enthought

Scientific Computing
with Python
[Advanced Topics]
Eric Jones
[email protected]

Travis Oliphant
[email protected]

Enthought
www.enthought.com

Brigham Young University

http://www.ee.byu.edu/

enthought

Topics

Python as Glue
Wrapping Fortran Code
Wrapping C/C++
Parallel Programming

enthought

Python as Glue

enthought

Why Python for glue?

Python reads almost like pseudo-code so its easy to
pick up old code and understand what you did.
Python has dynamic typing and dynamic binding --allows very flexible coding.
Python is object oriented.
Python has high-level data structures like lists,
dictionaries, strings, and arrays all with useful methods.
Python has a large module library (batteries included)
and common extensions covering internet protocols and
data, image handling, and scientific analysis.
Python development is 5-10 times faster than C/C++ and
3-5 times faster than Java

enthought

Electromagnetics Example
(1) Parallel simulation
(2) Create plot
(3) Build HTML page

(4) FTP page to Web Server

(5) E-mail users that results
are available.

enthought

How is Python glue?

Legacy
App.
Parallel
Process.

Internet
Python

Users

Equipment
New
Algorithm

enthought

Why is Python good glue?

Python can be embedded into any C or C++ application
Provides your legacy application with a powerful scripting language
instantly.

Python can interface seamlessly with Java

Jython www.jython.org
JPE
jpe.sourceforge.net
Python can interface with critical C/C++ and Fortran
subroutines
Rarely will you need to write a main-loop again.
Python does not directly call the compiled routines, it uses
interfaces (written in C or C++) to do it --- the tools for
constructing these interface files are fantastic (sometimes
making the process invisible to you).

enthought

Tools
C/C++ Integration
SWIG
SIP
Pyrex
boost
weave

www.swig.org
www.riverbankcomputing.co.uk/sip/index.php
nz.cosc.canterbury.ac.nz/~greg/python/Pyrex
www.boost.org/libs/python/doc/index.html
www.scipy.org/site_content/weave

FORTRAN Integration
f2py
PyFort

cens.ioc.ee/projects/f2py2e/
pyfortran.sourceforge.net

enthought

f2py
Author: Pearu Peterson at Center for
Nonlinear Studies Tallinn, Estonia
Automagically wraps Fortran 77/90/95
libraries for use in Python. Amazing.
f2py is specifically built to wrap Fortran
functions using NumPy arrays.

enthought

Simplest f2py Usage

Fortran
File

Python
PythonExtension
Extension
Module
Module

fcopy.f
fcopymodule.so
fcopymodule.so
f2py c fcopy.f m fcopy

Compile code and build an

extension module

Name the extension

module fcopy.

enthought

Simplest Usage Result

Fortran file fcopy.f
C
SUBROUTINE FCOPY(AIN,N,AOUT)
C
>>> import fcopy
DOUBLE COMPLEX AIN(*)
>>> info(fcopy)
INTEGER N
This module 'fcopy' is auto-generated with f2py
DOUBLE COMPLEX AOUT(*)
(version:2.37.233-1545).
DO 20 J = 1, N
Functions:
fcopy(ain,n,aout)
AOUT(J) = AIN(J)
>>> info(fcopy.fcopy)
20
CONTINUE
fcopy - Function signature:
END

>>> a = rand(1000) +
1j*rand(1000)

fcopy(ain,n,aout)
Required arguments:
ain : input rank-1 array('D') with bounds (*)
n : input int
aout : input rank-1 array('D') with bounds (*)

>>> b = zeros((1000,),D)
>>> fcopy.fcopy(a,1000,b)

Looks exactly like the

Fortran --- but now in Python!

enthought

More Sophisticated

Fortran
File

Interface
File

fcopy.f

fcopy.pyf
hand edit

Python
PythonExtension
Extension
Module
Module
fcopymodule.so
fcopymodule.so

f2py fcopy.f h fcopy.pyf m fcopy

f2py -c fcopy.pyf fcopy.f

enthought

More Sophisticated
Interface file fcopy.pyf
!
-*- f90 -*python module fcopy ! in
interface ! in :fcopy
subroutine fcopy(ain,n,aout) ! in :fcopy:fcopy.f
double complex dimension(n), intent(in) :: ain
integer, intent(hide),depend(ain) :: n=len(ain)
double complex dimension(n),intent(out) :: aout
end subroutine fcopy
end interface
end python module fcopy
! This file was auto-generated with f2py (version:2.37.233-1545).
! See http://cens.ioc.ee/projects/f2py2e/

fcopy - Function signature:

aout = fcopy(ain)
Required arguments:
ain : input rank-1 array('D') with
bounds (n)
Return objects:
aout : rank-1 array('D') with
bounds (n)

Give f2py
some hints as
to what these
variables are
used for and
how they may
be related in
Python.

