1. Scientific
Computing on
JRuby
github.com/prasunanand

2. Objective
● A Scientific library is memory intensive and speed counts.How to use
JRuby effectively to create a great tool/gem.
● A General Purpose GPU library for Ruby that can be used by industry in
production and academia for research.

3. ● Ruby Science Foundation
● SciRuby has been trying to push Ruby for scientific computing.
● Popular Rubygems:
1. NMatrix
2. Daru
3. Mixed_models

4. NMatrix
NMatrix is SciRuby’s numerical matrix core, implementing dense matrices as
well as two types of sparse (linked-list-based and Yale/CSR).
It currently relies on ATLAS/CBLAS/CLAPACK and standard LAPACK for several
of its linear algebra operations.

6. Daru

7. Mixed_models

8. Nyaplot

9. Why nya?

11. Contributors wanted
● IRC #sciruby
● Slack-channel #sciruby
● Google-group #sciruby

12. Known for performance JRuby is 10 times faster than CRuby.
With truffle it’s around 40 times faster than CRuby.

13. Say hello

14. NMatrix for JRuby
● Not a unified interface for Sciruby gems: MDArray.
● MDArray is a great gem for Linear Algebra.
● However, every gem that used NMatrix as dependency needed to be
reimplemented with MDArray.
● Hence, putting in effort for optimization.

15. NMatrix for JRuby
● Parallelism=> No Global Interpreter Lock as in case of MRI
● Easy Deployment(Warbler gem)

16. How NMatrix works
● N-Dimensional
● 2-Dimensional NMatrix

17. N-dimensional NMatrix
N-dimensional matrices are stored as a one-dimensional Array.

19. Elementwise Operation
● Iterate through the elements
● Access the array; do the operation, return it
● [:add, :subtract, :sin, :gamma]

20. Determinants and Factoriztion
● Two dimensional matrix operations
● In NMatrix-MRI, BLAS-III and LAPACK routines are implemented using
their respective libraries
● NMatrix-JRuby depends on Java functions.

21. Mixed models
● After NMAtrix for doubles was ready, I tested it with mixed_models.

22. Challenges
● Autoboxing and Multiple data type
● Minimise copying of data
● Handling large array

23. Autoboxing
● :float64 => double only
● Strict dtypes => creating data type in Java: not guessing
● Errors => that can’t be reproduced :P
[ 0. 11, 0.05, 0.34, 0.14 ] + [ 0. 21,0.05, 0.14, 0.14 ] = [ 0, 0, 0, 0]
([ 0. 11, 0.05, 0.34, 0.14 ] + 5) + ([ 0. 21, 0.05, 0.14, 0.14 ] + 5) - 10 =
[ 0.32, 0.1, 0.48, 0.28]

24. Minimise copying of data
● Make sure you make copies of data

25. Handling large arrays
● Array Size
● Accessing elements
● Chaining to java method
● Speed and Memory Required

26. Ruby Code
index =0
puts Benchmark.measure{
(0...15000).each do |i|
(0...15000).each do |j|
c[i][j] = b[i][j]
index+=1
end
end
}
#67.790000 0.070000 67.860000 ( 65.126546)
#RAM consumed => 5.4GB
b = Java::double[15_000,15_000].new
c = Java::double[15_000,15_000].new
index=0
puts Benchmark.measure{
(0...15000).each do |i|
(0...15000).each do |j|
b[i][j] = index
index+=1
end
end
}
#43.260000 3.250000 46.510000 ( 39.606356)

28. Java Code
public class MatrixGenerator{
public static void test2(){
for (int index=0, i=0; i < row ; i++){
for (int j=0; j < col; j++){
c[i][j]= b[i][j];
index++;
}
}
}
puts Benchmark.measure{MatrixGenerator.test2}
#0.034000 0.001000 00.034000 ( 00.03300)
#RAM consumed => 300MB
public class MatrixGenerator{
public static void test1(){
double[][] b = new double[15000][15000];
double[][] c = new double[15000][15000];
for (int index=0, i=0; i < row ; i++){
for (int j=0; j < col; j++){
b[i][j]= index;
index++;
}
}
}
puts Benchmark.measure{MatrixGenerator.test1}
#0.032000 0.001000 00.032000 ( 00.03100)

29. Results
Improves:
● 1000 times the speed
● 10times the memory

30. Benchmarking NMatrix functionalities

31. System Specifications
● CPU: AMD FX8350 0ctacore 4.2GHz
● RAM: 16GB

32. Addition

33. Subtraction

34. Gamma

35. Matrix Multiplication

36. Determinant

37. Factorization

38. Benchmark conclusion
● NMatrix-JRuby is incredibly faster for N-dimensional matrices when
elementwise operations are concerned.
● NMatrix-MRI is faster for 2-dimensional matrix when calculating matrix
multiplication, determinant calculation and factorization.

39. Improvements
● Make NMatrix-JRuby faster than NMatrix-MRI using BLAS level-3 and
LAPACK routines.
● How?
● Why not JBlas?

40. Future Work
● Add support for complex dtype.
● Convert NMatrix-JRuby Enumerators to Java code.
● Add sparse support.

41. Am I done?

42. Nope!

46. Enter GPU

47. A General-Purpose GPU library
● Combine the beauty of Ruby with transparent GPU processing
● This will work both on client computers and on servers that make use of
TESLA's and Intel Xeon Phi solutions.
● Developer activity and support for the current projects is mixed at best,
and they are tough to use as they involve writing kernels and require a lot
of effort to be put in buffer/RAM optimisation.

48. ArrayFire-rb
● Wraps ArrayFire library

49. Using ArrayFire

50. MRI
● C extension
● Architecture is inspired by NMatrix and NArray
● The C++ function is placed in a namespace (e.g., namespace af { }) or is
declared static if possible. The C function receives the prefix af_, e.g.,
af_multiply() (this function also happens to be static).
● C macros are capitalized and generally have the prefix AF_, as with
AF_DTYPE().
● C functions (and macros, for consistency) are placed within extern "C" { }
blocks to turn off C++ mangling.

51. JRuby
● The approach is same as NMatrix JRuby.
● Java Native Interface( JNI )
● Work on ArrayFire-Java

52. Benchmarking ArrayFire

53. System Specification
CPU: AMD FX Octacore 4.2GHz
RAM: 16GB
GPU: Nvidia GTX 750Ti
GPU RAM : 4GB DDR5

54. Matrix Addition

55. Matrix Multiplication

56. Matrix Determinant

57. Factorization

58. Transparency
● Integrate with Narray
● Integrate with NMatrix
● Integrate with Rails

59. Applications
● Endless possibilities ;)
● Bioinformatics
● Integrate Tensorflow
● Image Processing
● Computational Fluid Dynamics

60. Conclusion

61. Useful Links
● https://github.com/sciruby/nmatrix
● https://github.com/arrayfire/arrayfire-rb
● https://github.com/prasunanand/arrayfire-rb/tree/temp

62. Acknowlegements
1. Pjotr Prins
2. Charles Nutter
3. John Woods
4. Alexej Gossmann
5. Sameer Deshmukh
6. Pradeep Garigipati

63. Thank You
Github: prasunanand
Twitter: @prasun_anand
Blog: prasunanand.com