keyed state are scaled by repartitioning keys • Operators with operator list state are scaled by redistributing the list entries. • Operators with operator broadcast state are scaled up by copying the The number of key groups limits the maximum number of parallel tasks to which keyed state can be scaled. • Trade-off between flexibility in rescaling and the maximum overhead involved in indexing and

throughput matches the data input rate • In the case of known aggregation functions, results can be scaled using approximate query processing techniques, where accuracy is measured in terms of relative error

restart • only temporarily block the affected dataflow subgraph • usually the operator to be scaled and upstream channels • All-at-once • move state to be migrated in one operation • high latency