>>> a = rand(100,F)
>>> b = fcopy.fcopy(a)
>>> print b.typecode()
D

More Pythonic behavior

enthought

Simply Sophisticated
Fortran File
fcopy.f

Python
PythonExtension
Extension
Module
Module

hand edit
fcopymodule.so
fcopymodule.so
f2py c fcopy.f m fcopy

Compile code and build an

extension module

Name the extension

module fcopy.

enthought

Simply Sophisticated
Fortran file fcopy2.f
C
SUBROUTINE FCOPY(AIN,N,AOUT)
A few directives can help
C
CF2PY INTENT(IN), AIN
f2py interpret the source.
CF2PY INTENT(OUT), AOUT
CF2PY INTENT(HIDE), DEPEND(A), N=LEN(A)
DOUBLE COMPLEX AIN(*)
INTEGER N
DOUBLE COMPLEX AOUT(*)
DO 20 J = 1, N
>>> import fcopy
AOUT(J) = AIN(J)
>>> info(fcopy.fcopy)
20
CONTINUE
fcopy - Function signature:
END
aout = fcopy(ain)

>>> a = rand(1000)
>>> import fcopy
>>> b = fcopy.fcopy(a)

Much more Python like!

Required arguments:
ain : input rank-1 array('D') with bounds (n)
Return objects:
aout : rank-1 array('D') with bounds (n)

enthought

Saving the Module C-File

f2py h alib.pyf m alib *.f

compile

*.f

Library
libflib.a

Interface File
flib.pyf
hand edited

C-extension
Module
flibmodule.c

f2py c alibmodule.c *.f

either one

Library of
Fortran Files

f2py alib.pyf

Shared extension
Module
flibmodule.so
f2py c alibmodule.c l alib

enthought

Multidimensional array issues

Python and Numeric use C conventions for array storage (row
major order). Fortran uses column major ordering.
Numeric:
A[0,0], A[0,1], A[0,2],, A[N-1,N-2], A[N-1,N-1]
(last dimension varies the fastest)
Fortran:
A(1,1), A(2,1), A(3,1), , A(N-1,N), A(N,N)
(first dimension varies the fastest)
f2py handles the conversion back and forth between the
representations if you mix them in your code. Your code will
be faster, however, if you can avoid mixing the representations
(impossible if you are calling out to both C and Fortran
libraries that are interpreting matrices differently).

enthought

scipy_distutils
How do I distribute this great new extension module?
Recipient must have f2py and scipy_distutils installed (both
are simple installs)
Create setup.py file
Distribute *.f files with setup.py file.
Optionally distribute *.pyf file if youve spruced up the
interface in a separate interface file.

Supported Compilers
g77, Compaq Fortran, VAST/f90 Fortran, Absoft F77/F90,
Forte (Sun), SGI, Intel, Itanium, NAG, Lahey, PG

enthought

Complete Example
In scipy.stats there is a function written entirely in Python
>>> info(stats.morestats._find_repeats)
_find_repeats(arr)
Find repeats in the array and return a list of the
repeats and how many there were.

Goal: Write an equivalent fortran function and link it in to

Python with f2py so it can be distributed with scipy_base
(which uses scipy_distutils) and be available for stats.
Python algorithm uses sort and so we will
need a fortran function for that, too.

enthought

Complete Example
Fortran file futil.f
C
Sorts an array arr(1:N) into
SUBROUTINE DQSORT(N,ARR)
CF2PY INTENT(IN,OUT,COPY), ARR
CF2PY INTENT(HIDE), DEPEND(ARR), N=len(ARR)
INTEGER N,M,NSTACK
REAL*8 ARR(N)
PARAMETER (M=7, NSTACK=100)
INTEGER I,IR,J,JSTACK, K,L, ISTACK(NSTACK)
REAL*8 A,TEMP

END
C
CF2PY
CF2PY
CF2PY
CF2PY
CF2PY

Finds repeated elements of ARR

SUBROUTINE DFREPS(ARR,N,REPLIST,REPNUM,NLIST)
INTENT(IN), ARR
INTENT(OUT), REPLIST
INTENT(OUT), REPNUM
INTENT(OUT), NLIST
INTENT(HIDE), DEPEND(ARR), N=len(ARR)
REAL*8 REPLIST(N), ARR(N)
REAL*8 LASTVAL
INTEGER REPNUM(N)
INTEGER HOWMANY, REPEAT, IND, NLIST, NNUM

END

#Lines added to setup_stats.py

#add futil module
sources = [os.path.join(local_path,
'futil.f']
name = dot_join(package,futil)
ext = Extension(name,sources)
config['ext_modules'].append(ext)

#Lines added to morestats.py

# (under stats)
import futil
def find_repeats(arr):
"""Find repeats in arr and
return (repeats, repeat_count)
"""
v1,v2, n = futil.dfreps(arr)
return v1[:n],v2[:n]

enthought

Complete Example
Try It Out!!
>>> from scipy import *
>>> a = stats.randint(1,30,size=1000)
>>> reps, nums = find_repeats(a)
>>> print reps
[

1.
12.
23.

