Abstract: A streaming operation is performed by nodes of cluster that implement a database. A method includes a first node determining data segments from data in a first data file stored at the first node for transfer to a second node of the cluster. The first node generates segment offset data for each data segment defining an offset position of the data segment relative to positions in the first data file. The first node transfers sets of segment data, each set including a data segment and the segment offset data for the data segment, to the receiving node. The second writes the data segments to a second data file stored at the second node by mapping each data segment to a position in the second data file as defined by the offset position in the segment offset data for the data segment.

Abstract: Fault tolerant querying of data distributed across multiple nodes is accomplished by each node determining and reporting its own health status and indexing status to the other nodes in the cluster via a gossip protocol. A coordinator node then prioritizes replica nodes based on the received status of the other nodes and sends query requests to those nodes based on the prioritization. Should a node fail to provide an response to a query request, further query requests are sent to a next highest priority replica node containing the relevant data. This results in improved query performance by avoiding busy nodes and further provides a fault tolerant approach to data queries.

Abstract: Fault tolerant querying of data distributed across multiple nodes is accomplished by each node determining and reporting its own health status and indexing status to the other nodes in the cluster via a gossip protocol. A coordinator node then prioritizes replica nodes based on the received status of the other nodes and sends query requests to those nodes based on the prioritization. Should a node fail to provide an response to a query request, further query requests are sent to a next highest priority replica node containing the relevant data. This results in improved query performance by avoiding busy nodes and further provides a fault tolerant approach to data queries.