Julia Language
for electronic struture calculations
II LatinXChem - Twitter ConferenceFederal University of Santa CatarinaLetícia Maria Pequeno Madureira Group of Electronic Molecular Structure Supervisor: Giovanni Finoto Caramori September, 2021
Brief guide
How to navigate through this interactive presentation?
Index boxes move to referred pageArrows move to the next pagePlus signals indicates explanation boxesSocial media symbols open network web tabs
The graphs on results section are interactive! Click on the bars to see the values.
INDEX
METHODOLOGY
HARTREE-FOCK
JULIA LANG
REFERENCES
CONCLUSIONS
BENCHMARK
JULIA LANGUAGE
FASTER CODING
Motivations
A programming language is a formal code of communication that contains a set of instructions to generate a series of computer outputs through the implementation of algorithms. Nowadays, scientific computing is standing right to the next level of influence in knowledge discovery, aiming new theoretical analysis and time gain improving, accelerating the finding of new results in Chemistry, Biology, Physics, and Mathematics. In the last ten years, new programming languages have emerged, such as Julia, to improve scientific computation with better numerical programming tools and Machine Learning features. Thus, Julia Language was chosen to demonstrate the potential for accelerating electronic structure calculations (in this case Hartree Fock), ushering in a new phase for software development in chemistry, correlating syntactic ease with speed in a very versatile hybrid high-low-level language.
Python
Julia
Loop rates
The Python for loop take around 1.73 seconds to run, while Julia for loop lasted 2.00 ms, being almost 1000 times faster. Remembering that speeds depend on the processing resources of the machine, but still the results prove that Julia is faster than Python.
HARTREE-FOCK
SELF CONSISTENT FIELD THEORY
DENSITY MATRIX
CALCULATE POTENTIAL
GUESS WAVEFUNCTION
No
CONVERGENCE CRITERIA MET?
DENSITY MATRIX
SCHRÖDINGER EQUATION
COMPARISON
HARTREE-FOCK BENCHMARKING
HARTREE-FOCK
Linear Equation Systems
Linear Equation Systems
Timing Measures
Timing Measures
Basis Set
Basis Set
INITIAL RESULTS
JULIA BENCHMARKING
Run-time comparison
Relative Energies
Discussions
Apointments
- Julia Lang code is the fastest;
- Python Code is the laziest;
- Weak-electron molecules are better energetically described;
- The energy bars are nearby, but Python has the increased error.
moving forward
FUTURE PERSPECTIVES
PROCESS
RESEARCH INTENTIONS
USER FRIENDLY
Graphical interface interaction
FASTER USE
New integration methods benchmarking
WIDER USE
More basis set implementations
Aromatic parameters calculation
CHEMICAL INTERPRETER
ReFERENCES
THEORETICAL BASIS
A New Kid on the Block: Application of Julia to Hartree–Fock Calculations
- J. Chem. Theory Comput. 2020, 16, 8, 5006–5013
Modern quantum chemistry: introduction to advanced electronic structure theory
-Szabo, Attila, and Neil S. Ostlund. Courier Corporation, 2012.
Student-Friendly Guide to Molecular Integrals
- J. Chem. Educ. 2018, 95, 9, 1572–1578
THANKS
Untitled genially
Letícia Madureira
Created on September 14, 2021
Start designing with a free template
Discover more than 1500 professional designs like these:
View
Modern Zen Presentation
View
Newspaper Presentation
View
Audio tutorial
View
Pechakucha Presentation
View
Desktop Workspace
View
Decades Presentation
View
Psychology Presentation
Explore all templates
Transcript
Julia Language
for electronic struture calculations
II LatinXChem - Twitter ConferenceFederal University of Santa CatarinaLetícia Maria Pequeno Madureira Group of Electronic Molecular Structure Supervisor: Giovanni Finoto Caramori September, 2021
Brief guide
How to navigate through this interactive presentation?
Index boxes move to referred pageArrows move to the next pagePlus signals indicates explanation boxesSocial media symbols open network web tabs
The graphs on results section are interactive! Click on the bars to see the values.
INDEX
METHODOLOGY
HARTREE-FOCK
JULIA LANG
REFERENCES
CONCLUSIONS
BENCHMARK
JULIA LANGUAGE
FASTER CODING
Motivations
A programming language is a formal code of communication that contains a set of instructions to generate a series of computer outputs through the implementation of algorithms. Nowadays, scientific computing is standing right to the next level of influence in knowledge discovery, aiming new theoretical analysis and time gain improving, accelerating the finding of new results in Chemistry, Biology, Physics, and Mathematics. In the last ten years, new programming languages have emerged, such as Julia, to improve scientific computation with better numerical programming tools and Machine Learning features. Thus, Julia Language was chosen to demonstrate the potential for accelerating electronic structure calculations (in this case Hartree Fock), ushering in a new phase for software development in chemistry, correlating syntactic ease with speed in a very versatile hybrid high-low-level language.
Python
Julia
Loop rates
The Python for loop take around 1.73 seconds to run, while Julia for loop lasted 2.00 ms, being almost 1000 times faster. Remembering that speeds depend on the processing resources of the machine, but still the results prove that Julia is faster than Python.
HARTREE-FOCK
SELF CONSISTENT FIELD THEORY
DENSITY MATRIX
CALCULATE POTENTIAL
GUESS WAVEFUNCTION
No
CONVERGENCE CRITERIA MET?
DENSITY MATRIX
SCHRÖDINGER EQUATION
COMPARISON
HARTREE-FOCK BENCHMARKING
HARTREE-FOCK
Linear Equation Systems
Linear Equation Systems
Timing Measures
Timing Measures
Basis Set
Basis Set
INITIAL RESULTS
JULIA BENCHMARKING
Run-time comparison
Relative Energies
Discussions
Apointments
moving forward
FUTURE PERSPECTIVES
PROCESS
RESEARCH INTENTIONS
USER FRIENDLY
Graphical interface interaction
FASTER USE
New integration methods benchmarking
WIDER USE
More basis set implementations
Aromatic parameters calculation
CHEMICAL INTERPRETER
ReFERENCES
THEORETICAL BASIS
A New Kid on the Block: Application of Julia to Hartree–Fock Calculations
- J. Chem. Theory Comput. 2020, 16, 8, 5006–5013
Modern quantum chemistry: introduction to advanced electronic structure theory
-Szabo, Attila, and Neil S. Ostlund. Courier Corporation, 2012.
Student-Friendly Guide to Molecular Integrals
- J. Chem. Educ. 2018, 95, 9, 1572–1578
THANKS