1块K80到底能加速多少？
相对于最新的至强CPU

Pricing for Gaussian Products

http://gaussian.com/pricing/

What’s New in GaussView6
GaussView 6 Features at a Glance

Features new to GaussView 6 are in Blue; features enhanced in GaussView 6 are in Green.

This page is under construction.

Examine Molecular Structures
Rotate, translate and zoom in 3D with:
Mouse operations
Precision positioning toolbar
Available in every graphical display
View numeric value for any structural parameter
Use multiple synchronized or independent views of same structure
Customize display layout
Manipulate multiple structures individually or as an ensemble
Display formats: wire frame, tubes, ball & stick/bond type, space fill (CPK) style
View per-atom labels for element, serial number, NMR shielding (when available)
Visualize depth with fog feature
Display stereochemistry info
Highlight, display or hide atoms based on rich selection capabilities
Persistent highlighting available
Building/Modifying Molecules
Convenient palettes:
Atoms (including hybridization)
Functional groups
Rings
Amino acids (central fragment, amino- or carboxyl-terminated)
Nucleosides (central fragment, C3’-, C5’-terminated, free forms)
Custom fragment libraries
Import standard molecule file formats:
PDB, AMBER-created PDB
Gaussian input (.gjf and .com), output (.log), checkpoint (.chk and .fchk), cube (.cub), and frequency (.gfrq) files
Sybyl files: .mol2, .ml2
MDL files: .mol, .rxn, .sdf
Crystallographic Information files: .cif
Optionally include intermediate structures from optimizations etc.
Multi-structure .sdf and .mol2 files
Include/discard waters on PDB import
Optionally apply standard residue bonding on PDB import
Include/convert lone pairs for .mol2
Accurately add hydrogens automatically or manually to an entire molecule or to selected residues or secondary structures
An advanced open dialog, allowing options to be modified and retained:
Reading intermediate geometries
Saving the formatted checkpoint file, using the bond table, and weak bond inclusion
Gaussian log file load order
Gaussian input file load order
PDB file settings
Sybyl Mol2 file settings
Modify bond type/length, bond angles, dihedral angles
Rationalize structures with an advanced Clean function
Recompute bonding on demand
Constrain structure to specific point group symmetry; ability to reduce symmetry to a specified subgroup
Mirror invert structure
Invert structure about selected atom
Place atom/fragment at centroid position of selected atoms
Define named groups of atoms via:
Click, marquee, brush selection modes (customizable)
Complex filters combining atom type, number, MM settings, ONIOM layer
Select by PDB residue and/or secondary structure (e.g., helix, chain)
Expand selections by bond or proximity
Use groups for display purposes and in Gaussian input
Specify nonstandard isotopes
Customize fragment placement behavior
Specify bonding based on geometry: distances between atoms
Setup Features for Specific Job Types
Specify input for complex calculations via simple mouse/spreadsheet operations:

