Executor options

Specify executor options for the dsbulk command. These options define how DataStax Bulk Loader executes queries, and allow users to control throughput, throttling, concurrency, and query method execution.

--executor.maxPerSecond, --dsbulk.executor.maxPerSecond number

The maximum number of concurrent operations per second. This fixed value setting acts as a safeguard to prevent more requests than the cluster can handle. To set a variable throughput limit, use the --dsbulk.executor.maxInFlight number option.

Batch statements are counted by the number of statements included. Reduce this setting when the latencies get too high and a remote cluster cannot keep up with throughput, as dsbulk requests will eventually time out. This setting applies to all operations. When writing to the database, it applies to the number of statements executed. When reading from the database, it applies to the number of rows retrieved.

For important related information, see --dsbulk.engine.maxConcurrentQueries.

Default: -1 [Starting in 1.6.0, disabled by default -1. Zero 0 or any negative value also disables this option.]

When connecting to DataStax Astra DB clusters, a rate limit is always enforced. By default, the limit is 4,096 operations per second and per coordinator. Therefore, when --executor.maxPerSecond is left to its default setting, and the cluster is an Astra DB cluster, DataStax Bulk Loader automatically applies an appropriate limit and logs a message. Increasing this setting to load data at a higher rate can trigger Astra DB rate limit errors. See Get support for Astra DB for details about how to get support and to raise the rate limit.

--executor.maxInFlight, --dsbulk.executor.maxInFlight number

The maximum number of "in-flight" requests, or maximum number of concurrent requests waiting for a response from the server. This acts as a safeguard to prevent more requests than the cluster can handle. Batch statements count as one request. Reduce this value if you are facing out-of-memory errors.

For important related information, see --dsbulk.engine.maxConcurrentQueries.

Default: -1 [Starting in 1.6.0, disabled by default]

--executor.continuousPaging.enabled, --dsbulk.executor.continuousPaging.enabled {true | false}

Enable or disable continuous paging. If the target cluster does not support continuous paging, or if --datastax-java-driver.basic.request.consistency is not ONE or LOCAL_ONE, traditional paging is used regardless of this setting. Can be used with unload and count operations. Not applicable for load.

Default: true

Executor options

--executor.maxPerSecond, --dsbulk.executor.maxPerSecond number

--executor.maxInFlight, --dsbulk.executor.maxInFlight number

--executor.continuousPaging.enabled, --dsbulk.executor.continuousPaging.enabled {true | false}

--executor.continuousPaging.maxConcurrentQueries, --dsbulk.executor.continuousPaging.maxConcurrentQueries number

--executor.continuousPaging.maxPages, --dsbulk.executor.continuousPaging.maxPages number

--executor.continuousPaging.maxPagesPerSecond, --dsbulk.executor.continuousPaging.maxPagesPerSecond number

--executor.continuousPaging.pageSize, --dsbulk.executor.continuousPaging.pageSize number

--executor.continuousPaging.pageUnit, --dsbulk.executor.continuousPaging.pageUnit {ROWS | BYTES}