2.
13.
24.

3.
14.
25.

4.
15.
26.

5.
16.
27.

6.
17.
28.

7.
8.
18. 19.
29.]

9.
20.

10.
21.

11.
22.

>>> print nums

[29 37 29 30 34 39 46 20 30 32 35 42 40 39 35 26 38 33 40
29 34 26 38 45 39 38 29 39 29]

New function is 25 times faster

than the plain Python version

enthought

Complete Example
Packaged for Individual release
#!/usr/bin/env python
# File: setup_futil.py
from scipy_distutils.core import Extension
ext = Extension(name = 'futil',
sources = ['futil.f'])

python setup_futil.py install

With futil.f in current directory this builds
and installs on any platform with a C
compiler and a fortran compiler that
scipy_distutils recognizes.

if __name__ == "__main__":
from scipy_distutils.core import setup
setup(name = 'futil',
description
= "Utility fortran functions",
author
= "Travis E. Oliphant",
author_email
= "[email protected]",
ext_modules = [ext]
)
# End of setup_futil.py

enthought

Weave

enthought

weave
weave.blitz()
Translation of Numeric array expressions to C/C++ for
fast execution
weave.inline()
Include C/C++ code directly in Python code for on-thefly execution
weave.ext_tools
Classes for building C/C++ extension modules in
Python

enthought

weave.inline
>>> import weave
>>> a=1
>>> weave.inline('std::cout << a << std::endl;',['a'])
sc_f08dc0f70451ecf9a9c9d4d0636de3670.cpp
Creating library <snip>
1
>>> weave.inline('std::cout << a << std::endl;',['a'])
1
>>> a='qwerty'
>>> weave.inline('std::cout << a << std::endl;',['a'])
sc_f08dc0f70451ecf9a9c9d4d0636de3671.cpp
Creating library <snip>
qwerty
>>> weave.inline('std::cout << a << std::endl;',['a'])
qwerty

enthought

Support code example

>>> import weave
>>> a = 1
>>> support_code = int bob(int val) { return val;}
>>> weave.inline(return_val = bob(a);,['a'],support_code=support_code)
sc_19f0a1876e0022290e9104c0cce4f00c0.cpp
Creating library <snip>
1
>>> a = 'string'
>>> weave.inline(return_val = bob(a);,['a'],support_code = support_code)
sc_19f0a1876e0022290e9104c0cce4f00c1.cpp
C:\DOCUME~1\eric\LOCALS~1\Temp\python21_compiled\sc_19f0a1876e0022290e9104c0cce4
f00c1.cpp(417) : error C2664: 'bob' : cannot convert parameter 1 from 'class Py:
:String' to 'int' No user-defined-conversion operator available that can
perform this conversion, or the operator cannot be called
Traceback (most recent call last):
<snip>
weave.build_tools.CompileError: error: command '"C:\Program Files\Microsoft Visu
al Studio\VC98\BIN\cl.exe"' failed with exit status 2

enthought

ext_tools example
import string
from weave import ext_tools
def build_ex1():
ext = ext_tools.ext_module('_ex1')
# Type declarations define a sequence and a function
seq = []
func = string.upper
code = """
py::tuple args(1);
py::list result(seq.length());
for(int i = 0; i < seq.length();i++)
{
args[0] = seq[i];
result[i] = PyEval_CallObject(func,py::tuple(args[0]));
}
return_val = result;
"""
func = ext_tools.ext_function('my_map',code,['func','seq'])
ext.add_function(func)
ext.compile()
try:
from _ex1 import *
except ImportError:
build_ex1()
from _ex1 import *
if __name__ == '__main__':
print my_map(string.lower,['asdf','ADFS','ADSD'])

enthought

Efficiency Issues
PSEUDO C FOR STANDARD
NUMERIC EVALUATION
>>> c = a + b + c

FAST, IDIOMATIC C CODE

>>> c = a + b + c

tmp1
tmp2
// c code
// tmp1 = a + b
tmp1 = malloc(len_a * el_sz);
for(i=0; i < len_a; i++)
tmp1[i] = a[i] + b[i];
// tmp2 = tmp1 + c
tmp2 = malloc(len_c * el_sz);
for(i=0; i < len_c; i++)
tmp2[i] = tmp1[i] + c[i];

