Distributed, Eventually Consistent Computations

Strange Loop 2015

Transcript

1. Distributed
   Eventually Consistent
   Computations
   Christopher Meiklejohn // @cmeik
   Strange Loop 2015, September 24 - 26th, 2015
   1
   

   View Slide

2. RA
   RB
   

   View Slide

3. RA
   RB
   1
   set(1)
   

   View Slide

4. RA
   RB
   1
   set(1)
   3
   2
   set(2)
   set(3)
   

   View Slide

5. RA
   RB
   1
   set(1)
   3
   2
   set(2)
   set(3)
   ?
   ?
   

   View Slide

6. Synchronization
   • To enforce an order

   Makes programming easier
   6
   

   View Slide

7. Synchronization
   • To enforce an order

   Makes programming easier
   • Eliminate accidental nondeterminism

   Prevent race conditions
   6
   

   View Slide

8. Synchronization
   • To enforce an order

   Makes programming easier
   • Eliminate accidental nondeterminism

   Prevent race conditions
   • Techniques

   Locks, mutexes, semaphores, monitors, etc.
   6
   

   View Slide

9. Difficult Cases
   • “Internet of Things”

   Low power, limited memory and connectivity
   7
   

   View Slide

10. Difficult Cases
    • “Internet of Things”

    Low power, limited memory and connectivity
    • Mobile Gaming

    Offline operation with replicated, shared state
    7
    

    View Slide

11. Zero Synchronization
    • We want to have our cake and eat it too.
    8
    

    View Slide

12. Zero Synchronization
    • We want to have our cake and eat it too.
    • “Zero synchronization” as a starting point

    Add synchronization only where it is necessary
    8
    

    View Slide

13. Zero Synchronization
    • We want to have our cake and eat it too.
    • “Zero synchronization” as a starting point

    Add synchronization only where it is necessary
    • Explore the limits

    Can we design a model of computation where it’s
    easy to write coordination-free computations that
    are free from concurrency anomalies?
    8
    

    View Slide

14. 9
    Strong
    Weak
    Strong
    (Data Center)
    Weak
    (Open Internet)
    Sharing
    
    (Application Property)
    Coupling (Infrastructure Property)
    

    View Slide

15. 9
    Strong
    Weak
    Strong
    (Data Center)
    Weak
    (Open Internet)
    Sharing
    
    (Application Property)
    Coupling (Infrastructure Property)
    MapReduce SETI@Home
    

    View Slide

16. 9
    Strong
    Weak
    Strong
    (Data Center)
    Weak
    (Open Internet)
    Sharing
    
    (Application Property)
    Coupling (Infrastructure Property)
    Twitter
    MapReduce SETI@Home
    

    View Slide

17. 9
    Strong
    Weak
    Strong
    (Data Center)
    Weak
    (Open Internet)
    Sharing
    
    (Application Property)
    Coupling (Infrastructure Property)
    Twitter
    MapReduce SETI@Home
    Synchronization-Free
    Computations
    

    View Slide

18. The Fundamentals of
    Distributed Computation
    10
    

    View Slide

19. Distributed Computation
    • Distributed to concurrent programming

    Consistency and now partial failure
    11
    

    View Slide

20. Distributed Computation
    • Distributed to concurrent programming

    Consistency and now partial failure
    • Enforcing the “single system” illusion
    11
    

    View Slide

21. Distributed Computation
    • Distributed to concurrent programming

    Consistency and now partial failure
    • Enforcing the “single system” illusion
    • Consistency model as a contract
    11
    

    View Slide

22. Distributed Computation
    • Distributed to concurrent programming

    Consistency and now partial failure
    • Enforcing the “single system” illusion
    • Consistency model as a contract
    • Enforced through synchronization
    11
    

    View Slide

23. Distributed Computation
    • Distributed to concurrent programming

    Consistency and now partial failure
    • Enforcing the “single system” illusion
    • Consistency model as a contract
    • Enforced through synchronization
    • Consistency model

    We can think of a consistency model as analogous
    to a programming paradigm
    11
    

    View Slide

24. Why Is Synchronization
    Undesirable?
    12
    

    View Slide

25. The Avatars of Time
    • The problem of time

    Handling physical time in applications is difficult
    13
    

    View Slide

26. The Avatars of Time
    • The problem of time

    Handling physical time in applications is difficult
    • Time has three major avatars in computing
    13
    

