cluster_table_function_buckets_batch_size
Defines the approximate size of a batch (in bytes) used in distributed processing of tasks in cluster table functions withbucket split granularity. The system accumulates data until at least this amount is reached. The actual size may be slightly larger to align with data boundaries.
cluster_table_function_pushed_down_filter
This is an internal setting that should not be used directly. The initiator of a query with a cluster table function sets it for the replicas to the serialized filter on the table, so that the replicas drop the rows that do not pass it before sending them.cluster_table_function_split_by_split_offsets
Withbucket split granularity, when the table metadata records the file size and the row group offsets (split_offsets in Iceberg), cut files into byte ranges on these offsets without reading them; each task reads the row groups that start in its range. Otherwise, and on a hit in the query condition cache, the initiator reads the footer of every file.
Requires all servers of the cluster to support it: a task with a byte range cannot be sent to a server of an older version.
cluster_table_function_split_granularity
Controls how data is split into tasks when executing a CLUSTER TABLE FUNCTION. This setting defines the granularity of work distribution across the cluster:file— each task processes an entire file.bucket— tasks are created per internal data block within a file (for example, Parquet row groups).
bucket) can improve parallelism when working with a small number of large files.
For instance, if a Parquet file contains multiple row groups, enabling bucket granularity allows each group to be processed independently by different workers.