// c code
// 1. loops fused
// 2. no memory allocation
for(i=0; i < len_a; i++)
c[i] = a[i] + b[i] + c[i];

enthought

Finite Difference Equation

MAXWELLS EQUATIONS: FINITE DIFFERENCE TIME DOMAIN (FDTD),
UPDATE OF X COMPONENT OF ELECTRIC FIELD

Ex

x t
1
dH y
dH z
2 x
t
t
=

E +
x t dz
x t dy
x t x
x +
x +
1+
2
2
2 x

PYTHON VERSION OF SAME EQUATION, PRE-CALCULATED CONSTANTS

ex[:,1:,1:] =

ca_x[:,1:,1:]
* ex[:,1:,1:]
+ cb_y_x[:,1:,1:] * (hz[:,1:,1:] - hz[:,:-1,:])
- cb_z_x[:,1:,1:] * (hy[:,1:,1:] - hy[:,1:,:-1])

enthought

weave.blitz
weave.blitz compiles array
expressions to C/C++ code using
the Blitz++ library.
WEAVE.BLITZ VERSION OF SAME EQUATION
>>> from scipy import weave
>>> # <instantiate all array variables...>
>>> expr = ex[:,1:,1:] =
ca_x[:,1:,1:]
* ex[:,1:,1:]\
+ cb_y_x[:,1:,1:] * (hz[:,1:,1:] - hz[:,:-1,:])\
- cb_z_x[:,1:,1:] * (hy[:,1:,1:] - hy[:,1:,:-1])
>>> weave.blitz(expr)
< 1. translate expression to blitz++ expression>
< 2. compile with gcc using array variables in local scope>
< 3. load compiled module and execute code>

enthought

weave.blitz benchmarks
Equation

Numeric
(sec)

Inplace
(sec)

compiler
(sec)

Speed
Up

Float (4 bytes)
(512,512)

0.027

0.019

0.024

1.13

a = b + c + d (512x512)

0.060

0.037

0.029

2.06

5 pt. avg filter (512x512)

0.161

0.060

2.68

FDTD

0.890

0.323

2.75

a=b+c

(100x100x100)

Double (8 bytes)
(512,512)

0.128

0.106

0.042

3.05

a = b + c + d (512x512)

0.248

0.210

0.054

4.59

5 pt. avg filter (512x512)

0.631

0.070

9.01

FDTD

3.399

0.395

8.61

a=b+c

(100x100x100)

Pentium II, 300 MHz, Python 2.0, Numeric 17.2.0

Speed-up taken as ratio of scipy.compiler to standard Numeric
runs.

enthought

weave and Laplaces equation

Weave case study: An iterative solver for Laplaces Equation

u u
+ 2 =0
2
x
y
2

PURE PYTHON

1 Volt

2000 SECONDS

for i in range(1, nx-1):

for j in range(1, ny-1):
tmp = u[i,j]
u[i,j] = ((u[i-1, j] + u[i+1, j])*dy2 +
(u[i, j-1] + u[i, j+1])*dx2)
/ (2.0*(dx2 + dy2))
diff = u[i,j] tmp
err = err + diff**2
Thanks to Prabhu Ramachandran for designing and running this example. His
complete
write-up is available at:
www scipy org/site content/weave/python performance html

enthought

weave and Laplaces equation

USING NUMERIC

29.0 SECONDS

old_u = u.copy() # needed to compute the error.

u[1:-1, 1:-1] = ((u[0:-2, 1:-1] + u[2:, 1:-1])*dy2 +
(u[1:-1, 0:-2] + u[1:-1, 2:])*dx2)
* dnr_inv
err = sum(dot(old_u u))
WEAVE.BLITZ

10.2 SECONDS

old_u = u.copy() # needed to compute the error.

expr = \
u[1:-1, 1:-1] = ((u[0:-2, 1:-1] + u[2:, 1:-1])*dy2 +
(u[1:-1, 0:-2] + u[1:-1, 2:])*dx2)
* dnr_inv

weave.inline(expr,size_check=0)
err = sum((old_u u)**2)

enthought

weave and Laplaces equation

WEAVE.INLINE

4.3 SECONDS

code = """
#line 120 "laplace.py" (This is only useful for debugging)
double tmp, err, diff;
err = 0.0;
for (int i=1; i<nx-1; ++i) {
for (int j=1; j<ny-1; ++j) {
tmp = u(i,j);
u(i,j) = ((u(i-1,j) + u(i+1,j))*dy2 +
(u(i,j-1) + u(i,j+1))*dx2)*dnr_inv;
diff = u(i,j) - tmp;
err += diff*diff;
}
}
return_val = sqrt(err);
"""
err = weave.inline(code, ['u','dx2','dy2','dnr_inv','nx','ny'],
type_converters = converters.blitz,
compiler = 'gcc',
extra_compile_args = ['-O3','-malign-double'])

enthought

Laplace Benchmarks
Method

Speed
Up

1897.0

0.02

Numeric

29.0

1.00

weave.blitz

10.2

2.84

weave.inline

4.3

6.74

weave.inline (fast)

2.9

10.00

Python/Fortran (with f2py)

3.2

9.06

Pure C++ Program

2.4

12.08

Pure Python

Run Time
(sec)

Debian Linux, Pentium III, 450 MHz, Python 2.1, 192 MB RAM
Laplace solve for 500x500 grid and 100 iterations
Speed-up taken as compared to Numeric

enthought

SWIG

enthought

SWIG
Author: David Beazley at Univ. of Chicago
Automatically wraps C/C++ libraries for use in
Python. Amazing.
SWIG uses interface files to describe library
functions
No need to modify original library code
Flexible approach allowing both simple and complex
library interfaces

