Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Ctrie Data Structure

Sponsored · Your Podcast. Everywhere. Effortlessly. Share. Educate. Inspire. Entertain. You do you. We'll handle the rest. →

Ctrie Data Structure

The description of the Ctrie data structure from PPoPP 2012.

Aleksandar Prokopec

February 28, 2012

Transcript

124 slides
  1. Concurrent Tries with Efficient Non-blocking Snapshots Aleksandar Prokopec Phil Bagwell

    Martin Odersky École Polytechnique Fédérale de Lausanne Nathan Bronson Stanford
  2. Motivation val numbers = getNumbers() // compute square roots numbers

    foreach { entry => x = entry.root n = entry.number entry.root = 0.5 * (x + n / x) if (abs(entry.root - x) < eps) numbers.remove(entry) }
  3. Immutable HAMT • updates rewrite path from root to leaf

    4 12 16 20 25 33 37 0 1 8 9 3 4 12 8 9 11 insert(11)
  4. Immutable HAMT • updates rewrite path from root to leaf

    4 12 16 20 25 33 37 0 1 8 9 3 4 12 8 9 11 insert(11) efficient updates - logk (n)
  5. Node compression 48 57 48 57 1 0 1 0

    48 57 1 0 1 0 48 57 10 BITPOP(((1 << ((hc >> lev) & 1F)) – 1) & BMP)
  6. Node compression 48 57 48 57 1 0 1 0

    48 57 1 0 1 0 48 57 10 48 57
  7. Ctrie insert 4 9 12 16 20 25 33 37

    0 1 3 48 57 17 = 0100012
  8. Ctrie insert 4 9 12 16 20 25 33 37

    0 1 3 48 57 17 = 0100012 16 17 1) allocate
  9. Ctrie insert 4 9 12 20 25 33 37 0

    1 3 48 57 17 = 0100012 16 17 2) CAS
  10. Ctrie insert 4 9 12 20 25 33 37 0

    1 3 48 57 17 = 0100012 16 17
  11. Ctrie insert 4 9 12 33 37 0 1 3

    48 57 18 = 0100102 16 17 20 25
  12. Ctrie insert 4 9 12 33 37 0 1 3

    48 57 18 = 0100102 16 17 20 25 1) allocate 16 17 18
  13. Ctrie insert 4 9 12 33 37 0 1 3

    48 57 18 = 0100102 20 25 2) CAS 16 17 18
  14. Ctrie insert 4 9 12 33 37 0 1 3

    48 57 18 = 0100102 20 25 2) CAS 16 17 18 Unless…
  15. Ctrie insert 4 9 12 33 37 0 1 3

    48 57 18 = 0100102 16 17 20 25 T1-1) allocate 16 17 18 Unless… 28 = 0111002 T1 T2
  16. Ctrie insert 4 9 12 0 1 3 18 =

    0100102 16 17 20 25 T1-1) allocate 16 17 18 Unless… 28 = 0111002 T1 T2 20 25 28 T2-1) allocate
  17. Ctrie insert 4 9 12 0 1 3 18 =

    0100102 16 17 20 25 T1-1) allocate 16 17 18 28 = 0111002 T1 T2 20 25 28 T2-2) CAS
  18. Ctrie insert 4 9 12 0 1 3 18 =

    0100102 16 17 20 25 T1-2) CAS 16 17 18 28 = 0111002 T1 T2 20 25 28 T2-2) CAS
  19. Ctrie insert 4 9 12 0 1 3 18 =

    0100102 16 17 20 25 16 17 18 28 = 0111002 T1 T2 20 25 28 Lost insert!
  20. Ctrie insert – 2nd attempt 4 9 12 0 1

    3 16 17 20 25 Solution: I-nodes
  21. Ctrie insert – 2nd attempt 4 9 12 0 1

    3 16 17 20 25 18 = 0100102 28 = 0111002 T1 T2
  22. Ctrie insert – 2nd attempt 4 9 12 0 1

    3 16 17 T1 T2 20 25 18 = 0100102 28 = 0111002 16 17 18 20 25 28 T2-1) allocate T1-1) allocate
  23. Ctrie insert – 2nd attempt 4 9 12 0 1

    3 16 17 T1 T2 20 25 16 17 18 20 25 28 T2-2) CAS T1-2) CAS
  24. Ctrie insert – 2nd attempt 4 9 12 0 1