Build unit cells for polymers, 2D surfaces and crystals (periodic boundary conditions)
Constrain to specific space group symmetry
Assign atoms to ONIOM layers by:
Direct selection
Bond proximity to specified atom
Absolute distance from specified atom
PDB file residue, secondary structure
Complex selection criteria
View/specify MM atom types and charges
Add/redefine redundant internal coordinates
Specify frozen atoms/coordinates during geometry optimizations
Specify atom equivalences for QST2/QST3 transition state optimizations
Manipulate MOs: Select, rearrange and/or reoccupy orbitals for CASSCF etc
Define fragments for fragment guess/counterpoise calculations
Assign fragment-specific charges and spin multiplicities
Include PDB data in molecule specification
Select normal modes for frequency calculations
Specify atoms for NMR spin-spin coupling
Specify a search for Conformations using the GMMX add-on
Optional full AMPAC interface integration
Setup jobs that retrieve data from existing checkpoint files
Support for latest Gaussian 16 features
Prepare and Run Gaussian Calculations
Create input files via a straightforward menu-driven interface:
Select job/method/basis from pop-up menu; related options appear automatically
Supports all Gaussian 16 features
Convenient access to commonly-used general options (e.g., SCF=QC)
Extra input sections in imported files are retained
Preview the input file before it is submitted to Gaussian
Convert Files between different formats; Ampac Input files, Gaussian Input files, MDL Mol files, MDL SDF files, GMMX Input files, and Sybyl Mol2 files
Select solvent and specify other parameters for calculations in solution
Specify any Link 0 command
Specify setting for multiprocessor and cluster/network parallel jobs
Greater control over Link 0 files
Use calculation schemes to set up jobs from templates
“Quick launch” Gaussian jobs with a single mouse click
Molecule specification created automatically
Optional connectivity section
Monitor/control local Gaussian and utility processes
Stream log files in a text-searchable window
Initiate remote jobs via a customizable script
Generate job-specific input automatically
PBC translation vector for periodic structures like polymers and crystals
Orbital alterations
Multiple molecule specifications for QST2/QST3 transition state searches
Fragment guess and counterpoise per-fragment charge and spin multiplicity
Set up multiple job files through a one step process
Integrated queuing tool to run a list of jobs without additional actions
Examining and Visualizing Gaussian Results
Show calculation results summary; Expanded Thermo tab which displays predicted thermochemical quantities and Opt tab which displays optimization step information
A table displaying information about opened molecules that can be edited; which can be used to generate a table of results
Examine atomic changes: numerical values, color atoms by charge, dipole moment vector
Visualize atomic properties, predicted bond lengths, and predicted bond orders from a completed Gaussian calculation
Open any or all results from multi-job Gaussian input and output files
Create surfaces and contours for molecular orbitals, electron density, electrostatic potential, spin density, NMR shielding density
Display formats: 3D solid, translucent, wire mesh; 2D contour
Color surfaces by a separate property
Specify the desired contour plane
Load any cube created by Gaussian
Save computed cubes for future reuse
Perform operations on cubes (e.g., subtract for a difference density).
Display solvation cavity surface used in an SCRF calculation
Plot the results of an Optical Rotary Displacement calculation for any number of molecules in one group
Animate normal modes associated with vibrational frequencies
Indicate motion via displacement vector, dipole derivative unit vector
Displace structures any specified distance along normal mode
Select subset of modes for display
Save generated normal modes back to checkpoint file
Substitute isotopes in frequency analysis/normal modes
Ability to scale frequencies by a specified value
Options for looping, time delay between frame changes, and frames displayed per loop
Display spectra: IR, Raman, NMR, VCD, ROA, UV-Visible, Harmonic, Anharmonic etc.
Specify incident light frequency for frequency-dependent calculations
Display energy and other results from Gaussian trajectory calculations
Conformational Search Results:
View resulting energies on a plot
NMR Results:
Report absolute NMR chemical shifts or relative to reference compound
Export NMR summary data as text
Animate structure sequences: geometry optimizations, IRC reaction paths, potential energy surface scans, BOMD and ADMP trajectories
Single play or continuous looping
Play in reverse
Plots of related data are also produced
Display 3D surface plots for 2-variable scan calculations
Customize plot and spectra displays:
Zooming, scaling, inverting, etc.
Add molecular properties to plots
Mixture editor: Specify line/stick characteristics, plot a combination line and specify its appearance, assign weights to component data sets (Boltzmann averaging)
Advanced plot visualization customization; line color, canvas and background color, title, x- and y- axis settings, ect.
Save any image to a file (including customizations)
Produce web graphics: JPEG, PNG and other formats
Produce publication quality graphics files and printouts: TIFF, JPEG, vector graphics EPS and other formats
Create images at arbitrary size and resolution
Select full color or high quality grey scale formats
Specify custom colors and/or background
Save plots as images or textual data files
Save animations in GIF, MNG, MPEG4 format or as individual frames
Expanded controls over movie frame sequencing and delays
Customize GaussView
Set/save preferences for most aspects of GaussView functionality:

Control building toolbars individually
Colors: per-element, molecule window background, surfaces, transparency
Builder operation: atom and fragment join methods, adding hydrogens when needed, automated full or partial clean operations, etc.
Gaussian 16 calculation settings
Gaussian job execution methods
Display modes
Window placement and visibility
Icon sizes
File/directory locations
Image capture and printing defaults
Animation settings and movie defaults
Clean function parameters
Charge distribution display defaults
Custom bonding parameters defaults
GaussView Tips facility
Windows file extension associations
Help files integrated into individual dialogs
Quick Links
Basis Sets
Density Functional (DFT) Methods
Solvents List SCRF
References

http://gaussian.com/gv6features/

Gaussian 16 Rev. A.03 Release Notes
New Modeling Capabilities
TD-DFT analytic second derivatives for predicting vibrational frequencies/IR and Raman spectra and performing transition state optimizations and IRC calculations for excited states.
EOMCC analytic gradients for performing geometry optimizations.
Anharmonic vibrational analysis for VCD and ROA spectra: see Freq=Anharmonic.
Vibronic spectra and intensities: see Freq=FCHT and related options.
Resonance Raman spectra: see Freq=ReadFCHT.
New DFT functionals: M08 family, MN15, MN15L.
New double-hybrid methods: DSDPBEP86, PBE0DH and PBEQIDH.
PM7 semi-empirical method.
Adamo excited state charge transfer diagnostic: see Pop=DCT.
The EOMCC solvation interaction models of Caricato: see SCRF=PTED.
Generalized internal coordinates, a facility which allows arbitrary redundant internal coordinates to be defined and used for optimization constraints and other purposes. See Geom=GIC and GIC Info.
Performance Enhancements
NVIDIA K40 and K80 GPUs are supported under Linux for Hartree-Fock and DFT calculations. See the Using GPUs tab for details.
Parallel performance on larger numbers of processors has been improved. See the Parallel Performance tab for information about how to get optimal performance on multiple CPUs and clusters.
Gaussian 16 uses an optimized memory algorithm to avoid I/O during CCSD iterations.
There are several enhancements to the GEDIIS optimization algorithm.
CASSCF improvements for active spaces ≥ (10,10) increase performance and make active spaces of up to 16 orbitals feasible (depending on the molecular system).
Significant speedup of the core correlation energies for W1 compound model.
Gaussian 16 incorporates algorithmic improvements for significant speedup of the diagonal, second-order self-energy approximation (D2) component of composite electron propagator (CEP) methods as described in [DiazTinoco16]. See EPT.
Usage Enhancements
Tools for interfacing Gaussian with other programs, both in compiled languages such as Fortran and C and with interpreted languages such as Python and Perl. Refer to the Interfacing to Gaussian 16 page for details.
Parameters specified in Link 0 (%) input lines and/or in a Default.Route file can now also be specified via either command-line
arguments or environment variables. See the Link 0 Equivalences tab for details.
Compute the force constants are every nth step of a geometry optimization: see Opt=Recalc.

USING GPU:
Gaussian 16 can use NVIDIA K40 and K80 GPUs under Linux. Earlier GPUs do not have the computational capabilities or memory size to run the algorithms in Gaussian 16. Gaussian 16 does not yet support the Tesla-Pascal series.

Allocating sufficient amounts of memory to jobs is even more important when using GPUs than for CPUs, since larger batches of work must be done at the same time in order to use the GPUs efficiently. The K40 and K80 units can have up to 16 GB of memory. Typically, most of this should be made available to Gaussian. Giving Gaussian 8-9 GB works well when there is 12 GB total on each GPU; similarly, allocating Gaussian 11-12 GB is appropriate for a 16 GB GPU. In addition, at least an equal amount of memory must be available for each CPU thread which is controlling a GPU.

When using GPUs, it is essential to have the GPU controlled by a specific CPU. The controlling CPU should be as physically close as possible to the GPU it is controlling. The hardware arrangement on a system with GPUs can be checked using the nvidia-dmi utility. For example, this output is for a machine with two 16-core Haswell CPU chips and three K80 boards, each of which has two GPUs:

GPU0 GPU1 GPU2 GPU3 GPU4 GPU5 GPU6 GPU7 CPU  Affinity                  
GPU0    X  PIX  SOC  SOC  SOC  SOC  SOC SOC  0-15     cores on first chip
GPU1  PIX    X  SOC  SOC  SOC  SOC  SOC SOC  0-15                  
GPU2  SOC  SOC    X  PIX  PHB  PHB  PHB PHB  16-31    cores on second chip                  
GPU3  SOC  SOC  PIX    X  PHB  PHB  PHB PHB  16-31                  
GPU4  SOC  SOC  PHB  PHB    X  PIX  PXB PXB  16-31                  
GPU5  SOC  SOC  PHB  PHB  PIX    X  PXB PXB  16-31                  
GPU6  SOC  SOC  PHB  PHB  PXB  PXB    X PIX  16-31                  
GPU7  SOC  SOC  PHB  PHB  PXB  PXB  PIX   X  16-31                  
The important part of this output is the CPU affinity. This example shows that GPUs 0 and 1 (on the first K80 card) are connected to the CPUs on chip 0 while GPUs 2-5 (on the other two K80 cards) are connected to the CPUs on chip 1.

The GPUs to use for a calculation and their controlling CPUs are specified with the %GPUCPU Link 0 command. This command takes one parameter:

%GPUCPU=gpu-list=controlling-cpus                  
where gpu-list is a comma-separated list of GPU numbers, possibly including numerical ranges (e.g., 0-4,6), and controlling-cpus is a similarly-formatted list of controlling CPU numbers. To continue with the same example, a job which uses all the CPUs—20 CPUs doing parts of the computation and 6 controlling GPUs—would use the following Link 0 commands:

%CPU=0-31                  
%GPUCPU=0,1,2,3,4,5=0,1,16,17,18,19                  
This pins threads 0-31 to CPUs 0-31 and then uses GPU0 controlled by CPU 0, GPU1 controlled by CPU 1, GPU2 controlled by CPU 16, and so on. Note that the controlling CPUs are included in %CPU. The GPU and CPU lists could be expressed more tersely as:

%CPU=0-31                  
%GPUCPU=0-5=0-1,16-19                  
Normally one uses consecutive numbering in the obvious way, but things can be associated differently in special cases. For example, suppose on the same machine one already had one job using 6 CPUs running with %CPU=16-21. Then if one wanted to use the other 26 CPUs with 6 controlling GPUs one would specify:

%CPU=0-15,22-31                  
%GPUCPU=0-5=0-1,22-25                  
This would create 26 threads, with GPUs controlled by the threads on CPUs 0, 1, 22, 23, 24 and 25.

GPUs are not helpful for small jobs but are effective for larger molecules when doing DFT energies, gradients and frequencies (for both ground and excited states). They are not used effectively by post-SCF calculations such as MP2 or CCSD. Each GPU is several times faster than a CPU but since on modern machines there are typically many more CPUs than GPUs, it is important to use all the CPUs as well as the GPUs and the speedup from GPUs is reduced because many CPUs are also used effectively. For example, if the GPU is 5x faster than a CPU, then the speedup from going to 1 CPU to 1 CPU + 1 GPU would be 5x, but the speedup going from 32 CPUs to 32 CPUs + 8 GPUs (i.e., 24 CPUs + 8 GPUs) would be equivalent to 24 + 5×8 = 44 CPUs, for a speedup of 44/32 or about 1.4x.

GPUs on nodes in a cluster can be used. Since the %CPU and %GPUCPU specifications are applied to each node in the cluster, the nodes must have identical configurations (number of GPUs and their affinity to CPUs); since most clusters are collections of identical nodes, this is not usually a problem.
http://gaussian.com/relnotes/

大伙儿期待ing

墨灵格的肖经理说还要再等等

强烈期待GPU加速的测试

yjcmwgk 发表于 2017-1-13 09:11
从1994年开始算 Gaussian 94
4年后 Gaussian 98
5年后 Gaussian 03

哈哈，没毛病

从1994年开始算 Gaussian 94
4年后 Gaussian 98
5年后 Gaussian 03
6年后 Gaussian 09
7年后 Gaussian 16
坐等Gaussian 24就好了

都17年了～～～～～～～～～～～～～～～

谢谢分享

是真的吗？

有网上破解？

咨询一下老师：
这个高斯16开始卖了吗？linux的多少钱？
相比g09，主要改进在哪些地方？
要是买的话，是什么流程？
因为，我们这边老师让调研一下，想要买这个16.
谢谢指点