Linear solvers

To improve performance, Ipopt supports multiple linear solvers, the default one is MUMPS but you can modify the linear solver with the keyword argument linear_solver.

HSL

Obtain a license and download HSL_jll.jl from https://licences.stfc.ac.uk/products/Software/HSL/LibHSL.

There are two versions available: LBT and OpenBLAS. LBT is the recommended option for Julia ≥ v1.9.

Install this download into your current environment using:

import Pkg
Pkg.develop(path = "/full/path/to/HSL_jll.jl")

We provide an example with the linear solvers MA27 and MA57:

using HSL_jll, NLPModelsIpopt
stats_ma27 = ipopt(nlp, linear_solver="ma27")
stats_ma57 = ipopt(nlp, linear_solver="ma57")

SPRAL

If you use NLPModelsIpopt.jl with Julia ≥ v1.9, the linear solver SPRAL is available. You can use it by setting the linear_solver attribute:

using NLPModelsIpopt
stats_spral = ipopt(nlp, linear_solver="spral")

Note that the following environment variables must be set before starting Julia:

export OMP_CANCELLATION=TRUE
export OMP_PROC_BIND=TRUE

BLAS and LAPACK

With Julia v1.9 or later, Ipopt and the linear solvers MUMPS (default), SPRAL, and HSL are compiled with libblastrampoline (LBT), a library that can change between BLAS and LAPACK backends at runtime.

The default BLAS and LAPACK backend is OpenBLAS.

Using LBT, we can also switch dynamically to other BLAS backends such as Intel MKL and Apple Accelerate. Because Ipopt and the linear solvers heavily rely on BLAS and LAPACK routines, using an optimized backend for a particular platform can improve the performance.

MKL

If you have MKL.jl installed, switch to MKL by adding using MKL to your code:

using MKL  # Replace OpenBLAS by Intel MKL
using NLPModelsIpopt

AppleAccelerate

If you are using macOS ≥ v13.4 and you have AppleAccelerate.jl installed, you can replace OpenBLAS as follows:

using AppleAccelerate  # Replace OpenBLAS by Apple Accelerate
using NLPModelsIpopt

Display backends

Check what backends are loaded using:

import LinearAlgebra
LinearAlgebra.BLAS.lbt_get_config()