News

Online VASP workshop: Advanced optics and dielectric properties

Date: March 13 - 15, 2024 Venue: Zoom event

Meet the VASP developers and learn how to run your own VASP calculations! The agenda includes lectures and hands-on sessions with the option of attending morning or afternoon sessions to serve all time zones. We will cover various methods and levels of approximation to obtain dielectric properties from first principles. Afterward you will be able to draw the connection to experimental results, decide when density-functional theory (DFT) suffices, and when to opt for methods within many-body perturbation theory, like the GW approximation or solving the Bethe-Salpeter equation within time-dependent DFT.

Go to the event page!

Job announcement: Two open positions.

Company: VASP Software GmbH
Location: Berggasse 21/14, 1090 Vienna

We are looking for a VASP Support Scientist that creates tutorials, contributes to the VASP Wiki, and resolves user inquiries, as well as a VASP HPC Specialist that enhances the scalability of algorithms, implements parallel computing strategies and identifies bottlenecks.

1. Position: VASP Support Scientist

Are you passionate about computational materials science and interested in exploring the vast spectrum of VASP features? Do you have a knack for clear communication? If so, we have an opportunity for you!

Job Description:

As a VASP Support Scientist at VASP Software GmbH, you will play a pivotal role in educating and empowering the scientific community by creating informative and user-friendly tutorials in Jupyter notebooks and enhancing the VASP Wiki with comprehensive documentation. Your work will contribute to making VASP accessible and user-friendly for researchers and scientists worldwide.

Key Responsibilities:

  • Create detailed tutorials in Jupyter notebooks covering a wide range of VASP features.
  • Contribute to and maintain the VASP Wiki with well-organized, comprehensive documentation.
  • Collaborate with the VASP community to gather feedback and improve documentation.
  • Assist in troubleshooting and resolving user inquiries related to VASP.
  • Stay up-to-date with the latest developments and features in VASP.

Qualifications:

  • A PhD and possibly PostDoc experience in physics, chemistry, material science, biology or related fields.
  • A strong interest in the entire spectrum of VASP features.
  • Experience in running VASP calculations for a scientific project or contributing to a first-principles code with similar algorithms and features as VASP.
  • Background in one or more of the following fields would complement the team best: Density-functional-perturbation theory, disorder, impurities, organic chemistry, van der Waals systems, catalysis or superconductivity.

Requirements:

  • Proficiency in using VASP and a strong understanding of first-principles calculations.
  • Strong communication skills to convey complex topics in a simple and understandable manner.
  • Self-motivated and capable of working independently as well as collaboratively.
  • Strong problem-solving skills and attention to detail.
  • A commitment to producing high-quality documentation.

2. Position: VASP HPC Specialist

Are you an expert in High-Performance Computing (HPC) and interested in pushing the boundaries of computational materials science? We have a unique opportunity for an HPC Specialist to join our team!

Job Description:

As an HPC Specialist at VASP Software GmbH, you will be at the forefront of driving computational materials science to new heights. You will play a crucial role in optimizing and accelerating VASP simulations on HPC clusters and supercomputers, enabling groundbreaking scientific research. Your expertise will be instrumental in harnessing the full potential of VASP for solving complex materials-science challenges.

Key Responsibilities:

  • Assess and enhance the scalability of algorithms to ensure VASP can efficiently utilize a wide range of computational resources, from local clusters to national and international supercomputers.
  • Implement and maintain parallel computing strategies, taking advantage of multi-core processors, accelerators (e.g., GPUs), and distributed computing environments.
  • Conduct performance profiling and benchmarking to identify bottlenecks in VASP simulations, providing recommendations for improvement.
  • Offer technical support to the VASP user community, addressing HPC-related queries and troubleshooting issues to improve the user experience.
  • Contribute to the documentation of best practices for running VASP on HPC systems, sharing your knowledge and expertise with users.