    View Slide

27. The Avatars of Time
    • The problem of time

    Handling physical time in applications is difficult
    • Time has three major avatars in computing
    • Mutable state in sequential systems
    13
    

    View Slide

28. The Avatars of Time
    • The problem of time

    Handling physical time in applications is difficult
    • Time has three major avatars in computing
    • Mutable state in sequential systems
    • Nondeterminism in concurrent systems
    13
    

    View Slide

29. The Avatars of Time
    • The problem of time

    Handling physical time in applications is difficult
    • Time has three major avatars in computing
    • Mutable state in sequential systems
    • Nondeterminism in concurrent systems
    ★ Network latency in distributed systems
    13
    

    View Slide

30. The Avatars of Time
    • The problem of time

    Handling physical time in applications is difficult
    • Time has three major avatars in computing
    • Mutable state in sequential systems
    • Nondeterminism in concurrent systems
    ★ Network latency in distributed systems
    • Unavoidable

    Time is how users interact with programs
    13
    

    View Slide

31. Parable of the Car
    • “Car driving on a highway”
    14
    

    View Slide

32. Parable of the Car
    • “Car driving on a highway”
    • Friction needed for the car to grip the road
    14
    

    View Slide

33. Parable of the Car
    • “Car driving on a highway”
    • Friction needed for the car to grip the road
    • Motors want as little friction as possible
    14
    

    View Slide

34. Parable of the Car
    • “Car driving on a highway”
    • Friction needed for the car to grip the road
    • Motors want as little friction as possible
    • Time is like friction

    We can not completely eliminate friction, but aim to
    reduce it as much as possible
    14
    

    View Slide

35. Physical
    Time
    (real world)
    Parable of the Car
    15
    • Time only at the interface

    The interface is a small part of the system, and this is where we
    introduction friction
    

    View Slide

36. Physical
    Time
    (real world)
    Parable of the Car
    15
    • Time only at the interface

    The interface is a small part of the system, and this is where we
    introduction friction
    • Physical time-free execution

    Internally, avoid synchronization as much as possible and virtualize
    notion of time
    

    View Slide

37. No Time In The Box?
    16
    

    View Slide

38. No Time In The Box?
    16
    You program the box.
    

    View Slide

39. Weak Synchronization
    • Can we achieve anything without synchronization?

    Not really.
    17
    

    View Slide

40. Weak Synchronization
    • Can we achieve anything without synchronization?

    Not really.
    • Strong Eventual Consistency (SEC)

    “Replicas that deliver the same updates have equivalent state”
    17
    

    View Slide

41. Weak Synchronization
    • Can we achieve anything without synchronization?

    Not really.
    • Strong Eventual Consistency (SEC)

    “Replicas that deliver the same updates have equivalent state”
    • Primary requirement

    Eventual replica-to-replica communication
    17
    

    View Slide

42. Weak Synchronization
    • Can we achieve anything without synchronization?

    Not really.
    • Strong Eventual Consistency (SEC)

    “Replicas that deliver the same updates have equivalent state”
    • Primary requirement

    Eventual replica-to-replica communication
    • Order insensitive! (Commutativity)
    17
    

    View Slide

43. Weak Synchronization
    • Can we achieve anything without synchronization?

    Not really.
    • Strong Eventual Consistency (SEC)

    “Replicas that deliver the same updates have equivalent state”
    • Primary requirement

    Eventual replica-to-replica communication
    • Order insensitive! (Commutativity)
    • Duplicate insensitive! (Idempotent)
    17
    

    View Slide

44. RA
    RB
    

    View Slide

45. RA
    RB
    1
    set(1)
    

    View Slide

46. RA
    RB
    1
    set(1)
    3
    2
    set(2)
    set(3)
    

    View Slide

47. RA
    RB
    1
    3
    2
    3
    3
    set(1) set(2)
    set(3)
    max(2,3)
    max(2,3)
    

    View Slide

48. How can we succeed with
    Strong Eventual Consistency?
    22
    

    View Slide

49. Programming SEC
    1. Eliminate accidental nondeterminism

    (ex. deterministic, modeling non-monotonic operations monotonically)
    23
    

    View Slide

50. Programming SEC
    1. Eliminate accidental nondeterminism

    (ex. deterministic, modeling non-monotonic operations monotonically)
    2. Retain the properties of functional programming

