Abstract: Operations in a query plan generated for a query may be scaled when performed at a query engine. The performance of an operation in the query plan may be monitored to detect a scaling event. A performance distribution for the operation may be determined among a modified number of processing resources. The operation may then continue performing at the modified number of processing resources according to the determined performance distribution.

Abstract: Systems are disclosed for modifying weighting of database components by a query planner based on component volatility. The systems obtain a plurality of snapshots of one or more datasets that include one or more data components, and from the plurality of snapshots, the system identifies the volatility of each data component. The systems further generate a plurality of separate dynamic weights for each data component and initiates application of the separate dynamic weights by the query planner to the one or more data components. In some instances, the weights differentially affect the computational cost for query plans based in a time-dependent fashion and are applied prior to calculating costs of the query plans.

Abstract: Implementing a new node controller in a cluster environment. A method includes, at a primary node controller in the cluster environment, deploying a secondary node controller to be updated, associated with a flight container. The secondary node controller appears logically to the primary node controller as resources available to the primary node controller to which the primary node controller can assign cluster jobs. The primary node controller delegates cluster jobs to the secondary node controller. Sub-containers are started in the flight container for the secondary node controller to start delegated cluster jobs. Performance of the secondary node controller is monitored with respect to the started jobs.

Abstract: A relational database system may include a query optimizer and a query execution engine. The query optimizer may be configured to receive a query from a query-generating entity and to determine a sequence of operations for executing the query. The query execution engine may use real-time statistics to select physical operators for performing the sequence of operations.

Abstract: Systems are disclosed for modifying weighting of database components by a query planner based on component volatility. The systems obtain a plurality of snapshots of one or more datasets that include one or more data components, and from the plurality of snapshots, the system identifies the volatility of each data component. The systems further generate a plurality of separate dynamic weights for each data component and initiates application of the separate dynamic weights by the query planner to the one or more data components. In some instances, the weights differentially affect the computational cost for query plans based in a time-dependent fashion and are applied prior to calculating costs of the query plans.

Abstract: Scheduling garbage collection operations. A set of nodes are identified in a cluster. A relative factor of the nodes in the set of nodes is identified. A target length of time in which to complete a garbage collection process is identified. A subset of the set of nodes in which the garbage collection process could be completed in the identified target length of time is selected. Selecting the subset includes selecting nodes for the garbage collection process based on the relative factor and a probability that the garbage collection process will be completed within the identified target length of time while attempting to maximize an amount of garbage that can be collected. Garbage collection on the subset of the set of nodes is initiated.

Abstract: Implementing a new node controller in a cluster environment. A method includes, at a primary node controller in the cluster environment, deploying a secondary node controller to be updated, associated with a flight container. The secondary node controller appears logically to the primary node controller as resources available to the primary node controller to which the primary node controller can assign cluster jobs. The primary node controller delegates cluster jobs to the secondary node controller. Sub-containers are started in the flight container for the secondary node controller to start delegated cluster jobs. Performance of the secondary node controller is monitored with respect to the started jobs.