Maximum Threads

Default value: 62
rosepy parameter name: max_threads

The maximum threads option specifies the maximum number of threads the solver can use when solving the optimization model. Each thread allows the solver to work on a different part of the problem at the same time, parallelizing the solution process across the available CPU cores for improved performance on large or complex problems.

The maximum value is 62. Each compute node provides 64 physical CPU cores, and 2 are reserved for other resources, leaving 62 available to the solver.

Increasing the number of threads typically leads to faster solve times by enabling broader and deeper exploration of the search tree in parallel. This is especially beneficial for mixed-integer programming (MIP) models. A value of 1 means the solver will run on a single thread. Setting this to a higher value scales the solver across additional cores, up to the maximum of 62.

If you would like to increase your compute capacity, reach out to us.