Skip to main content

Using MarkLogic Content Pump (mlcp)

Concurrent mlcp Jobs

We do not recommend using concurrent mlcp jobs. Regardless of the version, mlcp does not support concurrent jobs if mlcp is importing from/exporting to the same data file. In addition, beginning in 10.0-4.2, each mlcp job uses the maximum number of threads available on the server as the default thread count. Therefore, using concurrent mlcp jobs will not improve performance, as one job is already using full concurrent capacity.

Here is more about threads and thread count:

  • A command line option called -max_threads refers to the maximum number of threads that run mlcp. This command line option is optional.

  • mlcp conducts initial polling to identify the available server threads on the port that handles mlcp requests. mlcp then uses half of this value as the default thread count.

  • You can overwrite this calculated value by specifying -thread_count in the command line.

  • If you specify -threads_per_split, each input split will run with the number you have specified. Note, however, that the total thread count is controlled by the newly calculated thread count or, if specified, -thread_count.

In this section: