Research: Higher Education and Supercomputing

COMPUTATIONAL CHEMISTRY AND BIOLOGY

Bioinformatics >

Arioc

Description:

High-throughput read alignment with GPU- accelerated exploration of the seed-and- extend search space

Features:

Single-end alignment, paired-end alignment Output in SAM or database-ready binary formats Multiple GPU implementation

Multi GPU support:

Yes

BarraCUDA

Description:

Sequence mapping software

Features:

Alignment of short sequencing reads, alignment of indels with gap openings and extensions.

Multi GPU support:

Yes

BEAGLE-lib

Description:

BEAGLE is a high-performance library that can perform the core calculations at the heart of most Bayesian and Maximum Likelihood phylogenetics packages. It can make use of highly-parallel processors such as those in graphics cards (GPUs) found in many PCs.

Features:

Evaluation of likelihood for sequence evolution on trees and Arbitrary models (e.g. nucleotide, amino acid, codon) Speed-ups (over CPU only version): nucleotide model = up to 25x, codon model = up to 50x.

Multi GPU support:

Yes

Campaign

Description:

An open-source library of GPU-accelerated data clustering algorithms and tools.

Features:

K-means (and Kps-means, a K-means variant for GPUs with parallel sorting for improved performance), K-medoids, K-centers (a K-medoids variant in which medoids are placed only once according to a heuristic), Hierarchical clustering and Self-organizing map.

Multi GPU support:

Single only

CUDASW++

Description:

Open source software for Smith-Waterman protein database searches on GPUs.

Features:

Parallel search of Smith-Waterman database.

Multi GPU support:

Yes

CUSHAW

Description:

Parallelized short read aligner

Features:

Parallel, accurate long read aligner for large genomes

Multi GPU support:

Yes

G-BLASTN

Description:

GPU-accelerated nucleotide alignment tool based on the widely used NCBI-BLAST.

Features:

Blastn and megablast modes of NCBI-BLAST

Multi GPU support:

Single only

GPU-Blast

Description:

Local search with fast k-tuple heuristic

Features:

Protein alignment according to BLASTP

Multi GPU support:

Single only

mCUDA-MEME based on MEME .

Description:

Ultrafast scalable motif discovery algorithm

Features:

Scalable motif discovery algorithm based on MEME.

Multi GPU support:

Yes

MUMmer GPU program

Description:

High-throughput local sequence alignment

Features:

Aligns multiple query sequences against reference sequence in parallel.

Multi GPU support:

TBD

NVBIO

Description:

NVBIO is an open source C++ library of reusable components designed to accelerate bioinformatics applications using CUDA.

Features:

Data structures, algorithms, and utility routines useful for building complex computational genomics applications on CPU-GPU systems.

Multi GPU support:

Yes

NVBowtie

Description:

A largely complete implementation of the Bowtie2 aligner on top of NVBIO.

Features:

Good coverage of Bowtie2 features and comparable quality results.

Multi GPU support:

Yes

PEANUT

Description:

Read mapper for DNA or RNA sequence reads to a known reference genome.

Features:

Achieves supreme sensitivity and speed compared to current state of the art read mappers like BWA MEM, Bowtie2 and RazerS3. PEANUT reports both only the best hits or all hits.

Multi GPU support:

Single only

REACTA

Description:

A modified version of GCTA with improved computational performance, support for Graphics Processing Units (GPUs), and additional features. The purpose of REACTA is to quantify the contribution of genetic variation to phenotypic variation for complex traits.

Features:

GRM creation, REML analysis, Regional Heritability (including multi-GPU).

Multi GPU support:

Yes

SeqNFind

Description:

SeqNFind® is a powerful tool suite that addresses the need for complete and accurate alignments of many small sequences against entire genomes utilizing a unique hardware/software cluster system for facilitating bioinformatics research in Next Generation sequencing and genomic comparisons.

Features:

Hardware and software for reference assembly, blast, SW, HMM, de novo assembly.

Multi GPU support:

Yes

SOAP3

Description:

GPU-based software for aligning short reads with a reference sequence. It can find all alignments with k mismatches, where k is chosen from 0 to 3.

Features:

Short read alignment tool that is not heuristic based; reports all answers.

Multi GPU support:

Yes

SOAP3-dp

Description:

SOAP3-dp: Ultra-fast GPU-based tool for short read alignment via index-assisted dynamic programming.

