Oh, you're so random

Randomness and pink ponies in Codemotion Rome 2012

Transcript

1. View Slide

2. select a random element
   

   View Slide

3. select a random element
   ‘tis one is ok.
   

   View Slide

4. View Slide

5. View Slide

6. Information
   Theory
   

   View Slide

7. hard TOPIC
   Information
   Theory
   

   View Slide

8. hard TOPIC
   dumb SPEAKER
   +
   Information
   Theory
   

   View Slide

9. 0≤H(X)≤1
   where X is a discrete
   random variable
   

   View Slide

10. 0≤H(X)≤1
    where X is a discrete
    random variable
    unpredictable
    

    View Slide

11. 0≤H(X)≤1
    where X is a discrete
    random variable
    unpredictable
    always
    the same
    

    View Slide

12. View Slide

13. ask a
    question.
    

    View Slide

14. View Slide

15. bool is_random(char *bytes, size_t n)
    {
    }
    

    View Slide

16. bool is_random(char *bytes, size_t n)
    {
    }
    AGHHH
    

    View Slide

17. UNIFORM
    distribution
    

    View Slide

18. UNIFORM
    distribution
    

    View Slide

19. select a random element
    array[rand() % array.size]
    

    View Slide

20. select a random element
    array[rand() % array.size]
    UNIFORM
    distribution
    

    View Slide

21. select a random element
    array[rand() % array.size]
    UNIFORM
    distribution
    

    View Slide

22. select a random element
    array[rand() % array.size]
    UNIFORM
    distribution
    AGHHH
    

    View Slide

23. This is how you kill
    the RANDOM
    pnrg
    array
    

    View Slide

24. This is how you kill
    the RANDOM
    a
    pnrg
    array
    

    View Slide

25. This is how you kill
    the RANDOM
    a
    pnrg
    array
    

    View Slide

26. This is how you kill
    the RANDOM
    a a
    pnrg
    array
    

    View Slide

27. This is how you kill
    the RANDOM
    a a
    pnrg
    array
    

    View Slide

28. This is how you kill
    the RANDOM
    a a a
    pnrg
    array
    

    View Slide

29. This is how you kill
    the RANDOM
    a a a
    pnrg
    array
    

    View Slide

30. This is how you kill
    the RANDOM
    a a a
    pnrg
    array
    

    View Slide

31. This is how you kill
    the RANDOM
    a a a
    b
    pnrg
    array
    

    View Slide

32. This is how you kill
    the RANDOM
    a a a
    b
    pnrg
    array
    

    View Slide

33. This is how you kill
    the RANDOM
    a a a
    b b
    pnrg
    array
    

    View Slide

34. This is how you kill
    the RANDOM
    a a a
    b b
    pnrg
    array
    

    View Slide

35. This is how you kill
    the RANDOM
    a a a
    b b
    pnrg
    array
    

    View Slide

36. This is how you kill
    the RANDOM
    a a a
    b b
    pnrg
    array
    

    View Slide

37. how to FIX:
    

    View Slide

38. how to FIX:
    1. Random is hard
    

    View Slide

39. how to FIX:
    1. Random is hard
    2. Run away
    

    View Slide

40. how to FIX:
    1. Random is hard
    2. Run away
    Math.random() // between 0.0 and 1.0
    Javascript
    

    View Slide

41. how to FIX:
    1. Random is hard
    2. Run away
    

    View Slide

42. how to FIX:
    1. Random is hard
    2. Run away
    prng.rand(5..9) #=> one of [5, 6, 7, 8, 9]
    prng.rand(5...9) #=> one of [5, 6, 7, 8]
    Ruby
    

    View Slide

43. Good.
    

    View Slide

44. Good.
    (but I don’t care)
    

    View Slide

45. View Slide

46. “PRNGs
    and Hash
    functions
    are in the same
    family of algorithms”
    

    View Slide

47. View Slide

48. hash tables
    out of nowhere!
    

    View Slide

49. hash tables
    out of nowhere!
    O(1)
    

    View Slide

50. hash tables
    out of nowhere!
    O(1) uniform
    

    View Slide

51. pathological
    average data set:
    O(1)
    

    View Slide

52. pathological
    average data set:
    O(1)
    

    View Slide

53. pathological
    average data set:
    O(1) O(n)
    

    View Slide

54. ONE fix
    

    View Slide

55. ONE fix
    INT_MAX % size == 0
    

    View Slide

56. collide
    make them
    

    View Slide

57. collide
    make them
    • Brute force
    

    View Slide

58. collide
    make them
    • Brute force
    • MITM
    

    View Slide

59. collide
    make them
    • Brute force
    • MITM
    • Equivalent substrings
    

    View Slide

60. collide
    make them
    • Brute force
    • MITM
    • Equivalent substrings
    

    View Slide

61. collide
    make them
    • Brute force
    • MITM
    • Equivalent substrings
    

    View Slide

62. collide
    make them
    • Brute force
    • MITM
    • Equivalent substrings
    

    View Slide

63. collide
    make them
    • Brute force
    • MITM
    • Equivalent substrings
    

    View Slide

64. collide
    make them
    • Brute force
    • MITM
    • Equivalent substrings
    

    View Slide

65. problem
    & that’s a
    

    View Slide

66. problem
    & that’s a
    painful
    comparisons
    

    View Slide

67. problem
    & that’s a
    painful
    comparisons
    ~700ms
    responses
    

    View Slide

68. MANY fixes
    

    View Slide

69. MANY fixes
    (but only one is right)
    

    View Slide

70. MANY fixes
    (but only one is right)
    1. Limiting request size
    

    View Slide

71. this is bad and you
    should feel bad!
    MANY fixes
    (but only one is right)
    1. Limiting request size
    

    View Slide

72. MANY fixes
    (but only one is right)
    2. Changing the hash table
    

    View Slide

73. MANY fixes
    (but only one is right)
    2. Changing the hash table
    (no comment)
    

    View Slide

74. MANY fixes
    (but only one is right)
    3. Bring back the random
    

    View Slide

75. View Slide

76. “Randomness is
    too important
    to be left to
    chance”
    

    View Slide

77. Thanks.
    “Randomness is
    too important
    to be left to
    chance”
    

    View Slide