Well Documented

enthought

SWIG Process
Interface
File

C Extension
File
SWIG

lib.i

lib_wrap.c

Writing this is
your responsibility (kinda)
compile
Library Files
*.h files *.c files
compile

Python
PythonExtension
Extension
Module
Module
libmodule.so
libmodule.so

enthought

Simple Example
fact.h

example.i

#ifndef FACT_H
#define FACT_H

// Define the modules name

%module example

int fact(int n);

// Specify code that should

// be included at top of
// wrapper file.
%{
#include fact.h
%}

#endif

fact.c
#include fact.h
int fact(int n)
{
if (n <=1) return 1;
else return n*fact(n-1);
}

// Define interface. Easy way

// out - Simply include the
// header file and let SWIG
// figure everything out.
%include fact.h

enthought

Building the Module

LINUX
# Create example_wrap.c file
[ej@bull ej]$ swig python example.i
# Compile library and example_wrap.c code using
# position independent code flag
[ej@bull ej]$ gcc c fpic example_wrap.c fact.c
I/usr/local/include/python2.1
I/usr/local/lib/python2.1/config
# link as a shared library.
[ej@bull ej]$ gcc shared example_wrap.o fact.o
-o examplemodule.so
# test it in Python
[ej@bull ej]$ python
...
>>> import example
>>> example.fact(4)
24

\
\

For notes on how to use SWIG with

VC++ on Windows, see
http://www.swig.org/Doc1.1/HTML/Python.html#n2

enthought

The Wrapper File

example_wrap.c
static PyObject *_wrap_fact(PyObject *self, PyObject *args) {
PyObject *resultobj;
int arg0 ;
int result ;
/* parse the Python input arguments and extract */
name of function to return in case of error

if(!PyArg_ParseTuple(args,"i:fact",&arg0)) return NULL;

first arg in args read into arg0 as int

/* call the actual C function with arg0 as the argument*/

result = (int )fact(arg0);
/* Convert returned C value to Python type and return it*/
resultobj = PyInt_FromLong((long)result);
return resultobj;
}

enthought

SWIG Example 2
vect.h
int* vect(int x,int y,int z);
int sum(int* vector);

vect.c
#include <malloc.h>
#include vect.h
int* vect(int x,int y, int z){
int* res;
res = malloc(3*sizeof(int));
res[0]=x;res[1]=y;res[2]=z;
return res;
}
int sum(int* v) {
return v[0]+v[1]+v[2];
}

example2.i
Identical to example.i if you replace
fact with vect.
TEST IN PYTHON
>>> from example2 import *
>>> a = vect(1,2,3)
>>> sum(a)
6 #works fine!
# Lets take a look at the
# integer array a.
>>> a
'_813d880_p_int'
# WHAT THE HECK IS THIS???

enthought

Complex Objects in SWIG

SWIG treats all complex objects as
pointers.
These C pointers are mangled into string
representations for Pythons
consumption.
This is one of SWIGs secrets to wrapping
virtually any library automatically,
But the string representation is pretty
primitive and makes it un-pythonic to
observe/manipulate the contents of the
object.

enthought

Typemaps
example_wrap.c
static PyObject *_wrap_sum(PyObject *self, PyObject *args) {
...
if(!PyArg_ParseTuple(args,"O:sum",&arg0))
return NULL;
...
result = (int )sum(arg0);
...
return resultobj;
}

Typemaps allow you to insert type

conversion code into various location
within the function wrapper.
Not for the faint of heart. Quoting David:
You can blow your whole leg off,
including your foot!

enthought

Typemaps
The result? Standard C pointers are mapped to
NumPy arrays for easy manipulation in Python.
YET ANOTHER EXAMPLE NOW WITH TYPEMAPS
>>> import example3
>>> a = example3.vect(1,2,3)
>>> a
# a should be an array now.
array([1, 2, 3], 'i') # It is!
>>> example3.sum(a)
6
The typemaps used for example3 are included in the handouts.
Another example that wraps a more complicated C function used in
the previous VQ benchmarks is also provided. It offers more generic
handling 1D and 2D arrays.

enthought

Parallel Programming in Python

Parallel Computing Tools

Python has threads (sorta)

pyMPI(pympi.sf.net/)
pyre (CalTech)
PyPAR
(datamining.anu.edu.au/~ole/pypar/)
SCIENTIFIC
(starship.python.net/crew/hinsen)
COW (www.scipy.org)

enthought

enthought

Cluster Computing with Python

cow.py
Pure Python Approach
Easy to Use
Suitable for embarrassingly parallel tasks

pyMPI (Message Passing Interface)

Developed by Patrick Miller, Martin Casado et al. at
Lawrence Livermore National Laboratories
De-facto industry standard for high-performance computing
Vendor optimized libraries on Big Iron
Possible to integrate existing HPFortran and HPC codes
such as Scalapack (parallel linear algebra) into Python.

