Accelerated Clustered TOPOlogical (ACTOPO) Data Structure

This repository contains the implementation of the ACTOPO data structure integrated into Topology Toolkit framework.

Organization

The repo contains the TTK framework with the necessary plugins that we used in the experiments. It has two branches: sequential-algos, which allows the topological data structures to run in parallel while the plugins run in sequential, and parallel-algos, which has both the data structure and plugins run in parallel.

The implementation of ACTOPO can be found in core/base/actopo. The implementation of each evaluation plugins can also be found in the core/base folder with the following folder names.

• testTopoRelations
• scalarFieldCriticalPoints
• discreteGradient
• morseSmaleComplex3D

How to Install

To install ACTOPO, you can simply follow the official guide to install TTK on your system. Here we provide a simplified walkthrough without the ParaView support on the Linux system (i.e., plugins can only run through the command line).

1. Clone the repository

Simply run the following command to clone this repo to your local machine.

git clone https://github.com/DaVisLab/ACTOPO.git

2. Install the dependencies

Several dependencies need to be installed in order to compile TTK from source. Please enter the following commands in a termainal to install them:

sudo apt-get install cmake-qt-gui libboost-system-dev libpython3.8-dev libxt-dev libxcursor-dev libopengl-dev
sudo apt-get install qt5-default qttools5-dev libqt5x11extras5-dev libqt5svg5-dev qtxmlpatterns5-dev-tools

3. Configure, build and install

Configuration

To enter the configuration menu of TTK's build, enter the following commands:

cd ACTOPO/
mkdir build-sequential
cd build-sequential
cmake-gui ../

The configuration window opens. Click on the Configure button to proceed. Since we do not need to activate ParaView support, please set the TTK_BUILD_PARAVIEW_PLUGINS to OFF, and also make sure TTK_BUILD_STANDALONE_APPS is ON.

Build

Now you can start the compilation process by entering the following command, where N is the number of available cores on your system (this probably takes a LONG time):

make -jN

Installation

Once the build is finished, enter the following command to install TTK on your system:

sudo make install

To check if you successfully installed the TTK, enter the following commands:

ls /usr/local/bin/*Cmd
ls /usr/local/bin/*Gui

Switch branches

To switch between sequential and parallel algorithms, we suggest that you create a separate build folder, e.g., build-parallel, and save the bulid configurations of parallel-algos branch. Every time you want to install a specific version on the system, just go to the corrseponding build folder and run sudo make install command.

How to Use

ACTOPO is activated in the same way as TTK CompactTriangulation, i.e., when the input dataset contains the vertex-based clustering information as a scalar field named _index (defined in the file core/base/common/Datatypes.h). If you already have such datasets, you can skip the clustering step.

Cluster the dataset

TTK provides a preconditioning plugin (ttkCompactTriangulationPreconditioning) that can be used to divide the input mesh based on the point region octree. To use it, enter the following command in the terminal (replace input.vtu and output):

ttkCompactTriangulationPreconditioningCmd -i <input.vtu> -o <output>

To verify if the generated dataset contains the _index field, you can run any standalone plugin with -l option. For example, the followig command checks all data arrays in output.vtu:

ttkTestTopoRelationsCmd -i <output.vtu> -l

Run with ACTOPO

After you have the dataset with clustering information, you can run TTK standalone apps with ACTOPO. Some command-line options are dedicated to ACTOPO data structure:

• -t <int> specifies the number of consumer threads (has no effects in sequential-algo build as algorithms run sequentially).
• -p <int> specifies the number of producer threads (including one leader producer and (num-1) worker producers).
• -b <float> specifies the buffer size as a ratio of total number of blocks in the dataset.

The following command runs the TestTopoRelations plugin with 6 producer threads, and the buffer size is set to 20% of the total number of blocks.

ttkTestTopoRelationsCmd -i <input.vtu> -a "field" -t 1 -p 6 -b 0.2

License

ACTOPO follows the BSD license used by TTK, see LICENSE for more details.

Acknowledgements

[To be updated]