Features:

Borrows-Wheeler Transformation, Dynamic Programming.

Multi GPU support:

Yes

UGene

Description:

Open source Smith-Waterman for SSE/ CUDA, Suffix array based repeats finder and dotplot.

Features:

Fast short read alignment.

Multi GPU support:

Yes

WideLM

Description:

Fits numerous linear models to a fixed design and response.

Features:

Parallel linear regression on multiple similarly-shaped models.

Multi GPU support:

Yes

Molecular Dynamics >

ACEMD

Description:

GPU simulation of molecular mechanics force fields, implicit and explicit solvent

Features:

Written for use only on GPUs.

Multi GPU support:

Yes

AMBER

Description:

Suite of programs to simulate molecular dynamics on biomolecule.

Features:

PMEMD Explicit Solvent and GB Implicit Solvent

Multi GPU support:

Yes

CHARMM

Description:

MD package to simulate molecular dynamics on biomolecule.

Features:

Implicit (5x), Explicit (2x) Solvent via OpenMM, now ported natively to GPUs.

Multi GPU support:

Yes

DESMOND

Description:

High-speed molecular dynamics simulations of biological systems.

Features:

The code uses novel parallel algorithms and numerical techniques to achieve high performance and accuracy.

Multi GPU support:

Yes

ESPResSo

Description:

Highly versatile software package for performing and analyzing scientific Molecular Dynamics many-particle simulations of coarse-grained atomistic or bead-spring models as they are used in soft-matter research in physics, chemistry and molecular biology.

Features:

Hydrodynamic / Electrokinetic forces P3M electrostatics.

Multi GPU support:

Yes

Description:

A distributed computing project that studies protein folding, misfolding, aggregation, and related diseases.

Features:

Powerful distributed computing molecular dynamics system; implicit solvent and folding.

Multi GPU support:

Yes

GPUgrid.net

Description:

A distributed computing project that uses GPUs for molecular simulations.

Features:

High-performance all-atom biomolecular simulations; explicit solvent and binding.

Multi GPU support:

Yes

GROMACS

Description:

Simulation of biochemical molecules with complicated bond interactions.

Features:

Implicit (5x), Explicit (2x) Solvent

Multi GPU support:

Yes

HALMD

Description:

Large-scale simulations of simple and complex liquids.

Features:

Simple fluids and binary mixtures (pair potentials, high-precision NVE and NVT, dynamic correlations).

Multi GPU support:

Single only

HOOMD-Blue

Description:

Particle dynamics package written grounds up for GPUs.

Features:

Written for use only on GPUs

Multi GPU support:

Yes

LAMMPS

Description:

Classical molecular dynamics package

Features:

Lennard-Jones, Gay-Berne, Tersoff, and dozens more potentials

Multi GPU support:

Yes

* MELD

Description:

OpenMM plugin written for GPUs

Features:

Integrative approach to combine physics and information Orders of magnitude faster protein folding than brute force MD

Multi GPU support:

Yes

NAMD

Description:

Designed for high-performance simulation of large molecular systems.

Features:

Full electrostatics with PME and most simulation features; 100M atom capable.

Multi GPU support:

Yes

OpenMM

Description:

Library and application for molecular dynamics for HPC with GPUs.

Features:

Implicit and explicit solvent, custom forces

Multi GPU support:

Yes

PolyFTS

Description:

Classical molecular simulation code for studying polymer self-assembly and thermodynamics.

Features:

Uses auxiliary fields as the fundamental simulation degrees of freedom, Uses cuFFT extensively (~ 80%), CUDA code is ~20%, Multi CPU or single GPU per job, 1x = Ivy Bridge E5-2690 CPU all 10 cores, 3-8X on K40 or K80 (utilizing 1/2 of the K80).

Multi GPU support:

Single only

SOP-GPU

Description:

SOP-GPU package, where SOP stands for the Self Organized Polymer Model fully implemented on a GPU, is a scientific software package designed to perform Langevin Dynamics Simulations of the mechanical or thermal unfolding, and mechanical indentation of large biomolecular systems in the experimental subsecond (millisecond-to-second) timescale.