enthought

Threads
Python threads are built on POSIX and
Windows threads (hooray!)
Python threads share a lock that
prevents threads from invalid sharing
Threads pass control to another thread
every few instructions
during blocking I/O (if properly guarded)
when threads die

enthought

The threading module

from threading import Thread
a lower level thread library exists, but this
is much easier to use

a thread object can fork a new

execution context and later be joined
to another
you provide the thread body either by
creating a thread with a function or by
subclassing it

enthought

Making a thread
we will work at the prompt!
>>> from threading import *
>>> def f(): print hello
>>> T = Thread(target=f)
>>> T.start()

enthought

Thread operations
currentThread()
T.start()
T.join()
T.getName() / T.setName()
T.isAlive()
T.isDaemon() / T.setDaemon()

enthought

Passing arguments to a thread

>>> from threading import *
>>> def f(a,b,c): print hello,a,b,c
>>> T = Thread(target=f,args=(11,22),kwargs={c: )
>>> T.start()

enthought

Subclassing a thread
from threading import *
class myThread(Thread):
def __init__(self,x,**kw):
Thread.__init__(self,**kw) #FIRST!
self.x = x
def run():
print self.getName()
print I am running,self.x
T = myThread(100)
T.start()

NOTE: Only __init__ and run() are available for overload

enthought

CAUTION!
Threads are really co-routines!
Only one thread can operate on Python
objects at a time
Internally, threads are switched
If you write extensions that are intended
for threading, use
PY_BEGIN_ALLOW_THREADS
PY_END_ALLOW_THREADS

enthought

cow

enthought

Electromagnetic Scattering
Airborne Radar

Monostatic BackScatter from Buried

Landmine, Theta = 30, Phi = 0
0

RCS (dB)

-5
-10
-15
-20
-25
100

Buried Land Mine

Inputs
environment, target
mesh, and
multiple frequencies
Mem: KB to Mbytes
SMALL

Computation
N3 CPU
N2 storage
Time: a few seconds
to days
Mem: MB to GBytes
LARGE!

150

200
250
300
Frequency (MHz)

350

Outputs
Radar Cross Section
values
Mem: KB to MBytes
SMALL

400

enthought

cow.py
Master
Python
>>>

Socket
Node 0

Node 1

Node 2

Python
>>>

Python
>>>

Python
>>>

58

enthought

Cluster Creation
Master
>>> import scipy.cow
#
[name, port]
>>> machines = [['s0',11500],['s1',11500],['s2',11500]]
>>> cluster = scipy.cow.machine_cluster(machines)
>>>

Port numbers below 1024 are reserved by the OS and generally

must run as root or system. Valid port numbers are between
1025-49151. Be sure another program is not using the port
you choose.

enthought

Starting remote processes

Master
>>> cluster = scipy.cow.cluster(machines)
>>> cluster.start()

start() uses ssh to

start an interpreter
listening on port 11500
on each remote machine
s0

s1

s2

Python
>>>

Python
>>>

Python
>>>

enthought

Dictionary Behavior of Clusters

Master
# Set a global variable on each of the machines.
>>> cluster['a'] = 1

s0
Python
>>> a = 1

s1
Python
>>> a = 1

s2
Python
>>> a = 1

enthought

Dictionary Behavior of Clusters

Master
# Set a global variable on each of the machines.
>>> cluster['a'] = 1
# Retrieve a global variable from each machine.
>>> cluster['a']
( 1, 1, 1)
#(s0,s1,s2)

s0
Python
>>> a = 1
>>> a
1

s1
Python
>>> a = 1
>>> a
1

s2
Python
>>> a = 1
>>> a
1

enthought

cluster.apply()
Master
# run a function on each remote machine
>>> import os
>>> cluster.apply(os.getpid)
(123,425,947)

s0
>>> os.getpid()
123

s1
>>> os.getid()
425

s2
>>> os.getpid()
947

enthought

cluster.exec_code()
Master
# run a code fragment on each remote machine
>>> cluster.exec_code('import os; pid = os.getpid()',
...
returns = ('pid',))
(123,425,947)

s0
>>> import os
>>> os.getpid()
123

s1
>>> import os
>>> os.getpid()
425

s2
>>> import os
>>> os.getpid()
947

enthought

cluster.loop_apply()
Master
# divide task evenly (as possible) between workers
>>> import string
>>> s = ['aa','bb','cc','dd']
>>> cluster.loop_apply(string.upper, loop_var=0, args=(s,) )
('AA','BB','CC','DD')

s0
>>> x=upper('aa')
>>> y=upper('bb')
>>> (x,y)
('AA','BB')

s1
>>> x=upper('cc')
>>> (x,)
('CC',)

s2
>>> x=upper('dd')
>>> (x,)
('DD',)

enthought

Cluster Method Review

apply(function, args=(), keywords=None)
Similar to Pythons built-in apply function. Call the given function
with the specified args and keywords on all the worker machines.
Returns a list of the results received from each worker.