    (ex. confluence, referential transparency over composition)
    23
    

    View Slide

51. Programming SEC
    1. Eliminate accidental nondeterminism

    (ex. deterministic, modeling non-monotonic operations monotonically)
    2. Retain the properties of functional programming

    (ex. confluence, referential transparency over composition)
    3. Distributed, and fault-tolerant runtime

    (ex. replication)
    23
    

    View Slide

52. Programming SEC
    1. Eliminate accidental nondeterminism

    (ex. deterministic, modeling non-monotonic operations monotonically)
    
    2. Retain the properties of functional programming

    (ex. confluence, referential transparency over composition)
    3. Distributed, and fault-tolerant runtime

    (ex. replication)
    24
    

    View Slide

53. Convergent Objects

    Conflict-Free 

    Replicated Data Types
    25
    SSS 2011
    

    View Slide

54. Conflict-Free 

    Replicated Data Types
    • Many types exist with different properties

    Sets, counters, registers, flags, maps, graphs
    26
    

    View Slide

55. Conflict-Free 

    Replicated Data Types
    • Many types exist with different properties

    Sets, counters, registers, flags, maps, graphs
    • Strong Eventual Consistency

    Instances satisfy SEC property per-object
    26
    

    View Slide

56. RA
    RB
    RC
    

    View Slide

57. RA
    RB
    RC
    {1}
    (1, {a}, {})
    add(1)
    

    View Slide

58. RA
    RB
    RC
    {1}
    (1, {a}, {})
    add(1)
    {1}
    (1, {b}, {})
    add(1)
    

    View Slide

59. RA
    RB
    RC
    {1}
    (1, {a}, {})
    add(1)
    {1}
    (1, {b}, {})
    add(1)
    {}
    (1, {b}, {b})
    remove(1)
    

    View Slide

60. RA
    RB
    RC
    {1}
    (1, {a}, {})
    add(1)
    {1}
    (1, {b}, {})
    add(1)
    {}
    (1, {b}, {b})
    remove(1)
    {1}
    {1}
    {1}
    (1, {a, b}, {b})
    (1, {a, b}, {b})
    (1, {a, b}, {b})
    

    View Slide

61. Programming SEC
    1. Eliminate accidental nondeterminism

    (ex. deterministic, modeling non-monotonic operations monotonically)
    
    2. Retain the properties of functional programming

    (ex. confluence, referential transparency over composition)
    3. Distributed, and fault-tolerant runtime

    (ex. replication)
    32
    

    View Slide

62. Convergent Programs
    Lattice Processing
    33
    PPDP 2015
    

    View Slide

63. Lattice Processing (Lasp)
    • Distributed, deterministic dataflow

    Distributed, deterministic dataflow programming
    model for “eventually consistent” computations
    34
    

    View Slide

64. Lattice Processing (Lasp)
    • Distributed, deterministic dataflow

    Distributed, deterministic dataflow programming
    model for “eventually consistent” computations
    • Convergent data structures

    Primary data abstraction is the CRDT
    34
    

    View Slide

65. Lattice Processing (Lasp)
    • Distributed, deterministic dataflow

    Distributed, deterministic dataflow programming
    model for “eventually consistent” computations
    • Convergent data structures

    Primary data abstraction is the CRDT
    • Enables composition

    Provides functional composition of CRDTs that
    preserves the SEC property
    34
    

    View Slide

66. 35
    %% Create initial set.
    S1 = declare(set),
    %% Add elements to initial set and update.
    update(S1, {add, [1,2,3]}),
    %% Create second set.
    S2 = declare(set),
    %% Apply map operation between S1 and S2.
    map(S1, fun(X) -> X * 2 end, S2).
    

    View Slide

67. 36
    %% Create initial set.
    S1 = declare(set),
    %% Add elements to initial set and update.
    update(S1, {add, [1,2,3]}),
    %% Create second set.
    S2 = declare(set),
    %% Apply map operation between S1 and S2.
    map(S1, fun(X) -> X * 2 end, S2).
    

