Is there any way to speed up the search process?

[rhtownsend]

Rich has already explained why the calculation is inescapably slow but I'm curious about your models. Specifically, I'd say they don't actually include the structure of the neutron star. e.g. in ns10.GYRE, there's a point at the centre and then the remaining structure starts from 10⁶ cm (= 10 km). This probably differs from calculations involving a white dwarf, for which MESA can presumably include the white dwarf's actual structure.

If we could delete the central point, and do the calculation for the thin surface layer only, then I imagine the calculations might proceed much faster. Chen, is it possible to re-run your MESA calculations with add_center_point_to_pulse_data = .FALSE. in the &controls inlist? Then, post the new model.

cheers,

Rich

[cdd]

Hi Rich,

I attach the new model files in this letter. But when i use the new model in gyre calculation, it runs into error. Below is the error message:

ubuntu64@ubuntu64-virtual-machine:~/software/gyre/work$ ../bin/gyre test.in
gyre [5.2]
----------

Compiler : GCC version 8.3.0
Compiler options : -I /usr/include/x86_64-linux-gnu -I /home/ubuntu64/software/mesasdk/include -mtune=generic -march=x86-64 -ggdb -Wall -Wno-unused-dummy-argument -Wno-maybe-uninitialized -std=f2008 -ffpe-summary=none -finit-real=snan -fbacktrace -fmax-errors=25 -ffpe-trap=invalid,overflow,zero -fcheck=all -finline-limit=0 -fopenmp
OpenMP Threads : 2
Input filename : test.in

Model Init
----------

Reading from MESA file
File name ns10.GYRE
File version 1.01
Read 1766 points
Added central point

Mode Search
-----------

Mode parameters
l : 2
m : 0

Building frequency scan
added scan interval : 0.4810E-06 -> 0.4820E-06 (100 points, LINEAR)

Building x grid

Program received signal SIGFPE: Floating-point exception - erroneous arithmetic operation.

Backtrace for this error:

Program received signal SIGFPE: Floating-point exception - erroneous arithmetic operation.

Backtrace for this error:
#0 0x7f2e8483ff1f in ???
#1 0x448df7 in gamma_
at /home/ubuntu64/software/gyre/src/build/gyre_grid_util.f90:237
#2 0x4489df in __gyre_grid_util_MOD_eval_c_
at /home/ubuntu64/software/gyre/src/build/gyre_grid_util.f90:263
#3 0x6ae657 in __core_func_MOD_eval_r_
at /home/ubuntu64/software/gyre/src/build/core_func.f90:96
#4 0x6adff9 in __core_func_MOD_narrow_bracket_
at /home/ubuntu64/software/gyre/src/build/core_func.f90:361
#5 0x6ad7d1 in __core_func_MOD_root_r_
at /home/ubuntu64/software/gyre/src/build/core_func.f90:427
#6 0x44959c in __gyre_grid_util_MOD_find_turn
at /home/ubuntu64/software/gyre/src/build/gyre_grid_util.f90:167
#7 0x8bf113 in __gyre_grid_factory_MOD_add_inner_._omp_fn.1
at /home/ubuntu64/software/gyre/src/build/gyre_grid_factory.f90:249
#8 0x7f2e85276cfe in GOMP_parallel
at /root/sdk2/build/gcc/libgomp/parallel.c:168
#9 0x8bc2e2 in add_inner_
at /home/ubuntu64/software/gyre/src/build/gyre_grid_factory.f90:255
#10 0x8bd72d in __gyre_grid_factory_MOD_grid_t_grid_spec_
at /home/ubuntu64/software/gyre/src/build/gyre_grid_factory.f90:174
#11 0x8be603 in __gyre_grid_factory_MOD_grid_t_context_
at /home/ubuntu64/software/gyre/src/build/gyre_grid_factory.f90:106
#12 0x408188 in gyre
at /home/ubuntu64/software/gyre/src/build/gyre.f90:240
#13 0x40adaf in main
at /home/ubuntu64/software/gyre/src/build/gyre.f90:54
Floating point exception (core dumped)


I notice the model deletes the first node which has the radius 0, but why it core dumped?

Best Regards
Chen

[rhtownsend]

Hi Chen --

GYRE is attempting to add its own central point, and because it does this by extrapolation from the surface layers, it ends up with a 'bad' model.

The solution is to prevent the addition of this central point, by adding

Code: Select all

add_center = .false.

in the &model namelist group.

Because your model doesn't extend all the way to the origin, you'll also have to tell GYRE which inner boundary condition to use. I recommend setting the radial displacement to zero at the bottom of the envelope. To do this, add

Code: Select all

inner_bound = 'ZERO_R'

to the &osc namelist group.

cheers,

Rich

[cdd]

Hi Rich,

Good! After i add these two parameters in the inlist, the program runs fast. As a comparison, i run ns9 and ns10 (with add_center_point_to_pulse_data = .false. ) both once and the time is:

ns9:
Building frequency scan
added scan interval : 0.4800E-06 -> 0.4810E-06 (100 points, LINEAR)

Building x grid
Found inner turning points, x range 0.9988 -> 0.9988
Adding 0 inner point(s)
Adding 76795 global point(s) in iteration 1
Adding 0 global point(s) in iteration 2
Final grid has 1 segment(s) and 77688 point(s):
Segment 1 : x range 0.9988 -> 1.0000 (1 -> 77688)

Starting search (adiabatic)

Root bracketing.....
Time elapsed : 319.614 s

Root Solving.....
Time elapsed : 1436.729 s

ns10:
Building frequency scan
added scan interval : 0.4810E-06 -> 0.4820E-06 (100 points, LINEAR)

Building x grid
Found inner turning points, x range 0.9988 -> 0.9988
Adding 0 inner point(s)
Adding 68444 global point(s) in iteration 1
Adding 2787 global point(s) in iteration 2
Adding 0 global point(s) in iteration 3
Final grid has 1 segment(s) and 72997 point(s):
Segment 1 : x range 0.9988 -> 1.0000 (1 -> 72997)

Starting search (adiabatic)

Root bracketing.....
Time elapsed : 326.510 s

Root Solving.....
Time elapsed : 1083.328 s

It is so many times faster than before! Thank you, Rich.......Cheers!
And i'm waiting for your new release of tidal program~~~

Chen