1. C. Varela; Adapted with permission from S. Haridi and P. Van Roy1 Oz, Declarative Concurrency, and Active Objects (VRH 4, 7.8) Carlos Varela RPI Adapted with permission from: Seif Haridi KTH Peter Van Roy UCL

2. C. Varela; Adapted with permission from S. Haridi and P. Van Roy2 Overview of concurrent programming Four basic approaches to programming are: –Sequential programming (no concurrency) –Declarative concurrency (streams in a functional language, Oz) –Message passing with active objects (Erlang, SALSA) –Atomic actions on shared state (Java) The atomic action approach is the most difficult, yet it is the one you will probably be most exposed to! But, if you have the choice, which approach to use? –Use the simplest approach that does the job: sequential if that is ok, else declarative concurrency if there is no observable nondeterminism, else message passing if you can get away with it.

3. C. Varela; Adapted with permission from S. Haridi and P. Van Roy3 Concurrency Some programs are best written as a set of activities that run independently (concurrent programs) Concurrency is essential for interaction with the external environment –Examples includes GUI (Graphical User Interfaces), operating systems, web services –Also programs that are written independently but interact only when needed (client-server, peer-to-peer applications) First, we will cover declarative concurrency, programs with no observable nondeterminism, the result is a function –Independent procedures that execute at their own pace and may communicate through shared dataflow variables Then, we will cover message passing, programs consisting of components with encapsulated state communicating asynchronously

4. C. Varela; Adapted with permission from S. Haridi and P. Van Roy4 Overview of declarative concurrency Programming with threads –The model is augmented with threads –Programming techniques: stream communication, order-determining concurrency, concurrent composition Lazy execution –demand-driven computations, lazy streams Soft real-time programming

5. C. Varela; Adapted with permission from S. Haridi and P. Van Roy5 The sequential model w = a z = person(age: y) x y = 42 u Single-assignment store Semantic Stack Statements are executed sequentially from a single semantic stack

6. C. Varela; Adapted with permission from S. Haridi and P. Van Roy6 The concurrent model w = a z = person(age: y) x y = 42 u Single-assignment store Semantic Stack 1 Semantic Stack N Multiple semantic stacks (threads)

7. C. Varela; Adapted with permission from S. Haridi and P. Van Roy7 Concurrent declarative model  s  ::= skip empty statement |  x  =  y  variable-variable binding |  x  =  v  variable-value binding |  s 1   s 2  sequential composition |local  x  in  s 1  end declaration | proc {  x   y 1  …  y n  }  s 1  end procedure introduction |if  x  then  s 1  else  s 2  end conditional |{  x   y 1  …  y n  } procedure application |case  x  of  pattern  then  s 1  else  s 2  end pattern matching | thread  s 1  end thread creation The following defines the syntax of a statement,  s  denotes a statement

8. C. Varela; Adapted with permission from S. Haridi and P. Van Roy8 The concurrent model Single-assignment store ST thread  s 1  end,E Top of Stack, Thread i

9. C. Varela; Adapted with permission from S. Haridi and P. Van Roy9 The concurrent model Single-assignment store ST Top of Stack, Thread i  s 1 ,E

10. C. Varela; Adapted with permission from S. Haridi and P. Van Roy10 Basic concepts The model allows multiple statements to execute ”at the same time”. Imagine that these threads really execute in parallel, each has its own processor, but share the same memory Reading and writing different variables can be done simultaneously by different threads, as well as reading the same variable Writing the same variable is done sequentially The above view is in fact equivalent to an interleaving execution: a totally ordered sequence of computation steps, where threads take turn doing one or more steps in sequence

11. C. Varela; Adapted with permission from S. Haridi and P. Van Roy11 Causal order In a sequential program all execution states are totally ordered In a concurrent program all execution states of a given thread are totally ordered The execution state of the concurrent program as a whole is partially ordered

12. C. Varela; Adapted with permission from S. Haridi and P. Van Roy12 Total order In a sequential program all execution states are totally ordered computation step sequential execution

13. C. Varela; Adapted with permission from S. Haridi and P. Van Roy13 Causal order in the declarative model In a concurrent program all execution states of a given thread are totally ordered The execution state of the concurrent program is partially ordered computation step thread T1 thread T2 thread T3 fork a thread

14. C. Varela; Adapted with permission from S. Haridi and P. Van Roy14 Causal order in the declarative model computation step thread T1 thread T2 thread T3 fork a thread bind a dataflow variable synchonize on a dataflow variable x y

15. C. Varela; Adapted with permission from S. Haridi and P. Van Roy15 Nondeterminism An execution is nondeterministic if there is a computation step in which there is a choice what to do next Nondeterminism appears naturally when there is concurrent access to shared state

