Distribution of workloads among online analytics processing systems
Technology for routing queries in a system with a plurality of nodes (for example online analytical processing sub-systems) where each node has an associated replicated local database and a local latency value and replication velocity values. The workload balancing for incoming received queries among and between the plurality of nodes is based, at least in part, on consideration of latency values and/or replication velocity values for the various nodes. The best node to handle a given query is thereby selected and the query is routed to the selected node for response.
1. A computer-implemented method (CIM) for distributing query response workload in an online processing system including a plurality of accelerators, with each accelerator including a workload balance module and a plurality of online processing sub-systems and with each online processing subsystem respectively including a database management module, a replicated version of a database and a server configured to respond to queries, the CIM comprising:
for each online processing sub-system of the plurality of online processing systems, receiving a current performance characteristics data set including at least: (i) an update latency value indicative of a latency time associated with the given online processing sub-system occurring during intermittent database replication updates, and (ii) an update replication velocity value indicative of a speed at which the given online processing system receives data during intermittent database replication updates;
during each intermittent database replication update, replicating update information to each accelerator of the plurality of accelerators;
adjusting each workload balance module of each accelerator based on the update information;
subsequent to the adjusting of each workload balance module, receiving, by a first workload balance module of a first accelerator of the plurality of accelerators, a query requesting selected information from the database; and
selecting, by machine logic of the first workload balance module, a target online processing sub-system for responding to the query based on the current performance data sets of the plurality of online processing sub-systems, with the selection of the target online processing sub-system including consideration of at least the update latency values and the update replication velocity values.