exec_code(code, inputs=None, returns=None)

Similar to Pythons built-in exec statement. Execute the given code
on all remote workers as if it were typed at the command line. inputs
is a dictionary of variables added to the global namespace on the
remote workers. returns is a list of variable names (as strings) that
should be returned after the code is executed. If returns contains a
single variable name, a list of values is returned by exec_code. If
returns is a sequence of variable names, exec_code returns a list of
tuples.

enthought

Cluster Method Review

loop_apply(function,loop_var,args=(),
keywords=None)
Call function with the given args and keywords. One of the
arguments or keywords is actually a sequence of arguments. This
sequence is looped over, calling function once for each value in the
sequence. loop_var indicates which variable to loop over. If an
integer, loop_var indexes the args list. If a string, it specifies a
keyword variable. The loop sequence is divided as evenly as possible
between the worker nodes and executed in parallel.

loop_code(code, loop_var, inputs=None,

returns=None)
Similar to exec_code and loop_apply. Here loop_var
indicates a variable name in the inputs dictionary that should be
looped over.

enthought

Cluster Method Review

ps(sort_by=cpu,**filters)
Display all the processes running on the remote machine much like
the ps Unix command. sort_by indicates which field to sort the
returned list. Also keywords allow the list to be filtered so that only
certain processes are displayed.

info()
Display information about each worker node including its name,
processor count and type, total and free memory, and current work
load.

enthought

Query Operations
>>> herd.cluster.info()
MACHINE
CPU
GHZ
s0
2xP3
0.5
s1
2xP3
0.5
s2
2xP3
0.5

MB TOTAL
960.0
960.0
960.0

MB FREE
930.0
41.0
221.0

>>> herd.cluster.ps(user='ej',cpu='>50')
MACHINE USER PID %CPU %MEM TOTAL MB RES MB
s0
ej
123 99.9
0.4
3.836
3.836
s1
ej
425 99.9
0.4
3.832
3.832
s2
ej
947 99.9
0.4
3.832
3.832

LOAD
0.00
1.00
0.99

CMD
python...
python...
python...

enthought

Simple FFT Benchmark

(1) STANDARD SERIAL APPROACH TO 1D FFTs
>>> b = fft(a) # a is a 2D array: 8192 x 512
(2) PARALLEL APPROACH WITH LOOP_APPLY
>>> b = cluster.loop_apply(fft,0,(a,))
(3) PARALLEL SCATTER/COMPUTE/GATHER APPROACH
>>> cluster.import_all(FFT)
# divide a row wise amongst workers
>>> cluster.row_split('a',a)
# workers calculate fft of small piece of a and stores as b.
>>> cluster.exec_code('b=fft(a)')
# gather the b values from workers back to master.
>>> b = cluster.row_gather('b')

enthought

FFT Benchmark Results

Method

CPUs

Run Time
(sec)

Speed
Up

Efficiency

(1) standard

2.97

(2) loop_apply

11.91

0.25

-400%

(3) scatter/compute/gather

13.83

0.21

-500%

Test Setup:
The array a is 8192 by 512. ffts are applied to each row
independently as is the default behavior of the FFT module.
The cluster consists of 16 dual Pentium II 450 MHz machines
connected using 100 Mbit ethernet.

enthought

FFT Benchmark Results

Method

CPUs

Run Time
(sec)

Speed
Up

Efficiency

(1) standard

2.97

(2) loop_apply

11.91

0.25

-400%

(3) scatter/compute/gather

13.83

0.21

-500%

(3) compute alone

1.49

2.00

100%

(3) compute alone

0.76

3.91

98%

(3) compute alone

16

0.24

12.38

78%

(3) compute alone

32

0.17

17.26

54%

Moral:
If data can be distributed among the machines once and then
manipulated in place, reasonable speed-ups are achieved.

enthought

Electromagnetics

EM Scattering Problem

CPUs

Run Time
(sec)

Speed
Up

Efficiency

Small Buried Sphere

64 freqs, 195 edges

32

8.19

31.40

98.0%

Land Mine
64 freqs, 1152 edges

32

285.12

31.96

99.9%

enthought

Serial vs. Parallel EM Solver

SERIAL VERSION
def serial(solver,freqs,angles):
results = []
for freq in freqs:
# single_frequency handles calculation details
res = single_frequency(solver,freq,angles)
results.append(res)
return results
PARALLEL VERSION
def parallel(solver,freqs,angles,cluster):
# make sure cluster is running
cluster.start(force_restart = 0)
# bundle arguments for loop_apply call
args = (solver,freqs,angles)
# looping handled by loop_apply
results = cluster.loop_apply(single_frequency,1,args)
return results

enthought

pyMPI

enthought

Simple MPI Program

# output is asynchronous
% mpirun -np 4 pyMPI
>>> import mpi
>>> print mpi.rank
3
0
2
1
# force synchronization
>>> mpi.synchronizedWrite(mpi.rank, \n)

