the first dataflow cluster from the configuration to direct subsequent queries from the gateway node to the second dataflow cluster.

Abstract: A distributed computing system is described that leverages a nearline storage layer to minimize the downtime required for bootstrapping a new computing cluster in the distributed computing system. The system executes a computing cluster comprising a set of computing nodes and determines a set of one or more data segments to be written to a nearline storage system. The system writes the data segments to the nearline storage system. In certain examples, the system receives a request to create a second computing cluster and responsive to the request, bootstraps the second computing cluster using the set of data segments stored on the nearline storage system. The system additionally leverages the nearline storage layer to accelerate query processing by the computing nodes of a computing cluster.

Abstract: The present disclosure relates to system and techniques for prediction of failures in resources deployed in a data plane of a cloud based infrastructure. The resource are selected from a plurality of cloud based resources arranged in a hierarchical manner and allocated to a client device. A predictor employs a first prediction model to obtain a first prediction of a failure of a resource, and a second prediction model to obtain a second prediction of the failure of the resource. Weights are assigned to the first prediction and second prediction based at least in part on a criterion. The predictor computes an overall prediction of the failure of the resource based at least in part on at least one of the first prediction, the second prediction or the respective weights assigned to the predictions. The overall prediction is utilized for restoring the failure of the resource.

Abstract: Systems, devices, and methods discussed herein are directed to intelligently adjusting the set of worker nodes within a computing cluster. By way of example, a computing device (or service) may monitor performance metrics of a set of worker nodes of a computing cluster. When a performance metric is detected that is below a performance threshold, the computing device may perform a first adjustment (e.g., an increase or decrease) to the number of nodes in the cluster. Training data may be obtained based at least in part on the first adjustment and utilized with supervised learning techniques to train a machine-learning model to predict future performance changes in the cluster. Subsequent performance metrics and/or cluster metadata may be provided to the machine-learning model to obtain output indicating a predicted performance change. An additional adjustment to the number of worker nodes may be performed based at least in part on the output.

Abstract: Systems, devices, and methods discussed herein are directed to intelligently adjusting the set of worker nodes within a computing cluster. By way of example, a computing device (or service) may monitor performance metrics of a set of worker nodes of a computing cluster. When a performance metric is detected that is below a performance threshold, the computing device may perform a first adjustment (e.g., an increase or decrease) to the number of nodes in the cluster. Training data may be obtained based at least in part on the first adjustment and utilized with supervised learning techniques to train a machine-learning model to predict future performance changes in the cluster. Subsequent performance metrics and/or cluster metadata may be provided to the machine-learning model to obtain output indicating a predicted performance change. An additional adjustment to the number of worker nodes may be performed based at least in part on the output.

Abstract: Systems, devices, and methods discussed herein are directed to intelligently adjusting the set of worker nodes within a computing cluster. By way of example, a computing device (or service) may monitor performance metrics of a set of worker nodes of a computing cluster. When a performance metric is detected that is below a performance threshold, the computing device may perform a first adjustment (e.g., an increase or decrease) to the number of nodes in the cluster. Training data may be obtained based at least in part on the first adjustment and utilized with supervised learning techniques to train a machine-learning model to predict future performance changes in the cluster. Subsequent performance metrics and/or cluster metadata may be provided to the machine-learning model to obtain output indicating a predicted performance change. An additional adjustment to the number of worker nodes may be performed based at least in part on the output.

Abstract: Techniques for providing improved distributed caching are disclosed. A distributed computing system can be implemented with a cluster including a plurality of worker nodes configured to host one or more executors for processing data related to a query. The worker nodes can host a cache accessible to the executors. The data can be processed as a plurality of data segments. The worker nodes can be uniformly assigned a plurality of token bounds defining a range of integer token values. A hashing algorithm can be used to compute a token for each data segment associated with the query. Tasks can be launched on the executors preferentially, such that the task for processing a data segment having a token within the token bounds associated with the preferred executor. Executors can be instructed to review the associated cache to identify outlier data segments and inform other nodes in the cluster.