16. C. Varela; Adapted with permission from S. Haridi and P. Van Roy16 Example of nondeterminism time Thread 1 x = 1 x y = 5 store time Thread 2 x = 3 The thread that binds x first will continue, the other thread will raise an exception

17. C. Varela; Adapted with permission from S. Haridi and P. Van Roy17 Nondeterminism An execution is nondeterministic if there is a computation step in which there is a choice what to do next Nondeterminism appears naturally when there is concurrent access to shared state In the concurrent declarative model when there is only one binder for each dataflow variable, the nondeterminism is not observable on the store (i.e. the store develops to the same final results) This means for correctness we can ignore the concurrency

18. C. Varela; Adapted with permission from S. Haridi and P. Van Roy18 Scheduling The choice of which thread to execute next and for how long is done by a part of the system called the scheduler A thread is runnable if its next statement to execute is not blocked on a dataflow variable, otherwise the thread is suspended A scheduler is fair if it does not starve a runnable thread, i.e. all runnable threads eventually execute Fair scheduling makes it easy to reason about programs and program composition Otherwise some correct program (in isolation) may never get processing time when composed with other programs

19. C. Varela; Adapted with permission from S. Haridi and P. Van Roy19 The semantics In the sequential model we had: (ST,  ) ST is a stack of semantic statements  is the single assignment store In the concurrent model we have: (MST,  ) MST is a (multi)set of stacks of semantic statements  is the single assignment store

20. C. Varela; Adapted with permission from S. Haridi and P. Van Roy20 The initial execution state ({ [ (  s ,  ) ] },  ) statement stack multiset store

21. C. Varela; Adapted with permission from S. Haridi and P. Van Roy21 Execution (the scheduler) At each step, one runnable semantic stack is selected from MST (the multiset of stacks), call it ST, s.t. MST = ST  MST’ Assume the current store is , one computation step is done that transforms ST to ST’ and  to  ’ The total computation state is transformed from (MST,  ) to (ST’  MST’,  ’) Which stack is selected, and how many steps are taken is the task of the scheduler, a good scheduler should be fair, i.e., each runnable ’thread’ will eventually be selected The computation stops when there are no runnable stacks

22. C. Varela; Adapted with permission from S. Haridi and P. Van Roy22 Example of runnable threads proc {Loop P N} if N > 0 then {P} {Loop P N-1} else skip end end thread {Loop proc {$} {Show 1} end 1000} end thread {Loop proc {$} {Show 2} end 1000} end This program will interleave the execution of two threads, one printing 1, and the other printing 2 We assume a fair scheduler

23. C. Varela; Adapted with permission from S. Haridi and P. Van Roy23 Dataflow computation Threads suspend on data unavailability in dataflow variables The {Delay X} primitive makes the thread suspends for X milliseconds, after that, the thread is runnable declare X {Browse X} local Y in thread {Delay 1000} Y = 10*10 end X = Y + 100*100 end

24. C. Varela; Adapted with permission from S. Haridi and P. Van Roy24 Illustrating Dataflow computation Enter incrementally the values of X0 to X3 When X0 is bound the thread will compute Y0=X0+1, and will suspend again until X1 is bound declare X0 X1 X2 X3 {Browse [X0 X1 X2 X3]} thread Y0 Y1 Y2 Y3 in {Browse [Y0 Y1 Y2 Y3]} Y0 = X0 + 1 Y1 = X1 + Y0 Y2 = X2 + Y1 Y3 = X3 + Y2 {Browse completed} end

25. C. Varela; Adapted with permission from S. Haridi and P. Van Roy25 Concurrent Map fun {Map Xs F} case Xs of nil then nil [] X|Xr then thread {F X} end|{Map Xr F} end This will fork a thread for each individual element in the input list Each thread will run only if both the element X and the procedure F is known

26. C. Varela; Adapted with permission from S. Haridi and P. Van Roy26 Concurrent Map Function fun {Map Xs F} case Xs of nil then nil [] X|Xr then thread {F X} end |{Map Xr F} end end What this looks like in the kernel language: proc {Map Xs F Rs} case Xs of nil then Rs = nil [] X|Xr then R Rr in Rs = R |Rr thread R = {F X} end {Map Xr F Rr} end end

27. C. Varela; Adapted with permission from S. Haridi and P. Van Roy27 How does it work? If we enter the following statements: declare F X Y Z {Browse thread {Map X F} end } A thread executing Map is created. It will suspend immediately in the case-statement because X is unbound. If we thereafter enter the following statements: X = 1|2|Y fun {F X} X*X end The main thread will traverse the list creating two threads for the first two arguments of the list,

28. C. Varela; Adapted with permission from S. Haridi and P. Van Roy28 How does it work? The main thread will traverse the list creating two threads for the first two arguments of the list thread {F 1} end, and thread {F 2} end, Y = 3|Z Z = nil will complete the computation of the main thread and the newly created thread thread {F 3} end, resulting in the final list [1 4 9].