    View Slide

68. 37
    %% Create initial set.
    S1 = declare(set),
    %% Add elements to initial set and update.
    update(S1, {add, [1,2,3]}),
    %% Create second set.
    S2 = declare(set),
    %% Apply map operation between S1 and S2.
    map(S1, fun(X) -> X * 2 end, S2).
    

    View Slide

69. 38
    %% Create initial set.
    S1 = declare(set),
    %% Add elements to initial set and update.
    update(S1, {add, [1,2,3]}),
    %% Create second set.
    S2 = declare(set),
    %% Apply map operation between S1 and S2.
    map(S1, fun(X) -> X * 2 end, S2).
    

    View Slide

70. 39
    %% Create initial set.
    S1 = declare(set),
    %% Add elements to initial set and update.
    update(S1, {add, [1,2,3]}),
    %% Create second set.
    S2 = declare(set),
    %% Apply map operation between S1 and S2.
    map(S1, fun(X) -> X * 2 end, S2).
    

    View Slide

71. Lattice Processing (Lasp)
    • Functional and set-theoretic operations on sets

    Product, intersection, union, filter, map, fold
    40
    

    View Slide

72. Lattice Processing (Lasp)
    • Functional and set-theoretic operations on sets

    Product, intersection, union, filter, map, fold
    • Metadata computation

    Performs transformation on the internal metadata of
    CRDTs allowing creation of “composed” CRDTs
    40
    

    View Slide

73. Lasp Processes
    • Replicas as monotonic streams

    Each replica of a CRDT produces a monotonic
    stream of states
    41
    

    View Slide

74. Lasp Processes
    • Replicas as monotonic streams

    Each replica of a CRDT produces a monotonic
    stream of states
    • Monotonic processes

    Read from one or more input replica streams and
    produce a single output replica stream
    41
    

    View Slide

75. Lasp Processes
    • Replicas as monotonic streams

    Each replica of a CRDT produces a monotonic
    stream of states
    • Monotonic processes

    Read from one or more input replica streams and
    produce a single output replica stream
    • Inflationary reads

    Read operation ensures that we only read
    inflationary updates to replicas
    41
    

    View Slide

76. Lattice Processing
    Monotonic Streams
    42
    

    View Slide

77. RA
    {}
    C1
    {}
    C2
    {}
    43
    

    View Slide

78. RA
    {}
    (1, {a}, {})
    C1
    (1, {a}, {})
    {}
    C2
    {}
    44
    

    View Slide

79. RA
    {}
    (1, {a}, {})
    (1, {a, b}, {})
    C1
    (1, {a}, {})
    {}
    C2
    (1, {b}, {})
    {}
    45
    

    View Slide

80. RA
    {}
    (1, {a}, {})
    (1, {a, b}, {})
    (1, {a, b}, {a})
    C1
    (1, {a}, {})
    {}
    (1, {a}, {a})
    C2
    (1, {b}, {})
    {}
    46
    

    View Slide

81. RA
    {}
    (1, {a}, {})
    (1, {a, b}, {})
    (1, {a, b}, {a})
    C1
    (1, {a}, {})
    {}
    (1, {a}, {a})
    C2
    (1, {b}, {})
    {}
    (1, {a, b}, {a})
    (1, {a, b}, {a}) (1, {a, b}, {a})
    47
    

    View Slide

82. RA
    {}
    (1, {a}, {})
    (1, {a, b}, {})
    (1, {a, b}, {a})
    C1
    (1, {a}, {})
    {}
    (1, {a}, {a})
    C2
    (1, {b}, {})
    {}
    (1, {a, b}, {a})
    (1, {a, b}, {a}) (1, {a, b}, {a})
    48
    

    View Slide

83. Lattice Processing
    Monotonic Processes
    49
    

    View Slide

84. RA
    {}
    P1 F(RA)
    {}
    50
    

    View Slide

85. RA
    {}
    P1 F(RA)
    {}
    strict_read({})
    51
    

    View Slide

86. RA
    {}
    P1 F(RA)
    {}
    strict_read({})
    (1, {a}, {})
    52
    

    View Slide

87. RA
    {}
    P1 F(RA)
    {}
    strict_read({})
    (1, {a}, {}) F((1, {a}, {}))
    53
    

    View Slide

88. RA
    {}
    P1 F(RA)
    {}
    strict_read({})
    (1, {a}, {}) F((1, {a}, {})) (2, {a}, {})
    54
    

    View Slide

89. RA
    {}
    P1 F(RA)
    {}
    strict_read({})
    (1, {a}, {}) F((1, {a}, {})) (2, {a}, {})
    strict_read((1, {a}, {})
    (1, {a}, {})
    55
    