Features:

Langevin dynamics simulations using the coarse-grained Self Organized Polymer (SOP) model, Multiple simulation trajectories can be performed simultaneously on a single GPU, Calpha and Calpha-Cbeta models are supported, Simulations of protein forced unfolding, Novel simulations of nanoindentation in silico, Support for hydrodynamic interactions, Up to ~100 ms of simulation time per day, Systems of up to 1,000,000 amino-acids (on GPUs with 6GB or great memory).

Multi GPU support:

Single only

Quantum Chemistry >

Abinit

Description:

Allows to find total energy, charge density and electronic structure of systems made of electrons and nuclei within DFT.

Features:

Local Hamiltonian, non-local Hamiltonian, LOBPCG algorithm, diagonalization/ orthogonalization.

Multi GPU support:

Yes

ACES III

Description:

Takes best features of parallel implementations of quantum chemistry methods for electronic structure.

Features:

Integrating scheduling GPU into SIAL programming language and SIP runtime environment.

Multi GPU support:

Yes

ADF

Description:

Density Functional Theory (DFT) software package that enables first-principles electronic structure calculations.

Features:

Geometry optimizations and frequency calculations with GGA functionals.

Multi GPU support:

Yes

BigDFT

Description:

Implements density functional theory by solving the Kohn-Sham equations describing the electrons in a material.

Features:

DFT; Daubechies wavelets, part of Abinit

Multi GPU support:

Yes

CASTEP [In development]

Description:

CASTEP is a leading code for calculating the properties of materials from first principles. Using density functional theory, it can simulate a wide range of properties of materials proprieties including energetics, structure at the atomic level, vibrational properties, electronic response properties etc.

Features:

TBD

Multi GPU support:

Yes

CP2K

Description:

Program to perform atomistic and molecular simulations of solid state, liquid, molecular and biological systems.

Features:

DBCSR (space matrix multiply library)

Multi GPU support:

Yes

GAMESS-UK

Description:

The general purpose ab initio molecular electronic structure program for performing SCF-, DFT- and MCSCF-gradient calculations.

Features:

(ss|ss) type integrals within calculations using Hartree-Fock ab initio methods and density functional theory. Supports organics and inorganics.

Multi GPU support:

Yes

GAMESS-US

Description:

Computational chemistry suite used to simulate atomic and molecular electronic structure.

Features:

Libqc with Rys Quadrature Algorithm, Hartree-Fock, MP2 and CCSD.

Multi GPU support:

Yes

Gaussian [In development]

Description:

Predicts energies, molecular structures, and vibrational frequencies of molecular systems.

Features:

Joint NVIDIA, PGI and Gaussian collaboration.

Multi GPU support:

Yes

GPAW

Description:

Real-space grid DFT code written in C and Python

Features:

Electrostatic poisson equation, orthonormalizing of vectors, residual minimization method (rmm-diis).

Multi GPU support:

Yes

gWL-LSMS

Description:

Materials code for investigating the effects of temperature on magnetism.

Features:

Generalized Wang-Landau method

Multi GPU support:

Yes

LATTE

Description:

Density matrix computations

Features:

CU_BLAS, SP2 Algorithm

Multi GPU support:

Yes

* LSDalton

Description:

Linear-scaling HF and DFT code suitable for large molecular systems, now also with some CCSD capabilities

Features:

(T) correction to the CCSD energy. • RI-MP2 energy/gradient (in development). • CCSD energy (in development). • GPU-based ERI generator (in development).

Multi GPU support:

Yes

MOLCAS

Description:

Methods for calculating general electronic structures in molecular systems in both ground and excited states.

Features:

CU_BLAS

Multi GPU support:

Single only Additional GPU support coming in Version 8

MOPAC2012

Description:

Semiempirical Quantum Chemistry

Features:

Pseudodiagonalization, full diagonalization, and density matrix assembling via Magma libraries.

Multi GPU support:

Single only

NWChem

Description:

Calculations

Features:

Triples part of Reg-CCSD(T), CCSD and EOMCCSD task schedulers.

Multi GPU support:

Yes

Octopus

Description:

Used for ab initio virtual experimentation and quantum chemistry calculations.

Features:

Full GPU support for ground-state, real-time calculations; Kohn-Sham Hamiltonian, orthogonalization, subspace diagonalization, poisson solver, time propagation.