29. C. Varela; Adapted with permission from S. Haridi and P. Van Roy29 Cheap concurrency and dataflow Declarative programs can be easily made concurrent Just use the thread statement where concurrency is needed fun {Fib X} if X=<2 then 1 else thread {Fib X-1} end + {Fib X-2} end

30. C. Varela; Adapted with permission from S. Haridi and P. Van Roy30 Understanding why fun {Fib X} if X=<2 then 1 else F1 F2 in F1 = thread {Fib X-1} end F2 = {Fib X-2} F1 + F2 end end Dataflow dependency

31. C. Varela; Adapted with permission from S. Haridi and P. Van Roy31 Execution of {Fib 6} F6 F5 F4F2 F3 F2 F1 F2 F3 F2 F1 F4 F1F3 F2 Fork a thread Synchronize on result Running thread

32. C. Varela; Adapted with permission from S. Haridi and P. Van Roy32 Fib

33. C. Varela; Adapted with permission from S. Haridi and P. Van Roy33 Streams A stream is a sequence of messages A stream is First-In First-Out (FIFO) channel The producer augments the stream with new messages, and the consumer reads the messages, one by one. x 5 x 4 x 3 x 2 x 1 producerconsumer

34. C. Varela; Adapted with permission from S. Haridi and P. Van Roy34 Stream Communication I The data-flow property of Oz easily enables writing threads that communicate through streams in a producer- consumer pattern. A stream is a list that is created incrementally by one thread (the producer) and subsequently consumed by one or more threads (the consumers). The consumers consume the same elements of the stream.

35. C. Varela; Adapted with permission from S. Haridi and P. Van Roy35 Stream Communication II Producer, that produces incrementally the elements Transducer(s), that transforms the elements of the stream Consumer, that accumulates the results producer transducer consumer thread 1thread 2thread 3 thread N

36. C. Varela; Adapted with permission from S. Haridi and P. Van Roy36 Program patterns The producer, transducers, and the consumer can, in general, be described by certain program patterns We show the various patterns

37. C. Varela; Adapted with permission from S. Haridi and P. Van Roy37 Producer fun {Producer State} if { More State} then X = { Produce State} in X | {Producer { Transform State}} else nil end end The definition of More, Produce, and Transform is problem dependent State could be multiple arguments The above definition is not a complete program!

38. C. Varela; Adapted with permission from S. Haridi and P. Van Roy38 Example Producer fun {Generate N Limit} if N=<Limit then N | {Generate N+1 Limit} else nil end end The State is the two arguments N and Limit The predicate More is the condition N=<Limit The Transform function (N,Limit)  (N+1,Limit) fun {Producer State} if { More State} then X = { Produce State} in X | {Producer { Transform State}} else nil end end

39. C. Varela; Adapted with permission from S. Haridi and P. Van Roy39 Consumer Pattern fun {Consumer State InStream} case InStream of nil then { Final State} [] X | RestInStream then NextState = { Consume X State} in {Consumer NextState RestInStream} end Final and Consume are problem dependent The consumer suspends until InStream is either a cons or a nil

40. C. Varela; Adapted with permission from S. Haridi and P. Van Roy40 Example Consumer fun {Sum A Xs} case Xs of X|Xr then {Sum A+X Xr} [] nil then A end The State is A Final is just the identity function on State Consume takes X and State  X + State fun {Consumer State InStream} case InStream of nil then { Final State} [] X | RestInStream then NextState = { Consume X State} in {Consumer NextState RestInStream} end

41. C. Varela; Adapted with permission from S. Haridi and P. Van Roy41 Transducer Pattern 1 fun {Transducer State Instream} case InStream of nil then nil [] X | RestInStream then NextState#TX = { Transform X State} TX | {Transducer NextState RestInStream} end A transducer keeps its state in State, receives messages on InStream and sends messages on OutStream

42. C. Varela; Adapted with permission from S. Haridi and P. Van Roy42 Transducer Pattern 2 fun {Transducer State Instream} case InStream of nil then nil [] X | RestInStream then if { Test X#State} then NextState#TX = { Transform X State} TX | {Consumer NextState RestInStream} else {Consumer NextState RestInStream} end end A transducer keeps its state in State, receives messages on InStream and sends messages on OutStream