    View Slide

90. RA
    {}
    P1 F(RA)
    {}
    strict_read({})
    (1, {a}, {}) F((1, {a}, {})) (2, {a}, {})
    strict_read((1, {a}, {})
    (1, {a}, {})
    (1, {a}, {a}) F((1, {a}, {a})) (2, {a}, {a})
    strict_read((1, {a}, {a})
    (1, {a}, {a})
    56
    

    View Slide

91. RA
    {}
    P1 F(RA)
    {}
    strict_read({})
    (1, {a}, {})
    (1, {a}, {})
    (1, {a}, {a}) F((1, {a}, {a})) (2, {a}, {a})
    strict_read((1, {a}, {a})
    (1, {a}, {a})
    57
    

    View Slide

92. Programming SEC
    1. Eliminate accidental nondeterminism

    (ex. deterministic, modeling non-monotonic operations monotonically)
    
    2. Retain the properties of functional programming

    (ex. confluence, referential transparency over composition)
    3. Distributed, and fault-tolerant runtime

    (ex. replication)
    58
    

    View Slide

93. Distributed Runtime
    Selective Hearing
    59
    W-PSDS 2015
    

    View Slide

94. Selective Hearing
    • Epidemic broadcast based runtime system

    Provide a runtime system that can scale to large numbers of
    nodes, that is resilient to failures and provides efficient execution
    60
    

    View Slide

95. Selective Hearing
    • Epidemic broadcast based runtime system

    Provide a runtime system that can scale to large numbers of
    nodes, that is resilient to failures and provides efficient execution
    • Well-matched to Lattice Processing (Lasp)
    60
    

    View Slide

96. Selective Hearing
    • Epidemic broadcast based runtime system

    Provide a runtime system that can scale to large numbers of
    nodes, that is resilient to failures and provides efficient execution
    • Well-matched to Lattice Processing (Lasp)
    • Epidemic broadcast mechanisms provide weak ordering but
    are resilient and efficient
    60
    

    View Slide

97. Selective Hearing
    • Epidemic broadcast based runtime system

    Provide a runtime system that can scale to large numbers of
    nodes, that is resilient to failures and provides efficient execution
    • Well-matched to Lattice Processing (Lasp)
    • Epidemic broadcast mechanisms provide weak ordering but
    are resilient and efficient
    • Lasp’s programming model is tolerant to message re-ordering,
    disconnections, and node failures
    60
    

    View Slide

98. Selective Hearing
    • Epidemic broadcast based runtime system

    Provide a runtime system that can scale to large numbers of
    nodes, that is resilient to failures and provides efficient execution
    • Well-matched to Lattice Processing (Lasp)
    • Epidemic broadcast mechanisms provide weak ordering but
    are resilient and efficient
    • Lasp’s programming model is tolerant to message re-ordering,
    disconnections, and node failures
    • “Selective Receive”

    Nodes selectively receive and process messages based on
    interest.
    60
    

    View Slide

99. Programming SEC
    1. Eliminate accidental nondeterminism

    (ex. deterministic, modeling non-monotonic operations monotonically)
    
    2. Retain the properties of functional programming

    (ex. confluence, referential transparency over composition)
    3. Distributed, and fault-tolerant runtime

    (ex. replication)
    61
    

    View Slide

100. What can we build?
     Advertisement Counter
     62
     

     View Slide

101. Advertisement Counter
     • Mobile game platform selling advertisement space

     Advertisements are paid according to a minimum
     number of impressions
     63
     

     View Slide

102. Advertisement Counter
     • Mobile game platform selling advertisement space

     Advertisements are paid according to a minimum
     number of impressions
     • Clients will go offline

     Clients have limited connectivity and the system still
     needs to make progress while clients are offline
     63
     