    3 16 17 18 20 25 28 Idea: once added to the Ctrie, I-nodes remain present.
  25. Ctrie insert – 2nd attempt 4 9 12 0 1

    3 16 17 18 20 25 28 Remove operation supported as well - details in the paper.
  26. Ctrie size 4 9 12 0 1 3 16 17

    18 20 25 28 size = 0
  27. Ctrie size 4 9 12 0 1 3 16 17

    18 20 25 28 size = 0
  28. Ctrie size 4 9 12 0 1 3 16 17

    18 20 25 28 size = 0
  29. Ctrie size 4 9 12 0 1 3 16 17

    18 20 25 28 size = 0
  30. Ctrie size 4 9 12 0 1 3 16 17

    18 20 25 28 size = 1
  31. Ctrie size 4 9 12 0 1 3 16 17

    18 20 25 28 size = 2
  32. Ctrie size 4 9 12 0 1 3 16 17

    18 20 25 28 size = 3
  33. Ctrie size 4 9 12 0 1 3 16 17

    18 20 25 28 size = 5
  34. Ctrie size 4 9 12 0 1 3 16 17

    18 20 25 28 size = 5 actual size = 12
  35. Ctrie size 4 9 12 0 1 3 16 17

    18 20 25 28 size = 5 0 1 actual size = 12
  36. Ctrie size 4 9 12 0 1 3 16 17

    18 20 25 28 size = 5 0 1 CAS actual size = 11
  37. Ctrie size 4 9 12 16 17 18 20 25

    28 size = 5 0 1 actual size = 11
  38. Ctrie size 4 9 12 16 17 18 20 25

    28 size = 6 0 1 actual size = 11
  39. Ctrie size 4 9 12 16 17 18 20 25

    28 size = 6 0 1 actual size = 11 19
  40. Ctrie size 4 9 12 16 17 18 20 25

    28 size = 6 0 1 actual size = 11 16 17 18 19
  41. Ctrie size 4 9 12 16 17 18 20 25

    28 size = 6 0 1 actual size = 12 16 17 18 19 CAS
  42. Ctrie size 4 9 12 20 25 28 size =

    6 0 1 actual size = 12 16 17 18 19
  43. Ctrie size 4 9 12 20 25 28 size =

    6 0 1 actual size = 12 16 17 18 19
  44. Ctrie size 4 9 12 20 25 28 size =

    7 0 1 actual size = 9 16 17 18 19
  45. Ctrie size 4 9 12 20 25 28 size =

    8 0 1 actual size = 12 16 17 18 19
  46. Ctrie size 4 9 12 20 25 28 size =

    9 0 1 actual size = 12 16 17 18 19
  47. Ctrie size 4 9 12 20 25 28 size =

    10 0 1 actual size = 12 16 17 18 19
  48. Ctrie size 4 9 12 20 25 28 size =

    11 0 1 actual size = 12 16 17 18 19
  49. Ctrie size 4 9 12 20 25 28 size =

    12 0 1 actual size = 12 16 17 18 19
  50. Ctrie size 4 9 12 20 25 28 size =

    13 0 1 actual size = 12 16 17 18 19
  51. Ctrie size 4 9 12 20 25 28 size =

    13 0 1 actual size = 12 16 17 18 19 But the size was never 13!
  52. Global state information 4 9 12 20 25 28 0

    1 16 17 18 19 • size • find • filter • iterator
  53. Global state information 4 9 12 20 25 28 0

    1 16 17 18 19 • size • find • filter • iterator  snapshot
  54. Snapshot using locks 4 9 12 20 25 28 0

    1 16 17 18 19 • copy expensive
  55. Snapshot using locks 4 9 12 20 25 28 0

    1 16 17 18 19 • copy expensive • not lock-free
  56. Snapshot using locks 4 9 12 20 25 28 0

    1 16 17 18 19 • copy expensive • not lock-free • can insert or remove remain lock-free? 0 1 2 CAS
  57. Snapshot using locks 4 9 12 20 25 28 0

    1 16 17 18 19 • copy expensive • not lock-free • can insert or remove remain lock-free? 0 1 2 CAS
  58. Snapshot using logs 4 9 12 20 25 28 0

    1 16 17 18 19 • keep a linked list of previous values in each I-node
  59. Snapshot using logs 4 9 12 20 25 28 0

