SLIDE 1
PyStokes: A case study of accelerating Python using Cython
Rajesh Singh Jan 31, 2015
1
Contributors
2
PyStokes: A case study of accelerating Python using Cython Rajesh - - PowerPoint PPT Presentation
PyStokes: A case study of accelerating Python using Cython Rajesh - - PowerPoint PPT Presentation
PyStokes: A case study of accelerating Python using Cython Rajesh Singh Jan 31, 2015 1 Contributors 2 Outline Stokes law Rigid body motion of active colloids 3 Outline Stokes law Rigid body motion of active colloids
SLIDE 2
SLIDE 3
Outline
- Stokes law
- Rigid body motion of active colloids
3
SLIDE 4
Outline
- Stokes law
- Rigid body motion of active colloids
- Python
3
SLIDE 5
Outline
- Stokes law
- Rigid body motion of active colloids
- Python
- Cython
3
SLIDE 6
Outline
- Stokes law
- Rigid body motion of active colloids
- Python
- Cython
- Python and Cython
3
SLIDE 7
Outline
- Stokes law
- Rigid body motion of active colloids
- Python
- Cython
- Python and Cython
- PyStokes
3
SLIDE 8
Outline
- Stokes law
- Rigid body motion of active colloids
- Python
- Cython
- Python and Cython
- PyStokes
- Benchmarks
3
SLIDE 9
Stokes law
4
SLIDE 10
Stokes law
Vn = Fn 6πηa
4
SLIDE 11
Stokes law
Vn = Fn 6πηa Ωn = Tn 8πηa3
4
SLIDE 12
Rigid body motion of active colloids
Vn =
N
- m=1
- µTT
nm · FB m + µTR nm · TB m
- +
- lσ, m
π(T, lσ)
nm
· V(lσ)
m ,
Ωn =
N
- m=1
- µRT
nm · FB m + µRR nm · TB m
- +
- lσ, m
π(R, lσ)
nm
· V(lσ)
m .
- J. Stat. Mech. (2015) P06017
5
SLIDE 13
Python
1 Free and open source 2 High level & interpreted 3 Interactive environment 4 Object-oriented
6
SLIDE 14
Python
1 Free and open source 2 High level & interpreted 3 Interactive environment 4 Object-oriented 5 Speed 6 Dictionary lookups 7 Function calling overheads 8 GIL - global interpreter lock
6
SLIDE 15
Cython
- Attempt to make a superset of python
7
SLIDE 16
Cython
- Attempt to make a superset of python
- High level coolness of python along with the speed of C
7
SLIDE 17
Cython
- Attempt to make a superset of python
- High level coolness of python along with the speed of C
- Compiled
7
SLIDE 18
Cython
- Attempt to make a superset of python
- High level coolness of python along with the speed of C
- Compiled
- Cdef variables, attributes, functions
7
SLIDE 19
Cython
- Attempt to make a superset of python
- High level coolness of python along with the speed of C
- Compiled
- Cdef variables, attributes, functions
- Supports parallelism (openMP) by opening GIL
7
SLIDE 20
Cython
- Attempt to make a superset of python
- High level coolness of python along with the speed of C
- Compiled
- Cdef variables, attributes, functions
- Supports parallelism (openMP) by opening GIL
- Memoryviews allow efficient access to memory buffers like
numpy arrays without python overhead
7
SLIDE 21
Cython
- Attempt to make a superset of python
- High level coolness of python along with the speed of C
- Compiled
- Cdef variables, attributes, functions
- Supports parallelism (openMP) by opening GIL
- Memoryviews allow efficient access to memory buffers like
numpy arrays without python overhead
- Steps involves in building a cython code
1 A .pyx file is compiled by cython to .c
7
SLIDE 22
Cython
- Attempt to make a superset of python
- High level coolness of python along with the speed of C
- Compiled
- Cdef variables, attributes, functions
- Supports parallelism (openMP) by opening GIL
- Memoryviews allow efficient access to memory buffers like
numpy arrays without python overhead
- Steps involves in building a cython code
1 A .pyx file is compiled by cython to .c 2 The .c file is then compiled by the C compiler
7
SLIDE 23
Cython
- Attempt to make a superset of python
- High level coolness of python along with the speed of C
- Compiled
- Cdef variables, attributes, functions
- Supports parallelism (openMP) by opening GIL
- Memoryviews allow efficient access to memory buffers like
numpy arrays without python overhead
- Steps involves in building a cython code
1 A .pyx file is compiled by cython to .c 2 The .c file is then compiled by the C compiler 3 Building can be done in a single step using a setup.py
7
SLIDE 24
Python and Cython
8
SLIDE 25
Python and Cython
9
SLIDE 26
Python and Cython
10
SLIDE 27
Python and Cython
- Cython is three orders of magnitude FASTER than Python!
10
SLIDE 28
Python and Cython
- Cython is three orders of magnitude FASTER than Python!
- The Cython code is as fast as C code!!
10
SLIDE 29
PyStokes: README.md
- Cython library for computing Stokes flows produced by spheres
- The library computes flow and RBM
- Geometries supported
- unbounded
- wall-bounded
- periodic
- Free and open source
- Planned developments
- linear solvers
- fast multipole accelerations
- data layout
- References
- R. Singh, A. Laskar, and R. Adhikari. PyStokes: Hampi, Nov 2014.
- S. Ghose and R. Adhikari. Phys. Rev. Lett., 112(11):118102, 2014.
- R. Singh, S. Ghose and R. Adhikari. J. Stat. Mech, P06017, 2015
11
SLIDE 30
PyStokes: Usage
import pystokes , p y f o r c e s import numpy as np a , Np = 1 , 100 L , dim = 128 , 3 dt , T = 0.01 , 100 v = np . ze ros ( dim∗Np ) ; r = v ; F = v ; pRbm = pystokes . p e r i o d i c .Rbm(a , Np, L) f f = p y f o r c e s . f o r c e F i e l d s . Forces (Np) for t t in range (T) : f f . sedimentation (F , g=−10)
- pRbm. s t o k e s l e t V ( v ,
r , F) r = ( r + (F/(0.75∗ a ) + v )∗ dt)%L
12
SLIDE 31
PyStokes: Example
Figure: Crowley instability in a sedimenting lattice.
13
SLIDE 32
PyStokes: Benchmarks
Figure: Propulsion matrix calculation using the PyStokes library
Present implementation scales
- linearly with # cores
- quadratic with # particles
14
SLIDE 33
Summary
- Free and open source library
- Efficient and fast evaluation of Stokes flow
- A python front end for the user
- Present implementation scales
- linearly with # cores
- quadratic with # particles
15