Qualifications:

  • A PhD in a relevant field, such as computer science, computational materials science, physics or a related discipline.
  • Extensive experience in optimizing and running scientific applications, particularly in the realm of computational materials science.
  • Proficiency in parallel computing, distributed computing, and familiarity with HPC architectures and clusters.
  • Strong programming skills, with proficiency in languages such as Fortran, C++, Python, and parallel programming models such as MPI, OpenMP, and/or OpenACC.
  • Experience with profiling and performance analysis tools to identify and address bottlenecks.
  • Knowledge of supercomputing environments and experience with batch scheduling systems (e.g., Slurm, Torque, PBS).
  • Excellent problem-solving skills and the ability to work collaboratively with interdisciplinary teams.
  • Strong communication skills to convey complex HPC concepts to a non-technical audience and provide technical support to users.
  • A commitment to staying up-to-date with the latest HPC technologies and trends to continuously improve VASP’s performance on modern computing systems.

What we offer:

At VASP Software GmbH, we are committed to providing our team members with a rewarding and stable career. As part of the VASP team, you’ll enjoy a competitive income and the vibrant atmosphere of Vienna, Austria. While we offer flexibility with home office arrangements, we place a premium on a healthy office culture and do not endorse pure remote work. Unlike the uncertainty of short-term contracts in academia, our positions offer long-term job security. You will have the opportunity to pursue scientific projects and contribute to the growth of computational materials science while being part of a dynamic and dedicated team.

How to Apply:

If you are enthusiastic about VASP and possess the qualifications and skills we are looking for, we’d love to hear from you! Please write a motivational letter expressing what makes you the ideal candidate. Send this letter along with a curriculum vitae and a list of all publications to marie-therese.huebsch@vasp.at (subject: Job application) and in cc to christian.hirsch@vasp.at. Preselection of candidates will begin November 20, although we might accept applications later, if the positions are not yet filled.

Deadline for Applications: December 12, 2023

VASP Software GmbH is an equal opportunity employer. We encourage applications from candidates of all backgrounds. Join our team and help shape the future of computational materials science by making VASP more accessible and user-friendly. Your expertise and passion can make a real difference!

VASP support extension for Visual Studio Code

For users of Visual Studio Code, we highly recommend the VASP Support extension. It provides up-to-date information for INCAR tags, including descriptions, default values and links to the VASP Wiki. Additionally, a heap of useful information is provided when editing POSCAR files.

Try it yourself right now by installing the extension either from the Visual Studio Marketplace or directly inside Visual Studio Code using the extensions view.

NEW RELEASE: VASP.6.4.2

A new release of VASP is available for download now!

This release differs from VASP.6.4.1 by the following changes/additions (amongst others):

Fixes the following known issues:

  • Bugs in interface to wannier90.
  • Incorrect MLFF fast-mode predictions for some triclinic geometries.
  • Fast-mode predictions will crash together with finite difference (IBRION=5,6).

and additionally:

  • Added a proper exit if MLFF is encounters multiple atom types with the same label.
  • Blocking of scaLAPACK redistribution in MLFF (to reduce memory demands) was not correctly implemented.
  • Assorted small fixes and improvements.

Go to the VASP Portal to get VASP.6.4.2. This requires a VASP.6 license issued by the VASP Software GmbH.
Get a license!

Online VASP workshop: Moving ions for the advanced user

November 13 - 17, 2023

Schedule: 12:00 – 17:30 CET (lectures and hands-on sessions)
Venue: Online (Zoom)

In 5 days of lectures and hands-on sessions, this workshop covers ab-initio molecular dynamics, machine learning force fields, phonons, and high-performance computing. During the hands-on sessions, you will learn to perform your own ab-initio simulations using the Vienna Ab-initio Simulation Package (VASP). Additionally, you have the extraordinary opportunity to meet the team developing VASP to learn from them how to simulate the properties of materials.

Registration fee: 250€
Capacity: 100 (At least 15% will be free of charge.)
Deadline: October 13, 2023

Register here!

Note: The lectures are live, and a recording will be available for participants afterward. For hands-on sessions, active participation is required!