     View Slide

103. Ads
     Rovio Ad
     Counter 1
     Rovio Ad
     Counter 2
     Riot Ad
     Counter 1
     Riot Ad
     Counter 2
     Contracts
     Ads
     Contracts
     Ads
     With
     Contracts
     Riot Ads
     Rovio
     Ads
     Filter
     Product
     Read
     50,000
     Remove
     Increment
     Read
     Union
     Lasp Operation
     User-Maintained CRDT
     Lasp-Maintained CRDT
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Client Side, Single Copy at Client
     64
     

     View Slide

104. Ads
     Rovio Ad
     Counter 1
     Rovio Ad
     Counter 2
     Riot Ad
     Counter 1
     Riot Ad
     Counter 2
     Contracts
     Ads
     Contracts
     Riot Ads
     Rovio
     Ads
     Product
     Read
     50,000
     Remove
     Increment
     Union
     65
     Ads
     Rovio Ad
     Counter 1
     Rovio Ad
     Counter 2
     Riot Ad
     Counter 1
     Riot Ad
     Counter 2
     Contracts
     Ads
     Contracts
     Ads
     With
     Contracts
     Riot Ads
     Rovio
     Ads
     Filter
     Product
     Read
     50,000
     Remove
     Increment
     Read
     Union
     Lasp Operation
     User-Maintained CRDT
     Lasp-Maintained CRDT
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Client Side, Single Copy at Client
     

     View Slide

105. Ads
     Rovio Ad
     Counter 1
     Rovio Ad
     Counter 2
     Riot Ad
     Counter 1
     Riot Ad
     Counter 2
     Contracts
     Ads
     Contracts
     Ads
     With
     Contracts
     Riot Ads
     Rovio
     Ads
     Filter
     Product
     Read
     50,000
     Remove
     Increment
     Read
     Union
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Client
     66
     Ads
     Rovio Ad
     Counter 1
     Rovio Ad
     Counter 2
     Riot Ad
     Counter 1
     Riot Ad
     Counter 2
     Contracts
     Ads
     Contracts
     Ads
     With
     Contracts
     Riot Ads
     Rovio
     Ads
     Filter
     Product
     Read
     50,000
     Remove
     Increment
     Read
     Union
     Lasp Operation
     User-Maintained CRDT
     Lasp-Maintained CRDT
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Client Side, Single Copy at Client
     

     View Slide

106. Ads
     ovio Ad
     ounter 1
     ovio Ad
     ounter 2
     iot Ad
     ounter 1
     iot Ad
     ounter 2
     Contracts
     Ads
     Contracts
     Ads
     With
     Contracts
     Riot Ads
     Rovio
     Ads
     Filter
     Product
     Read
     50,000
     Remove
     Increment
     Read
     Union
     Rovio Ad
     Counter
     1
     Ro
     C
     Rovio Ad
     Counter
     1
     Ro
     C
     Rovio Ad
     Counter
     1
     Ro
     C
     Rovio Ad
     Counter
     1
     Ro
     C
     Client Side, Sing
     67
     Ads
     Rovio Ad
     Counter 1
     Rovio Ad
     Counter 2
     Riot Ad
     Counter 1
     Riot Ad
     Counter 2
     Contracts
     Ads
     Contracts
     Ads
     With
     Contracts
     Riot Ads
     Rovio
     Ads
     Filter
     Product
     Read
     50,000
     Remove
     Increment
     Read
     Union
     Lasp Operation
     User-Maintained CRDT
     Lasp-Maintained CRDT
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Client Side, Single Copy at Client
     

     View Slide

107. Ads
     Contracts
     Ads
     Contracts
     Ads
     With
     Contracts
     Riot Ads
     Rovio
     Ads
     Filter
     Product
     move Read
     Union
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Client Side, Single Copy at Client
     68
     Ads
     Rovio Ad
     Counter 1
     Rovio Ad
     Counter 2
     Riot Ad
     Counter 1
     Riot Ad
     Counter 2
     Contracts
     Ads
     Contracts
     Ads
     With
     Contracts
     Riot Ads
     Rovio
     Ads
     Filter
     Product
     Read
     50,000
     Remove
     Increment
     Read
     Union
     Lasp Operation
     User-Maintained CRDT
     Lasp-Maintained CRDT
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Client Side, Single Copy at Client
     

