Nightly Regression Tests

The following regression tests are run nightly with Nyx. The plotfiles generated in each night’s test are compared with the benchmark plotfiles using the AMReX fcompare utility to compare the mesh data and particle_compare to compare the particle data.

The results of these tests can be found at https://ccse.lbl.gov/pub/RegressionTesting/Nyx. These tests are also run on an NVIDIA GPU and those results can be found at https://ccse.lbl.gov/pub/GpuRegressionTesting/Nyx.

We have a number of compile-time options – this chart summarizes which regression tests test which compile-time options. (Note that the USE_GRAV and AMREX_USE_PARTICLES macros which are used in the code source are both set to TRUE in the build system.)

Test

NO_HYDRO

USE_HEATCOOL

AMR-density

FALSE

TRUE

DR_restart

FALSE

FALSE

DoubleRarefaction

FALSE

FALSE

DrivenTurbulence

FALSE

FALSE
DrivenTurbulence-OMP

(CPU only)

FALSE

FALSE

LyA

FALSE

TRUE

LyA-adiabatic

FALSE

FALSE
LyA-OMP

(CPU only)

FALSE

TRUE
LyA_Neutrinos

(CPU only)

FALSE

TRUE

MiniSB

FALSE

FALSE

MiniSB-ppm

FALSE

FALSE

MiniSB-ref

FALSE

FALSE

MiniSB-ref-ppm

FALSE

FALSE

Part-2line

FALSE

FALSE

Part-2line-nbody

FALSE

FALSE

Part-2line-nomesh

TRUE

FALSE

Part-2line_restart

FALSE

FALSE

Part-2line_restart-nbody

FALSE

FALSE

Part-2line_restart-nomesh

TRUE

FALSE

Part-mass.nosub

FALSE

FALSE

Part-mass.nosub-nbody

FALSE

FALSE

Part-mass.nosub-nomesh

TRUE

FALSE

Part-mass.sub

FALSE

FALSE

Part-mass.sub-nbody

FALSE

FALSE

Part-mass.sub-nbody

FALSE

FALSE

Part-mass.sub-nomesh

TRUE

FALSE

Sedov

FALSE

FALSE

Sedov-ppm

FALSE

FALSE

Sod

FALSE

FALSE

Sod-ppm

FALSE

FALSE

StrongShockTube

FALSE

FALSE

StrongShockTube-ppm

FALSE

FALSE

Below nx,ny,nz are the number of cells in each coordinate direction at the coarsest level; Npa = number of particles, Np = number of MPI ranks / number of OMP threads per rank. (If just a single number then pure MPI.)

Test

nx ny nz

Nl

Npa

Np

What does this test?

AMR-density

64 64 64

3

262144

4

Start from chk00300

DR_restart

32 4 4

3

0

2/2

Hydro only - restart

DoubleRarefaction

32 4 4

3

0

2/2

Hydro only

DrivenTurbulence

32 32 32

1

0

4

Turbulent forcing
DrivenTurbulence-OMP

(CPU only)

32 32 32

1

0

1/4

Turbulent forcing with OMP

LyA

32 32 32

1

32768

4

Includes h/c
Lya-OMP

(CPU only)

32 32 32

1

32768

1/4

LyA with OMP

LyA-adiabatic

32 32 32

1

32768

4

No h/c
LyA_Neutrinos

(CPU only)

32 32 32

1

32768

1

LyA with OMP

MiniSB

32 32 32

1

2500

2

Small version of SB

MiniSB-ppm

32 32 32

1

2500

2

Small version of SB

MiniSB-ref

32 32 32

2

2500

2

Small version of SB with refinement

MiniSB-ref-ppm

32 32 32

2

2500

2

Small version of SB with refinement

Part-2line

16 16 16

3

16

2/2

Particle-only

Part-2line-nbody

16 16 16

3

16

2/2

Particle-only

Part-2line-nomesh

16 16 16

3

16

2/2

Particle-only

Part-2line_restart

16 16 16

3

16

2/2

Particle-only

Part-2line_restart-nbody

16 16 16

3

16

2/2

Particle-only

Part-2line_restart-nomesh

16 16 16

3

16

2/2

Particle-only

Part-mass.nosub

16 16 16

3

8

2/2

Particle-only

Part-mass.nosub-nbody

16 16 16

3

8

2/2

Particle-only

Part-mass.nosub-nomesh

16 16 16

3

8

2/2

Particle-only

Part-mass.sub

16 16 16

3

8

2/2

Particle-only

Part-mass.sub-nbody

16 16 16

3

8

2/2

Particle-only

Part-mass.sub-nomesh

16 16 16

3

8

2/2

Particle-only

Sedov

32 32 32

1

0

1

Hydro only (PLM)

Sedov-ppm

32 32 32

1

0

1

Hydro only (PPM)

Sod

4 32 4

3

0

1

Hydro only (PLM)

Sod-ppm

4 32 4

3

0

1/2

Hydro only (PPM)

StrongShockTube

32 4 4

3

0

1/2

Hydro only (PLM)

StrongShockTube-ppm

32 4 4

3

0

1

Hydro only (PPM)