NEW RELEASE: VASP.6.4.1

A new release of VASP is available for download now!

This release differs from VASP.6.4.0 by the following changes/additions (amongst others):

  • Fixes:
    • Corrupt ML_FF files on some systems due to parallel writing and opening of these files.
    • ML_MODE=RUN broken for fast version if production run is carried out for subtypes of types in ML_FF file.
    • Bug in sparsification routine for default values of ML_MB.
    • Memory estimates too low in ML_LOGFILE for ML_IALGO_LINREG=4 (SVD).
    • Numerical issues for small Bayesian error estimates (square root of negative numbers is now caught).
  • Improvements:
    • ML_MB can be set to smaller numbers than the number of local reference configurations on the ML_AB file. This way the accuracy can be controlled via ML_MB.
    • Improved default values for many machine learning related variables.
    • ML_ICRITERIA=3 new default for ML_MODE=SELECT.
    • Significantly less memory needed for ML_IALGO_LINREG=4 (SVD).

Go to the VASP Portal to get VASP.6.4.1. This requires a VASP.6 license issued by the VASP Software GmbH.
Get a license!

NEW RELEASE: VASP.6.4.0

A new release of VASP is available for download now!

This release differs from VASP.6.3.2 by the following changes/additions (amongst others):

  • DFT and hybrid functionals:
    • IVDW_NL: A new tag to choose the version of the nonlocal vdW-DF functional.
    • Spin-polarized formulation for nonlocal vdW functionals.
    • Hybrid functionals that combine MGGA and Hartree-Fock.
    • Laplacian-dependent MGGA.
    • Local mBJ potential.
  • ACFDT and GW:
    • Added EVGW0R algorithm (NATURALO=4).
    • all-in-one GW mode improved (LALL_IN_ONE).
    • OpenACC port to GPU of ALGO = ACFDTR.
  • Machine learned force fields:
    • New fast-prediction mode (typically 20-100 times faster MD trajectories are possible).
    • Improved neighbor list algorithm (cell lists + Verlet lists, supports fast-prediction mode).
    • New “super”-tag ML_MODE for easy training and application of ML force fields.
    • New human-readable header for ML_FF file containing most important parameters, corresponding time stamp in ML_LOGFILE.
    • Regularized SVD for refitting with optimized default parameters (ML_IALGO_LINREG = 4).
    • Additional ML tags presented in ML_LOGFILE.
  • Miscellaneous:
    • Added Nose-hoover chain thermostat
    • Added option to set Fermi energy in the middle of the band gap (EFERMI = MIDGAP).
    • Created new command line argument “–dry-run” to check the setup.
    • Allow faster wavefunction initialization with parallel random numbers (RANDOM_GENERATOR = PCG_32).
    • Write additional output to vaspout.h5 for extraction with py4vasp (pair correlation function, dielectric function, velocities).
  • Various improvements and bugfixes.

Go to the VASP Portal to get VASP.6.4.0. This requires a VASP.6 license issued by the VASP Software GmbH.
Get a license!

Offer to VASP.5 users: Upgrade to VASP.6

We are pleased that you are using VASP.5 in your research. Perhaps some recent developments in VASP.6 will help you to run larger simulations or investigate new material properties.

Highlights:

  • The new on-the-fly algorithms to machine-learn force fields accelerate molecular-dynamics simulations by many orders of magnitude. This allows you to investigate larger systems for significantly longer simulation times.

  • Adiabatically compressing exchange for hybrid functionals and the cubic scaling RPA and GW code accelerate the computation of the exchange-correlation energy in these advanced techniques.

  • Many high-performance computing centers offer GPU accelerators. With the OpenACC port of VASP.6, you can utilize this computational power efficiently.

You will find a more comprehensive list of the new features in our wiki (https://www.vasp.at/wiki/index.php/Category:VASP6).