     View Slide

108. Ads
     Contracts
     Ads
     Contracts
     Ads
     With
     Contracts
     Filter
     Product
     Read
     Union
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Client Side, Single Copy at Client
     69
     Ads
     Rovio Ad
     Counter 1
     Rovio Ad
     Counter 2
     Riot Ad
     Counter 1
     Riot Ad
     Counter 2
     Contracts
     Ads
     Contracts
     Ads
     With
     Contracts
     Riot Ads
     Rovio
     Ads
     Filter
     Product
     Read
     50,000
     Remove
     Increment
     Read
     Union
     Lasp Operation
     User-Maintained CRDT
     Lasp-Maintained CRDT
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Client Side, Single Copy at Client
     

     View Slide

109. Ads
     Rovio Ad
     Counter 1
     Rovio Ad
     Counter 2
     Riot Ad
     Counter 1
     Riot Ad
     Counter 2
     Contracts
     Ads
     Contracts
     Ads
     With
     Contracts
     Riot Ads
     Rovio
     Ads
     Filter
     Product
     Read
     50,000
     Remove
     Increment
     Read
     Union
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Client Side, Single Copy at Client
     70
     Ads
     Rovio Ad
     Counter 1
     Rovio Ad
     Counter 2
     Riot Ad
     Counter 1
     Riot Ad
     Counter 2
     Contracts
     Ads
     Contracts
     Ads
     With
     Contracts
     Riot Ads
     Rovio
     Ads
     Filter
     Product
     Read
     50,000
     Remove
     Increment
     Read
     Union
     Lasp Operation
     User-Maintained CRDT
     Lasp-Maintained CRDT
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Client Side, Single Copy at Client
     

     View Slide

110. Ads
     Rovio Ad
     Counter 1
     Rovio Ad
     Counter 2
     Riot Ad
     Counter 1
     Riot Ad
     Counter 2
     Contracts
     Ads
     Contracts
     Riot Ads
     Rovio
     Ads
     Fil
     Product
     Read
     50,000
     Remove
     Increment
     Union
     71
     Ads
     Rovio Ad
     Counter 1
     Rovio Ad
     Counter 2
     Riot Ad
     Counter 1
     Riot Ad
     Counter 2
     Contracts
     Ads
     Contracts
     Ads
     With
     Contracts
     Riot Ads
     Rovio
     Ads
     Filter
     Product
     Read
     50,000
     Remove
     Increment
     Read
     Union
     Lasp Operation
     User-Maintained CRDT
     Lasp-Maintained CRDT
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Client Side, Single Copy at Client
     

     View Slide

111. Ads
     Rovio Ad
     Counter 1
     Rovio Ad
     Counter 2
     Riot Ad
     Counter 1
     Riot Ad
     Counter 2
     Contracts
     Ads
     Contracts
     Ads
     With
     Contracts
     Riot Ads
     Rovio
     Ads
     Filter
     Product
     Read
     50,000
     Remove
     Increment
     Read
     Union
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Client Side, Single Copy at Client
     72
     Ads
     Rovio Ad
     Counter 1
     Rovio Ad
     Counter 2
     Riot Ad
     Counter 1
     Riot Ad
     Counter 2
     Contracts
     Ads
     Contracts
     Ads
     With
     Contracts
     Riot Ads
     Rovio
     Ads
     Filter
     Product
     Read
     50,000
     Remove
     Increment
     Read
     Union
     Lasp Operation
     User-Maintained CRDT
     Lasp-Maintained CRDT
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Rovio Ad
     Counter
     1
     Rovio Ad
     Counter
     2
     Riot Ad
     Counter
     1
     Client Side, Single Copy at Client
     

     View Slide

112. Advertisement Counter
     • Completely monotonic

     Disabling advertisements and contracts are all
     modeled through monotonic state growth
     73
     

     View Slide

113. Advertisement Counter
     • Completely monotonic

     Disabling advertisements and contracts are all
     modeled through monotonic state growth
     • Arbitrary distribution

     Use of convergent data structures allows
     computational graph to be arbitrarily distributed
     73
     

     View Slide

114. Advertisement Counter
     • Completely monotonic

     Disabling advertisements and contracts are all
     modeled through monotonic state growth
     • Arbitrary distribution

     Use of convergent data structures allows
     computational graph to be arbitrarily distributed
     • Divergence