Multi GPU support:

TBD

ONETEP [In development]

Description:

ONETEP (Order-N Electronic Total Energy Package) is a linear-scaling code for quantum-mechanical calculations based on density-functional theory.

Features:

TBD

Multi GPU support:

Yes

PEtot

Description:

First principles materials code that computes the behavior of the electron structures of materials.

Features:

Density functional theory (DFT) plane wave pseudopotential calculations.

Multi GPU support:

Yes

PWMat

Description:

The fastest plane wave pseudopotential code for density functional theory simulations based on GPU.

Features:

It can perform extremely fast plane wave DFT calculations based on GPU machines and single precision and double precision mixed algorithm. It deploys the state-of- the-art electronic structure calculation methods with many new features and algorithm innovations. It performs ab initio material science simulations, designed for both theoretical and experimental groups.

Multi GPU support:

Yes

Q-CHEM

Description:

Computational chemistry package designed for HPC clusters.

Features:

Various features including RI-MP2

Multi GPU support:

Single Only

QMCPACK

Description:

Solves the many-body Schrodinger equation for electronic structures using a quantum Monte Carlo method.

Features:

Main features

Multi GPU support:

Yes

Quantum Espresso/ PWscf

Description:

An integrated suite of computer codes for electronic structure calculations and materials modeling at the nanoscale.

Features:

PWscf package: linear algebra (matix multiply), explicit computational kernels, 3D FFTs.

Multi GPU support:

Yes

QUICK

Description:

QUICK is a GPU-enabled ab intio quantum chemistry software package.

Features:

Running Hartree-Fock and DFT energy on GPU, Supports s, p, d, f orbitals on energy calculation, HF gradient with s,p,d orbital support, GPU-based ERI generator.

Multi GPU support:

Yes

TeraChem

Description:

Quantum chemistry software designed to run on NVIDIA GPU.

Features:

Full GPU-based solution; Performance compared to GAMESS CPU version.

Multi GPU support:

Yes

VASP

Description:

Complex package for performing ab-initio quantum-mechanical molecular dynamics (MD) simulations using pseudopotentials or the projector-augmented wave method and a plane wave basis set.

Features:

Blocked Davidson (ALGO = NORMAL & FAST), RMM-DIIS (ALGO = VERYFAST & FAST), K-Points and optimization for critical step in exact exchange calculations.

Multi GPU support:

Yes

Visualization and Docking >

Amira®

Description:

A multifaceted software platform for visualizing, manipulating, and understanding Life Science and bio-medical data.

Features:

3D visualization of volumetric data and surfaces

Multi GPU support:

Single only

BINDSURF

Description:

A virtual screening methodology that uses GPUs to determine protein binding sites.

Features:

Allows fast processing of large ligand databases

Multi GPU support:

Single only

BUDE

Description:

Molecular docking program

Features:

Empirical Free Energy Force field

Multi GPU support:

Single only

FastROCS

Description:

Molecule shape comparison application

Features:

Real-time shape similarity searching/ comparison

Multi GPU support:

Yes

* Interactive Molecule Visualizer

Description:

Experimental interactive molecule visualizer based on a ray-tracing engine.

Features:

Targeting high quality images and ease of interaction, IMV uses the latest GPUcomputing acceleration techniques, combined with natural user interfaces such as Kinect and Wiimotes.

Multi GPU support:

Single only

Molegro Virtual Docker 6

Description:

Method for performing high accuracy flexible molecular docking.

Features:

Energy grid computation, pose evaluation and guided differential evolution.

Multi GPU support:

Single only

PaPaRa 2.0

Description:

A Vectorized Algorithm for Probabilistic Phylogeny-Aware Alignment Extension.

Features:

Up to 15-fold run time improvements by deploying SIMD vector intrinsics to accelerate the alignment kernel.

Multi GPU support:

Single only

PIPER Protein Docking

Description:

Protein-protein docking program

Features:

Molecule docking

Multi GPU support:

TBD

PyMol

Description:

User-sponsored molecular visualization system on an open-source foundation

Features:

Lines: 460% increase Cartoons: 1246% increase Surface: 1746% increase Spheres: 753% increase Ribbon: 426% increase

Multi GPU support:

Single only

VEGA ZZ

Description:

Molecular Modeling Toolkit

Features:

Virtual logP, molecular surface values

Multi GPU support:

Single only

VMD

Description:

Visualization and analyzing large bio- molecular systems in 3-D graphics

Features:

High quality rendering, large structures (100M atoms), analysis and visualization tasks, multiple GPU support for display of molecular orbitals

Multi GPU support:

Yes

NUMERICAL ANALYTICS

>

Accelereyes- ArrayFire

Description:

Comprehensive GPU function library

Features:

Hundreds of functions for math, signal/ image processing, statistics, and more.