Abstract: A distributed computing system is described that leverages a nearline storage layer to minimize the downtime required for bootstrapping a new computing cluster in the distributed computing system. The system executes a computing cluster comprising a set of computing nodes and determines a set of one or more data segments to be written to a nearline storage system. The system writes the data segments to the nearline storage system. In certain examples, the system receives a request to create a second computing cluster and responsive to the request, bootstraps the second computing cluster using the set of data segments stored on the nearline storage system. The system additionally leverages the nearline storage layer to accelerate query processing by the computing nodes of a computing cluster.

Abstract: The present disclosure relates to system and techniques for prediction of failures in resources deployed in a data plane of a cloud based infrastructure. The resource are selected from a plurality of cloud based resources arranged in a hierarchical manner and allocated to a client device. A predictor employs a first prediction model to obtain a first prediction of a failure of a resource, and a second prediction model to obtain a second prediction of the failure of the resource. Weights are assigned to the first prediction and second prediction based at least in part on a criterion. The predictor computes an overall prediction of the failure of the resource based at least in part on at least one of the first prediction, the second prediction or the respective weights assigned to the predictions. The overall prediction is utilized for restoring the failure of the resource.

Abstract: The present embodiments relate to updating a dataflow interactive cluster with zero downtime. A request to update a first dataflow cluster can be received, and a second dataflow cluster can be generated as a replacement cluster to execute received queries. Generating the second dataflow cluster can include identifying a second series of executor nodes that are configured to execute queries from the gateway node via a second driver node. A first update to a configuration of a host configuration node can be performed to register the second dataflow cluster as an active endpoint and identify the first dataflow cluster as an inactive endpoint. When no active queries exist, a second update to the configuration can be provided to remove the first dataflow cluster from the configuration to direct subsequent queries from the gateway node to the second dataflow cluster.

Abstract: A distributed computing system is described that leverages a nearline storage layer to minimize the downtime required for bootstrapping a new computing cluster in the distributed computing system. The system executes a computing cluster comprising a set of computing nodes and determines a set of one or more data segments to be written to a nearline storage system. The system writes the data segments to the nearline storage system. In certain examples, the system receives a request to create a second computing cluster and responsive to the request, bootstraps the second computing cluster using the set of data segments stored on the nearline storage system. The system additionally leverages the nearline storage layer to accelerate query processing by the computing nodes of a computing cluster.

Abstract: Systems, devices, and methods discussed herein are directed to intelligently adjusting the set of worker nodes within a computing cluster. By way of example, a computing device (or service) may monitor performance metrics of a set of worker nodes of a computing cluster. When a performance metric is detected that is below a performance threshold, the computing device may perform a first adjustment (e.g., an increase or decrease) to the number of nodes in the cluster. Training data may be obtained based at least in part on the first adjustment and utilized with supervised learning techniques to train a machine-learning model to predict future performance changes in the cluster. Subsequent performance metrics and/or cluster metadata may be provided to the machine-learning model to obtain output indicating a predicted performance change. An additional adjustment to the number of worker nodes may be performed based at least in part on the output.

Abstract: Embodiments of the invention employ a distributed algorithm to enable sensor nodes in a sensor network to adaptively self-configure into sensor clusters to provide a desired higher-level functionality.

Abstract: A software system with a two tier arrangement for threads support that enhances the adaptability of a virtual machine to differing platforms. The software system includes a virtual machine with a threads interface layer having a set of methods that provide thread support in the virtual machine according to a standard threads interface associated with the virtual machine. The software system includes a native threads interface layer that provides a set of methods that adapt the methods of the threads interface layer to a platform which underlies the software system. The native threads interface layer shields the virtual machine from the particulars of the underlying operating system while the threads interface layer provides a stable interface for application programs and other tasks that benefit from thread support in the virtual machine.