If you are interested in any of these features, please get in contact with our licensing team and benefit from the upgrade price of about 1500 € (academic)* and about 3000 € (governmental and non-profit)*. Please note that this offer will expire on June 30, 2023. After that, this price will only apply to upgrades from an expired VASP.6 license.

We hope you will consider this offer and look forward to hearing from you.

Kind regards, The VASP team

*This offer is not binding. The actual final price may be up to 20% higher and depends on sales region and applicable taxes. For a final binding offer, please contact licensing@vasp.at.

NEW RELEASE: VASP.6.3.2

A new release of VASP is available for download now!

This release consist of several bugfixes and performance improvements w.r.t. VASP.6.3.1.

Go to the VASP Portal to get VASP.6.3.2. This requires a VASP.6 license issued by the VASP Software GmbH.
Get a license!

Online VASP workshop: Electronic correlation from first principles

November 15 - 18, 2022

Schedule: 4.00 pm – 9.00 pm CET (lectures and hands-on sessions)
Venue: Online (Zoom)

In 4 days of lectures and hands-on sessions, this workshop covers exchange-correlation density functionals, hybrid functionals, optical properties within linear-response theory, and methods based on many-body perturbation theory. The latter includes total-energy calculations within the random-phase approximation, electronic structure calculations in the GW approximation and using the Bethe-Salpeter equations. During the hands-on sessions, you will learn to perform your own ab-initio simulations using the Vienna Ab-initio Simulation Package (VASP). Additionally, you have the extraordinary opportunity to meet the team developing VASP to learn from them how to simulate the properties of materials.

Registration fee: 250€
Capacity: 100 (At least 15% will be free of charge.)
Deadline: September 19, 2022

Register here!

Note: The lectures are live, and a recording will be available for participants afterward. For hands-on sessions, active participation is required!

Research Software Engineers

We are seeking two Research Software Engineers in Vienna, Austria. We offer you

  • A combination of software development and scientific projects
  • To learn cutting-edge technologies for high-performance computing
  • A competitively-paid long-term perspective without grant or tenure track constraints

The VASP Software GmbH develops the de-facto standard for the simulation of condensed matter systems—the Vienna Ab Initio Software Package (VASP). We have a dynamic international team and would like to add two new members that

  • Have a proven track record in Computational Solid-State Physics or possibly Computational Chemistry
  • Enjoy programming and have experience with Fortran or Python
  • Want to understand materials from first principles
  • Like improving their software developers’ capabilities

You would stand out, if you

  • Have prior experience with VASP
  • Contributed to a large software project
  • Designed graphical applications and interfaces
  • Worked with workflow tools (ASE, AiiDA, pymatgen, …)
  • Developed code on modern high-performance architectures
  • Have extensive experience with teaching and workshop planning

Please write in a motivational letter expressing what makes you the ideal candidate. Send this letter along with a curriculum vitae and a list of all publications to martin.schlipf@vasp.at (subject: position Research Software Engineer) and in cc to christian.hirsch@vasp.at.

Preselection of candidates will begin 6th of July.

Online VASP workshop: Moving ions with VASP

September 27 - 30, 2022

Schedule: 4.00 pm – 9.00 pm CEST (lectures and hands-on sessions)
Venue: Online (Zoom)

In 4 days of lectures and hands-on sessions, this workshop covers ab-initio molecular dynamics, machine learning force fields, structure prediction, and phonons. During the hands-on sessions, you will learn to perform your own ab-initio simulations using the Vienna Ab-initio Simulation Package (VASP). Additionally, you have the extraordinary opportunity to meet the team developing VASP to learn from them how to simulate the properties of materials.

Registration fee: 250€
Capacity: 100 (At least 15% will be free of charge.)
Deadline: August 1, 2022

Register here!

Note: The lectures are live, and a recording will be available for participants afterward. For hands-on sessions, active participation is required!

NEW RELEASE: VASP.6.3.1

A new release of VASP is available for download now!

This release consist of several bugfixes and performance improvements w.r.t. VASP.6.3.0.