Multi GPU support:

Yes

HiPLAR

Description:

3High Performance Linear Algebra in R

Features:

Supports GPU and multi-core platforms, compatible with legacy R code, no new data types or operators, auto-tuning, support for R Matrix package.

Multi GPU support:

Yes (for algebra functions via Magma 1.5 or later)

Mathematica Wolfram

Description:

A symbolic technical computing language and development environment.

Features:

Development environment for CUDA and OpenCL. GPU acceleration for Wolfram Finance Platform.

Multi GPU support:

Yes

Mathworks - MATLAB

Description:

GPU acceleration for MATLAB (high-level technical computing language).

Features:

Support for 200+ of most used MATLAB functions (incl. Signal Processing, Image Processing, Communications Systems, etc).

Multi GPU support:

Yes

NMath Premium

Description:

GPU-accelerated math and statistics for .NET, automatically detects the presence of a CUDA-enabled GPU at runtime and seamlessly redirects appropriate computations to it.

Features:

Automatically offloads computations to the GPU.

Multi GPU support:

Single only

PHYSICS

AWP

Description:

The Anelastic Wave Propagation, AWP- ODC, independently simulates the dynamic rupture and wave propagation that occurs during an earthquake. Dynamic rupture produces friction, traction, slip, and slip rate information on the fault. The moment function is constructed from this fault data and used to initialize wave propagation.

Features:

3D Finite Difference Computation

Multi GPU support:

Single only

BQCD

Description:

Lattice quantum chromodynamics application, used for nuclear ad high energy physics calculations.

Features:

Wilson-clover fermion linear solver

Multi GPU support:

Yes

* CASTRO

Description:

A multicomponent compressible hydrodynamic code for astrophysical flows including self-gravity, nuclear reactions and radiation. CASTRO uses an Eulerian grid and incorporates adaptive mesh refinement (AMR). The approach uses a nested hierarchy of logically-rectangular grids with simultaneous refinement in both space and time.

Features:

Gravitational Field Solver

Multi GPU support:

Yes

Changa

Description:

Astrophysics code performs collisionless N-body simulations. It can perform cosmological simulations with periodic boundary conditions in comoving coordinates or simulations of isolated stellar systems.

Features:

Gravitational Model has been accelerated using CUDA

Multi GPU support:

Single only

Chemora

Description:

Chemora is a system for performing simulations of systems described by differential equations running on accelerated computational clusters.

Features:

Chemora embeds the equations’ computational kernels into dynamically compiled loop nests shaped for input size and GPU structure.

Multi GPU support:

Yes

Chroma

Description:

Lattice Quantum Chromodynamics (LQCD)

Features:

Wilson-clover fermions, Krylov solvers, Domain-decomposition

Multi GPU support:

Yes

CPS

Description:

Lattice quantum chromodynamics application, used for nuclear ad high energy physics calculations.

Features:

Wilson, domain-wall and Möbius fermion linear solvers

Multi GPU support:

Yes

ENZO

Description:

3D block-structured AMR code for cosmological structure formation.

Features:

Accelerated magneto hydrodynamics solvers

Multi GPU support:

Yes

GTC

Description:

Simulates microturbulence and transport in magnetically confined fusion plasma.

Features:

Electron push and shift (accounting for >80% of run time)

Multi GPU support:

Yes

GTC-P

Description:

A development code for optimization of plasma physics. Full science and data sets are included, but in a simplified form to allow performance testing and tuning.

Features:

Optimized with CUDA. OpenACC development underway

Multi GPU support:

Yes

GTS

Description:

Simulates microturbulence and the motion of charged particles and interactions in fusion plasma.

Features:

Push and shift for both electron and ion dynamics

Multi GPU support:

Yes

HACC

Description:

Simulates N-Body Astrophysics

Features:

This code has been optimized with CUDA runs in full production mode.

