The Toolkit for Interactive Computation (TINC) provides a set of C++ and python classes to assist in the interactive exploration of large datasets by managing parameter spaces, interactive computation and caching of data.
TINC allows exposing C++ application controls to the network. This simplifies the development of distributed applications as well as creating applications that can be controlled through python without having the application itself depend on python. A great use case is by interacting with the C++ application through a jupyter notebook.
TINC can also be used standalone in python to assist exploration of complex datasets that are spread out through the filesystem through interactive computation prototyping.
Here are some videos demonstrating the functionality and API of TINC
TINC: ParameterSpaces in Python TINC: Client in Python interacting with C++
We have set up to repositories to assist learning and using TINC. The tinc-playground is a quick way to build and explore the examples included with TINC. It provides a convenience run script to easily build single file applications, or build all the examples. The [tinc-template] (https://github.com/AlloSphere-Research-Group/tinc-template) repo shows how to set up a more complex project using TINC as a submodule through CMake. This template is useful to build larger applications made from multiple files or with multiple target that might require additional dependencies.
The TINC API documentation is online.
Join the user and development mailing lists at:
Please file feature and bug reports through the github issue tracker.
TINC depends on NetCDF4. You can use a binary installation and set the NETCDF4_INSTALL_ROOT cmake variable using ccmake or on the command line with:
-DNETCDF4_INSTALL_ROOT="C:/Program Files/netCDF 4.7.4/"
If you use a binary installation, make sure you download a netCDF version that includes hdf5 AND DAP support.
Or you can use the *./build_deps.sh script on *nix platforms. You will need ninja and curl installed for these scripts to work.
TINC is funded by NSF grant # OAC 2004693: "Elements: Cyber-infrastructure for Interactive Computation and Display of Materials Datasets."