43. C. Varela; Adapted with permission from S. Haridi and P. Van Roy43 Example Transducer fun {Filter Xs F} case Xs of nil then nil [] X|Xr then if {F X} then X|{Filter Xr F} else {Filter Xr F} end end Generate Filter IsOdd 6 5 4 3 2 15 3 1 Filter is a transducer that takes an Instream and incremently produces an Outstream that satisfies the predicate F local Xs Ys in thread Xs = {Generate 1 100} end thread Ys = {Filter Xs IsOdd} end thread {Browse Ys } end end

44. C. Varela; Adapted with permission from S. Haridi and P. Van Roy44 Larger Example: The sieve of Eratosthenes Produces prime numbers It takes a stream 2...N, peals off 2 from the rest of the stream Delivers the rest to the next sieve Sieve FilterSieve Xs Xr X Ys Zs X|Zs

45. C. Varela; Adapted with permission from S. Haridi and P. Van Roy45 Sieve fun {Sieve Xs} case Xs of nil then nil [] X|Xr then Ys in thread Ys = {Filter Xr fun {$ Y} Y mod X \= 0 end} end X | {Sieve Ys} end The program forks a filter thread on each sieve call

46. C. Varela; Adapted with permission from S. Haridi and P. Van Roy46 Example Call local Xs Ys in thread Xs = {Generate 2 100000} end thread Ys = {Sieve Xs} end thread for Y in Ys do {Show Y} end end end Filter 3Sieve Filter 2 Filter 5 7 | 11 |...

47. C. Varela; Adapted with permission from S. Haridi and P. Van Roy47 Limitation of eager stream processing The producer might be much faster than the consumer This will produce a large intermediate stream that requires potentially unbounded memory storage x 5 x 4 x 3 x 2 x 1 producerconsumer

48. C. Varela; Adapted with permission from S. Haridi and P. Van Roy48 Solutions There are three alternatives: 1.Play with the speed of the different threads, i.e. play with the scheduler to make the producer slower 2.Create a bounded buffer, say of size N, so that the producer waits automatically when the buffer is full 3.Use demand-driven approach, where the consumer activates the producer when it needs a new element (lazy evaluation) The last two approaches introduce the notion of flow-control between concurrent activities (very common)

49. C. Varela; Adapted with permission from S. Haridi and P. Van Roy49 Time In concurrent computation one would like to handle time proc {Time.delay T} – The running thread suspends for T milliseconds proc {Time.alarm T U} – Immediately creates its own thread, and binds U to unit after T milliseconds

50. C. Varela; Adapted with permission from S. Haridi and P. Van Roy50 Example local proc {Ping N} for I in 1..N do {Delay 500} {Browse ping} end {Browse 'ping terminate'} end proc {Pong N} for I in 1..N do {Delay 600} {Browse pong} end {Browse 'pong terminate'} end in.... end local.... in {Browse 'game started'} thread {Ping 1000} end thread {Pong 1000} end end

51. C. Varela; Adapted with permission from S. Haridi and P. Van Roy51 Thread Priority and Real Time Try to run the program using the following statement: –{Consumer thread {Producer 5000000} end} Switch on the panel and observe the memory behavior of the program. You will quickly notice that this program does not behave well. The reason has to do with the asynchronous message passing. If the producer sends messages i.e. create new elements in the stream, in a faster rate than the consumer can consume, increasingly more buffering will be needed until the system starts to break down. One possible solution is to control experimentally the rate of thread execution so that the consumers get a larger time-slice than the producers do.

52. C. Varela; Adapted with permission from S. Haridi and P. Van Roy52 Priorities There are three priority levels:  high,  medium, and  low (the default) A priority level determines how often a runnable thread is allocated a time slice. In Oz, a high priority thread cannot starve a low priority one. Priority determines only how large piece of the processor-cake a thread can get. Each thread has a unique name. To get the name of the current thread the procedure Thread.this/1 is called. Having a reference to a thread, by using its name, enables operations on threads such as:  terminating a thread, or  raising an exception in a thread. Thread operations are defined the standard module Thread.