Multi GPU support:

Yes

* MAESTRO

Description:

A low Mach number stellar hydrodynamics code that can be used to simulate long- time, low-speed flows that would be prohibitively expensive to model using traditional compressible code.

Features:

Gravitational Field Solver

Multi GPU support:

Yes

MILC

Description:

Lattice Quantum Chromodynamics (LQCD) codes simulate how elemental particles are formed and bound by the “strong force” to create larger particles like protons and neutrons.

Features:

Staggered fermions, Krylov solvers, Gauge- link fattening.

Multi GPU support:

Yes

OSIRIS

Description:

Simulates Plasma Physics including Laser interaction

Features:

2 dimensions of the particle push have been optimized with CUDA. Additional optimization is being planned with OpenACC.

Multi GPU support:

Yes

PIConGPU

Description:

A relativistic Particle-in-Cell code that describes the dynamics of a plasma by computing the motion of electrons and ions subject to the Maxwell-Vlasov equation.

Features:

Simulation of laser-wakefield acceleration of electrons.

Multi GPU support:

Yes

PPM

Description:

Piecewise parabolic method, a higher- order extension of Godunov’s method which uses spatial interpolation and allows for a steeper representation of discontinuities, particularly contact discontinuities.

Features:

Turbulent, compressible mixing of gases in the context of stars near the ends of their lives and also in inertial confinement fusion.

Multi GPU support:

Single only

QUDA

Description:

Library for Lattice QCD calculations using GPUs.

Features:

CUDA supports the following fermion formulations: Wilson,Wilson-clover,Twisted mass,Improved staggered (asqtad or HISQ) and Domain wall.

Multi GPU support:

Yes

RAMSES

Description:

Simulates astrophysical problems on different scales (e.g. star formation, galaxy dynamics, cosmological structure formation).

Features:

CUDA acceleration is applied for radiative transfer for reionization, and the hydrodynamic solver using AMR.

Multi GPU support:

Yes

XGC

Description:

Simulates edge effects for MHD plasma physics

Features:

The particle push portion has been optimized with CUDA and is being fully optimized with OpenACC and CUDA.

Multi GPU support:

Yes

SCIENTIFIC VISUALIZATION

3D Slicer

Description:

Medical visualization & segmentation

Features:

Rendering, image processing

Multi GPU support:

Single only

CEI EnSight

Description:

Visualization and analysis application for CAE

Features:

Rendering

Multi GPU support:

Yes

FluoRender (SCI, U of Utah)

Description:

Interactive rendering tool for confocal microscopy data visualization.

Features:

Multi-channel volume rendering

Multi GPU support:

Single only

GPULib for IDL

Description:

Data analysis application

Features:

Analysis tasks

Multi GPU support:

Single only

HVR (LCSE, U of Minnesota)

Description:

Interactive volume rendering application

Features:

Volume rendering

Multi GPU support:

Yes

ImageVis3D (SCI, U of Utah)

Description:

Simple, scalable, and interactive volume rendering application.

Features:

Out-of-core volume rendering

Multi GPU support:

Single only

IntelligentLight FieldView

Description:

Visualization application for CFD

Features:

Rendering

Multi GPU support:

Single only

MathWorks - MATLAB

Description:

Data analysis and visualization application

Features:

Rendering and analysis tasks

Multi GPU support:

Single only

ParaView

Description:

Scalable data anlysis and visualization application

Features:

Rendering and analysis tasks

Multi GPU support:

Yes

Seg3D (SCI, U of Utah)

Description:

Segmentation application for medical data

Features:

Rendering, image processing

Multi GPU support:

Single only

Visulalization Toolkit (VTK)

Description:

Data anlysis and visualization toolkit

Features:

Rendering

Multi GPU support:

Single only

VisIt

Description:

Scalable data anlysis and visualization application

Features:

Rendering and analysis tasks

Multi GPU support:

Yes

vl3 (Argonne National Lab)

Description:

Large dataset visualization in cosmology, astrophysics, and biosciences fields.

Features:

Volume rendering of particles

Multi GPU support:

Yes

VMD (U of Illionis, Urbana-Champaign)

Description:

Visualization and analysis of large bio- molecular systems in 3-D graphics.

Features:

High-qulity rendering, large structures (100M atoms), analysis and visualization tasks, multiple GPU support for display of molecular orbitals.

Multi GPU support:

Yes