Skip to content

Commit

Permalink
Merge pull request #43 from bhourahine/bhTypos
Browse files Browse the repository at this point in the history
Minor typesetting bugs
  • Loading branch information
bhourahine committed Jun 15, 2022
2 parents 354f95b + 8f7986b commit 2be9397
Show file tree
Hide file tree
Showing 17 changed files with 262 additions and 283 deletions.
Binary file modified docs/_figures/linresp/abs-emi.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified docs/_figures/linresp/homo.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified docs/_figures/linresp/lumo.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified docs/_figures/linresp/macromol_abs.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified docs/_figures/linresp/no-tio2_abs.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified docs/_figures/linresp/no-tio2_geo.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified docs/_figures/moleculardynamics/md/md_spectra.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
460 changes: 208 additions & 252 deletions docs/_figures/moleculardynamics/md/md_spectra.svg
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
29 changes: 16 additions & 13 deletions docs/interfaces/ipi/ipi.rst
Expand Up @@ -114,7 +114,7 @@ Geometry optimisation

[Input: `recipes/interfaces/ipi/zundel/`]

An example using a Zundel ion (H:sub:`5`O:sub:`2`:sup:`+`) with
An example using a Zundel ion (H\ :sub:`5`\ O\ :sub:`2`:sup:`+`) with
on-site corrected DFTB2 embedded in a implicit water solvent is
provided as an example. The implicit solvent lacks the
hydrogen-bonding network that would be present in real water.
Expand Down Expand Up @@ -174,18 +174,21 @@ continued for longer) is shown below
400 steps is insufficent to thermalise the system to the target
temperature, but the dynamics of the individual beads, and the *path
centroid* (i.e., the average position of the quantum particles) are
output, along with data gathered during the run in the files ::

+----------------------+----------------+
|simulation.xc.xyz | path centroid |
+----------------------+----------------+
|RESTART |restart data |
+----------------------+----------------+
|simulation.out |collected data |
+----------------------+----------------+
|simulation.pos_\*.xyz |individual beads|
+----------------------+----------------+

output, along with data gathered during the run in the files

+-----------------------+-----------------+
| File | Contents |
+=======================+=================+
| simulation.xc.xyz | path centroid |
+-----------------------+-----------------+
| RESTART | restart data |
+-----------------------+-----------------+
| simulation.out | collected data |
+-----------------------+-----------------+
| simulation.pos_*.xyz | individual beads|
+-----------------------+-----------------+


The frequency at which these files are appended is set in `nvtPI.xml`

Stopping programs
Expand Down
6 changes: 3 additions & 3 deletions docs/linresp/index.rst
@@ -1,8 +1,8 @@
.. _linear-response:

###################
Linear response
###################
###########################
Linear response excitations
###########################

In this chapter we will go through a couple of recipes for the
computation of excited state properties using linear response
Expand Down
1 change: 0 additions & 1 deletion docs/linresp/relax.rst
Expand Up @@ -137,7 +137,6 @@ illustrates the energetic landscape:

.. _fig_aee:
.. figure:: ../_figures/linresp/abs-emi.png
:height: 36ex
:align: center
:alt: homo real

Expand Down
3 changes: 2 additions & 1 deletion docs/moleculardynamics/analysis.rst
Expand Up @@ -57,7 +57,8 @@ frequencies). The spectrum data is in frequency inits of wavenumbers
:align: center
:alt: Autocorrelation spectra of vibrations

Anharmonic spectra
Anharmonic spectra from velocity autocorrelation (excluding large
peak at ~0 cm\ :sup:`-1`.

Try different sizes of the window and see what happens to the spectra.

Expand Down
13 changes: 7 additions & 6 deletions docs/moleculardynamics/annealing.rst
Expand Up @@ -44,9 +44,10 @@ Annealing to multiple minima
[Input: `recipes/moleculardynamics/annealV2/`]
The annealing process can also be used to find alternative local minina. Here
two vacancies in a graphene sheet are heated. Depending on the initial
conditions, the system anneals to different final structures. The starting
velocities are chosen at random, so depending on the seed value for the random
generator (several different cases are given in the input file) a different
final defect geometry is obtained by the same cycle of heating and cooling.
The annealing process can also be used to find alternative local
minina. Here two vacancies in a graphene sheet are heated. Depending
on the initial conditions, this system anneals to different final
structures. The starting velocities are chosen at random, so depending
on the seed value for the random number generator (several different
cases are given in the input file) a different final defect geometry
is obtained by the same cycle of heating and cooling.
5 changes: 3 additions & 2 deletions docs/moleculardynamics/plumed2.rst
Expand Up @@ -67,5 +67,6 @@ by the `FILE` tag. The `PRINT` function tells PLUMED2 to print `d1` and `d2`
every 100 steps to the file specified by the `FILE` tag. Finally, the `ENDPLUMED`
tag indicates to PLUMED2 that input is complete.

More complete information on the functionality of PLUMED2 can be found in its
`user manual <http://plumed.github.io/doc-v2.5/user-doc/html/index.html>`_.
More complete information on the functionality of PLUMED2 can be found
in it's `user manual
<http://plumed.github.io/doc-v2.5/user-doc/html/index.html>`_.
7 changes: 4 additions & 3 deletions docs/moleculardynamics/restarting.rst
Expand Up @@ -90,10 +90,11 @@ Example of restarting
[Input: `recipes/moleculardynamics/thermalise/`]
Stop a running MD calculation by either
Stop a running MD calculation by

* pressing `ctrl + c` or in the DFTB+ working directory create a
stop file ::
* pressing `ctrl + c` on the attached terminal

* in the DFTB+ working directory create a stop file ::

touch stop_driver

Expand Down
6 changes: 6 additions & 0 deletions docs/parallel/benchmarking.rst
Expand Up @@ -246,3 +246,9 @@ Topics to be discussed:
calculations with spin polarisation and/or k-points.

* Effects of latency on code performance.


GPU computing
^^^^^^^^^^^^^

Description coming soon!
15 changes: 13 additions & 2 deletions docs/parallel/introduction.rst
Expand Up @@ -26,8 +26,8 @@ memory per computing core).
Types of parallel hardware
--------------------------

The two types of parallel computer that DFTB+ currently can make use of are
either
The types of parallel computer that DFTB+ currently can make use of
are

#. **shared memory** machines, where all processes have access to the same
data. These are typically small to medium size systems. Most modern CPUs have
Expand All @@ -37,6 +37,11 @@ either
typically larger scale systems with higher numbers of processors and a
dedicated high speed network between them.

#. **GPU accelerated** where either shared memory or distributed
systems have access to graphical processor units, which provide
extremely parallel calculations but currently only for ground-state
calculations.

The different system types require distinct program models to make use of the
hardware (however code designed for a distributed memory system can often be
also used for shared memory architectures).
Expand Down Expand Up @@ -87,3 +92,9 @@ Hybrid parallelism
DFTB+ can be compiled with both MPI and openMP parallelism combined. However
using this can require system specific settings for thread affinity to provide
efficiency and this is beyond the scope of this tutorial.

GPU accelerated
---------------

Either the MAGMA (single node) or ELPA (distributed memory) solvers
and libraries are required.

0 comments on commit 2be9397

Please sign in to comment.