0
1
2
3

enthought

Broadcasting Data
import mpi
import math
if mpi.rank == 0:
data = [sin(x) for x in range(0,10)]
else:
data = None
common_data = mpi.bcast(data)

enthought

mpi.bcast()
bcast() broadcasts a value from the
root process (default is 0) to all other
processes
bcasts arguments include the message
to send and optionally the root sender
the message argument is ignored on all
processors except the root

enthought

Scattering an Array
# You can give a little bit to everyone
import mpi
from math import sin,pi
if mpi.rank == 0:
array = [sin(x*pi/99) for x in
range(100)]
else:
array = None
# give everyone some of the array
local_array = mpi.scatter(array)

enthought

mpi.scatter()
scatter() splits an array, list, or tuple
evenly (roughly) across all processors
the function result is always a [list]
an optional argument can change the
root from rank 0
the message argument is ignored on all
processors except the root

enthought

Gathering wandering data

# Sometimes everyone has a little data to bring
# together
import mpi
import math
local_data = [sin(mpi.rank*x*pi/99) for x in range(100)]
print local_data
root_data = mpi.gather(local_data)
print root_data

enthought

mpi.gather() / mpi.allgather()
gather appends lists or tuples into a
master list on the root process
if you want it on all ranks, use
mpi.allgather() instead
every rank must call the gather()

enthought

Reductions
# You can bring data together in interesting ways
import mpi
x_cubed = mpi.rank**3
sum_x_cubed = mpi.reduce(x_cubed,mpi.SUM)

enthought

mpi.reduce() / mpi.allreduce()
The reduce (and allreduce) functions
apply an operator across data from all
participating processes
You can use predefined functions
mpi.SUM, mpi.MIN, mpi.MAX, etc

you can define your own functions too

you may optionally specify an initial
value

enthought

3D Visualization with VTK

enthought

Visualization with VTK

Visualization Toolkit from Kitware
www.kitware.com

Large C++ class library

Wrappers for Tcl, Python, and Java
Extremely powerful, but
Also complex with a steep learning curve

enthought

VTK Gallery

enthought

VTK Pipeline
PIPELINE

Pipeline view from Visualization Studio

at http://www.principiamathematica.com

OUTPUT

enthought

Cone Example
SETUP
# VTK lives in two modules
from vtk import *
# Create a renderer
renderer = vtkRenderer()
# Create render window and connect the renderer.
render_window = vtkRenderWindow()
render_window.AddRenderer(renderer)
render_window.SetSize(300,300)
# Create Tkinter based interactor and connect render window.
# The interactor handles mouse interaction.
interactor = vtkRenderWindowInteractor()
interactor.SetRenderWindow(render_window)

enthought

Cone Example (cont.)

PIPELINE
# Create cone source with 200 facets.
cone = vtkConeSource()
cone.SetResolution(200)
# Create color filter and connect its input
# to the cone's output.
color_filter = vtkElevationFilter()
color_filter.SetInput(cone.GetOutput())
color_filter.SetLowPoint(0,-.5,0)
color_filter.SetHighPoint(0,.5,0)
# map colorized cone data to graphic primitives
cone_mapper = vtkDataSetMapper()
cone_mapper.SetInput(color_filter.GetOutput())

enthought

Cone Example (cont.)

DISPLAY
# Create actor to represent our
# cone and connect it to the
# mapper
cone_actor = vtkActor()
cone_actor.SetMapper(cone_mapper)
# Assign actor to
# the renderer.
renderer.AddActor(cone_actor)
# Initialize interactor
# and start visualizing.
interactor.Initialize()
interactor.Start()

enthought

Mesh Generation
POINTS AND CELLS

3
2

0
id
0
1
2
3

points
x y z temp
0 0 0 10
1 0 0 20
0 1 0 20
0 0 1 30
triangles
id x y z
0
0 1 3
1
0 3 2
2
1 2 3
3
0 2 1

# Convert list of points to VTK structure

verts = vtkPoints()
temperature = vtkFloatArray()
for p in points:
verts.InsertNextPoint(p[0],p[1],p[2])
temperature.InsertNextValue(p[3])

# Define triangular cells from the vertex

# ids (index) and append to polygon list.
polygons = vtkCellArray()
for tri in triangles:
cell = vtkIdList()
cell.InsertNextId(tri[0])
cell.InsertNextId(tri[1])
cell.InsertNextId(tri[2])
polygons.InsertNextCell(cell)

enthought

Mesh Generation
POINTS AND CELLS
# Create a mesh from these lists
mesh = vtkPolyData()
mesh.SetPoints(verts)
mesh.SetPolys(polygons)
mesh.GetPointData().SetScalars( \
...
temperature)
# Create mapper for mesh
mapper = vtkPolyDataMapper()
mapper.SetInput(mesh)
# If range isnt set, colors are
# not plotted.
mapper.SetScalarRange( \
...
temperature.GetRange())
Code for temperature bar not shown.

enthought

VTK Demo