Go to the VASP Portal to get VASP.6.3.0. This requires a VASP.6 license issued by the VASP Software GmbH.
Get a license!

py4vasp available via PyPI

Over the last two years, we have developed a new post-processing tool for VASP. Built with Python, you can integrate it easily into your scripts. This new package uses the HDF5 support of VASP to give you direct access to the data so that you can avoid parsing the text-based output.

If you use Jupyter Notebooks, you will appreciate the interactive plotting tools. For example, with a few easy lines of code, you see the structure or the density of states VASP produced.

At a glance: Watch the teaser to see what you can do with py4vasp!

Start using py4vasp

py4vasp is used extensively in our new tutorials. Please check them out to see how this package can assist your research.

1st online VASP workshop: Introduction to ab-initio simulation

November 8 - 12, 2021

Schedule: 4.00 pm – 8.00 pm CET (lectures), 12.00 am – 4.00 pm CET (hands-on sessions) Venue: Online (Zoom)

In 8 introductory lectures and 8 hands-on sessions, this workshop covers density-functional theory, the projector-augmented-wave method, selected exchange-correlation functionals, corrections for correlated electron systems on the level of GW, magnetism, relativistic corrections, and molecular dynamics using our novel machine-learning force-fields module. During the hands-on sessions, you will learn to perform your own ab-initio simulations using the Vienna Ab-initio Simulation Package (VASP), and you can have discussions with the developers of VASP.

Registration fee: 280€ Capacity: 64 (At least 15% will be free of charge.) Deadline: October 1, 2021

Note: The lectures are live and will be available on-demand for participants afterward. For hands-on sessions active participation is required!

Register here!

NEW RELEASE: VASP.6.2.1

A new release of VASP, version 6.2.1, is available for download now (for all holders of a valid “VASP.6” license)!

For all VASP users who have already purchased a VASP.6 license from the VASP Software GmbH, VASP.6.2.1 is now available for download.

This release consist of several (minor) bugfixes and performance improvements w.r.t. VASP.6.2.0.

VASP users who have a valid VASP.5 license from the VASP Software GmbH or the University of Vienna are entitled to upgrade to VASP.6. Because of withholding taxes and other surcharges the upgrade fee varies between 1.500,- Euro and about 2.000,- Euro net of VAT for six academic users. It requires the signing of a new license agreement between the VASP Software GmbH and the licensee. The new license agreement will replace the license agreement between the University of Vienna and the licensee.

To request an upgrade or new license send an email to licensing@vasp.at.

NEW RELEASE: VASP.6.2.0

A new release of VASP, version 6.2.0, is available for download now (for all holders of a valid “VASP.6” license)!

For all VASP users who have already purchased a VASP.6 license from the VASP Software GmbH, VASP.6.1.0 is now available for download.

VASP users who have a valid VASP.5 license from the VASP Software GmbH or the University of Vienna are entitled to upgrade to VASP.6. Because of withholding taxes and other surcharges the upgrade fee varies between 1.500,- Euro and about 2.000,- Euro net of VAT for six academic users. It requires the signing of a new license agreement between the VASP Software GmbH and the licensee. The new license agreement will replace the license agreement between the University of Vienna and the licensee.

To request an upgrade or new license send an email to licensing@vasp.at.

Bugfix: testsuite in VASP.6.1.0 to 6.1.2 on MacOS and BSD

The vasp.6.1.0, vasp.6.1.1 and vasp.6.1.2 releases have a potentially serious issue that can lead to loss of user data unrelated to vasp in the home directory! This issue has yet been only observed on MacOS and BSD systems, but might affect other OS as well. Avoid to use

 make test
 make test_all
 make cleantest

or the use of any script called

 cleanall

on MacOS or BSD systems (and potentially other OS) until we release a fix.

This script is used to clean the files produced by VASP after running the test suite. The readlink and dirname commands take different arguments on Linux and BSD systems including macOS. On BSD systems this results in the $dir variable becoming empty, cd to the home folder and removing ALL files recursively that do not match the pre-defined patterns.