2 docs tagged with "compute"

Three compute techniques that run on the node that owns the key: broadcast with local scan, single-key affinityCall, and cross-cache affinityCall. A scale experiment from 1,000 to 100,000 customers prints wall-clock and byte counts for each approach against pull-to-client.

Ship Java closures to the nodes that own the data instead of pulling the data to the client. Measure pull-to-client, broadcast, and affinityCall against the same question on a three-node cluster.