     Divergence is a factor of synchronization period
     73
     

     View Slide

115. We’ve build up from zero
     synchronization
     74
     

     View Slide

116. We’ve build up from zero
     synchronization
     74
     Instead of working to
     remove synchronization
     

     View Slide

117. What’s next?
     75
     

     View Slide

118. What’s Next?
     • Reduce metadata

     Metadata can grow very large
     76
     

     View Slide

119. What’s Next?
     • Reduce metadata

     Metadata can grow very large
     • Provide a more general ‘fold’ operation

     Currently very restrictive, how general can it be?
     76
     

     View Slide

120. What’s Next?
     • Reduce metadata

     Metadata can grow very large
     • Provide a more general ‘fold’ operation

     Currently very restrictive, how general can it be?
     • Auto-placement

     What can the dataflow graph tell us about
     placement?
     76
     

     View Slide

121. What have we learned?
     77
     

     View Slide

122. Programming SEC
     1. Eliminate accidental nondeterminism

     Convergent Objects: Conflict-Free Replicated Data Types
     78
     

     View Slide

123. Programming SEC
     1. Eliminate accidental nondeterminism

     Convergent Objects: Conflict-Free Replicated Data Types
     2. Retain the properties of functional programming

     Convergent Programs: Lattice Processing
     78
     

     View Slide

124. Programming SEC
     1. Eliminate accidental nondeterminism

     Convergent Objects: Conflict-Free Replicated Data Types
     2. Retain the properties of functional programming

     Convergent Programs: Lattice Processing
     3. Distributed, and fault-tolerant runtime

     Distributed Runtime: Selective Hearing
     78
     

     View Slide

125. Key Points
     • Synchronization is expensive

     Locking and other synchronization mechanisms limit
     scalability to the critical section
     79
     

     View Slide

126. Key Points
     • Synchronization is expensive

     Locking and other synchronization mechanisms limit
     scalability to the critical section
     • Synchronization is sometimes not possible

     Mobile and “Internet of Things” applications make
     synchronization for replicated state impractical
     79
     

     View Slide

127. Key Points
     • Synchronization is expensive

     Locking and other synchronization mechanisms limit
     scalability to the critical section
     • Synchronization is sometimes not possible

     Mobile and “Internet of Things” applications make
     synchronization for replicated state impractical
     • Apply synchronization only where required

     Global invariants, atomic visibility, etc.
     79
     

     View Slide

128. How do I learn more?
     80
     

     View Slide

129. Publications
     • “Lasp: A Language for Distributed, Coordination-Free Programming” 

     ACM SIGPLAN PPDP 2015
     81
     

     View Slide

130. Publications
     • “Lasp: A Language for Distributed, Coordination-Free Programming” 

     ACM SIGPLAN PPDP 2015
     • “Selective Hearing: An Approach to Distributed, Eventually Consistent Edge
     Computation”

     IEEE W-PSDS 2015
     81
     

     View Slide

131. Publications
     • “Lasp: A Language for Distributed, Coordination-Free Programming” 

     ACM SIGPLAN PPDP 2015
     • “Selective Hearing: An Approach to Distributed, Eventually Consistent Edge
     Computation”

     IEEE W-PSDS 2015
     • “The Implementation and Use of a Generic Dataflow Behaviour in Erlang”

     ACM SIGPLAN Erlang Workshop ’15
     81
     

     View Slide

132. Publications
     • “Lasp: A Language for Distributed, Coordination-Free Programming” 

     ACM SIGPLAN PPDP 2015
     • “Selective Hearing: An Approach to Distributed, Eventually Consistent Edge
     Computation”

     IEEE W-PSDS 2015
     • “The Implementation and Use of a Generic Dataflow Behaviour in Erlang”

     ACM SIGPLAN Erlang Workshop ’15
     • “Lasp: A Language for Distributed, Eventually Consistent Computations with
     CRDTs"

     PaPoC 2015
     81
     

     View Slide

133. 82
     Lasp Release 0.0.1
     http://lasp-lang.org
     

     View Slide

134. SyncFree is a European research project taking place
     for 3 years, staring October 2013, and is funded by
     the European Union, grant agreement n° 609551.
     83
     

     View Slide

135. Thanks!
     84
     Christopher Meiklejohn
     @cmeik
     

     View Slide