    1 16 17 18 19 0 1 2 • keep a linked list of previous values in each I-node
  60. Snapshot using logs 4 9 12 20 25 28 0

    1 16 17 18 19 • keep a linked list of previous values in each I-node • when is it safe to delete old entries? 0 1 2
  61. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 root
  62. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 snapshot! root
  63. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 snapshot! #2 root 1) create new I-node at #2
  64. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 snapshot! #2 root 2) set snapshot snapshot #1
  65. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 snapshot! #2 root 3) CAS root to new I-node snapshot #1
  66. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 subsequent insert #2 root snapshot #1 2
  67. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 subsequent insert #2 root snapshot #1 2 generation #2 - ok!
  68. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 subsequent insert #2 root snapshot #1 2 generation #1 not ok, too old!
  69. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 subsequent insert #2 root 1) create updated node at #2 snapshot #1 2 #2 #2
  70. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 subsequent insert #2 root 2) CAS to the updated node snapshot #1 2 #2 #2
  71. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 subsequent insert #2 root snapshot #1 2 #2 #2 #1 too old!
  72. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 subsequent insert #2 root snapshot #1 2 #2 #2 4 9 12 #2 1) create updated node at #2
  73. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 subsequent insert #2 root snapshot #1 2 #2 #2 4 9 12 #2 2) CAS
  74. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 subsequent insert #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 finally, create a new leaf and CAS
  75. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 another insert #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 3
  76. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 another insert #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 0 1 2 3
  77. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 But... this won't really work... why? #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 0 1 2 3
  78. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 0 1 2 3 T2: remove 19 16 17 18
  79. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 0 1 2 3 T2: remove 19 16 17 18 CAS
  80. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 0 1 2 3 T2: remove 19 16 17 18 CAS How to fail this last CAS?
  81. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 0 1 2 3 T2: remove 19 16 17 18 DCAS How to fail this last CAS? DCAS
  82. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 0 1 2 3 T2: remove 19 16 17 18 How to fail this last CAS? DCAS - software based DCAS
  83. Snapshot using immutability 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 0 1 2 3 T2: remove 19 16 17 18 How to fail this last CAS? DCAS - software based ...creates intermediate objects DCAS
  84. GCAS - generation-compare-and-swap 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 3 T2: remove 19 16 17 18 prev 1) set prev field
  85. GCAS - generation-compare-and-swap 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 3 T2: remove 19 16 17 18 prev 2) CAS
  86. GCAS - generation-compare-and-swap 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 3 T2: remove 19 16 17 18 prev 3) read root generation
  87. GCAS - generation-compare-and-swap 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 3 16 17 18 prev 4) if root generation changed CAS prev to FailedNode(prev) FN
  88. GCAS - generation-compare-and-swap 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 3 16 17 18 prev 4) if root generation changed CAS prev to FailedNode(prev) FN
  89. GCAS - generation-compare-and-swap 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 3 16 17 18 prev 5) CAS to previous value FN
  90. GCAS - generation-compare-and-swap 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 3 16 17 18 prev 4) if root generation unchanged CAS prev to null
  91. GCAS - generation-compare-and-swap 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 3 16 17 18 4) if root generation unchanged CAS prev to null
  92. GCAS - generation-compare-and-swap 4 9 12 20 25 28 0

    1 16 17 18 19 #1 #1 #1 #1 #1 #2 root snapshot #1 #2 #2 4 9 12 #2 0 1 2 3 1) Replace all CAS with GCAS 2) Replace all READ with GCAS_READ (which checks if prev field is null)
  93. Snapshot-based size def size = { val sz = 0

    val it = iterator while (it.hasNext) sz += 1 sz }
  94. Snapshot-based size def size = { val sz = 0

    val it = iterator while (it.hasNext) sz += 1 sz } Above is O(n). But, by caching size in nodes - amortized O(logk n)! (see source code)
  95. Snapshot-based atomic clear def clear() = { val or =

    READ(root) val nr = new INode(new Gen) if (!CAS(root, or, nr)) clear() } (roughly)
  96. Conclusion • snapshots are linearizable and lock-free • snapshots take

    constant time • snapshots are horizontally scalable • snapshots add a non-significant overhead to the algorithm if they aren't used • the approach may be applicable to tree-based lock-free data-structures in general (intuition)

More Decks by Aleksandar Prokopec

Other Decks in Programming