53. C. Varela; Adapted with permission from S. Haridi and P. Van Roy53 Thread priority and thread control fun {Thread.state T} % returns thread state proc{Thread.injectException T E} % exception E injected into thread fun {Thread.this} % returns 1st class reference to thread proc{Thread.setPriority T P} % P is high, medium or low proc{Thread.setThisPriority P} % as above on current thread fun{Property.get priorities} % get priority ratios proc{Property.put priorities(high:H medium:M)}

54. C. Varela; Adapted with permission from S. Haridi and P. Van Roy54 Thread Priorities Oz has three priority levels. The system procedure {Property.put priorities p(medium:Y high:X)}  Sets the processor-time ratio to X:1 between high-priority threads and medium- priority thread.  It also sets the processor-time ratio to Y:1 between medium-priority threads and low-priority threads. X and Y are integers. –Example: {Property.put priorities p(high:10 medium:10)} Now let us make our producer-consumer program work. We give the producer low priority, and the consumer high. We also set the priority ratios to 10:1 and 10:1.

55. C. Varela; Adapted with permission from S. Haridi and P. Van Roy55 The program local L in {Property.put priorities p(high:10 medium:10)} thread {Thread.setThisPriority low} L = {Producer 5000000} end thread {Thread.setThisPriority high} {Consumer L} end end

56. C. Varela; Adapted with permission from S. Haridi and P. Van Roy56 Concurrent control abstraction We have seen how threads are forked by ’thread... end’ A natural question to ask is: how can we join threads? fork join threads

57. C. Varela; Adapted with permission from S. Haridi and P. Van Roy57 Termination detection This is a special case of detecting termination of multiple threads, and making another thread wait on that event. The general scheme is quite easy because of dataflow variables: thread  S1  X1 = unit end thread  S2  X2 = X1 end... thread  Sn  X n = X n-1 end {Wait Xn} % Continue main thread When all threads terminate the variables X 1 … X N will be merged together labeling a single box that contains the value unit. {Wait X N } suspends the main thread until X N is bound.

58. C. Varela; Adapted with permission from S. Haridi and P. Van Roy58 Concurrent Composition conc S 1 [] S 2 [] … [] S n end {Conc [ proc{$} S1 end proc{$} S2 end... proc{$} Sn end] } Takes a single argument that is a list of nullary procedures. When it is executed, the procedures are forked concurrently. The next statement is executed only when all procedures in the list terminate.

59. C. Varela; Adapted with permission from S. Haridi and P. Van Roy59 Conc local proc {Conc1 Ps I O} case Ps of P|Pr then M in thread {P} M = I end {Conc1 Pr M O} [] nil then O = I end end in proc {Conc Ps} X in {Conc1 Ps unit X} {Wait X} end end This abstraction takes a list of zero-argument procedures and terminate after all these threads have terminated

60. C. Varela; Adapted with permission from S. Haridi and P. Van Roy60 Example local proc {Ping N} for I in 1..N do {Delay 500} {Browse ping} end {Browse 'ping terminate'} end proc {Pong N} for I in 1..N do {Delay 600} {Browse pong} end {Browse 'pong terminate'} end in.... end local.... in {Browse 'game started'} { Conc [ proc {$} {Ping 1000} end proc {$} {Pong 1000} end ]} {Browse ’game terminated’} end

61. C. Varela; Adapted with permission from S. Haridi and P. Van Roy61 Futures A future is a read-only capability of a single-assignment variable. For example to create a future of the variable X we perform the operation !! to create a future Y : Y = !!X A thread trying to use the value of a future, e.g. using Y, will suspend until the variable of the future, e.g. X, gets bound. One way to execute a procedure lazily, i.e. in a demand-driven manner, is to use the operation {ByNeed +P ?F}. ByNeed takes a zero-argument function P, and returns a future F. When a thread tries to access the value of F, the function {P} is called, and its result is bound to F. This allows us to perform demand-driven computations in a straightforward manner.

62. C. Varela; Adapted with permission from S. Haridi and P. Van Roy62 Example declare Y {ByNeed fun {$} 1 end Y} {Browse Y} we will observe that Y becomes a future, i.e. we will see Y in the Browser. If we try to access the value of Y, it will get bound to 1. One way to access Y is by perform the operation {Wait Y} which triggers the producing procedure.

63. C. Varela; Adapted with permission from S. Haridi and P. Van Roy63 Why not always use declarative concurrency? The concurrent declarative model is much simpler –Programs give the same results as if they were sequential, but they give the results incrementally (assuming a single binder per dataflow variable) Why is this model so easy? –Because dataflow variables can be bound to only one value. A thread that shares a variable with another thread does not have to worry that the other thread will change the binding. So why not stick with this model? –In many cases, we can stick with this model –But not always. For example, two clients that communicate with one server cannot be programmed in this model. Why not? Because there is an observable nondeterminism. The concurrent declarative model is deterministic. If the program we write has an observable nondeterminism, then we cannot use the model.

64. C. Varela; Adapted with permission from S. Haridi and P. Van Roy64 Concurrent stateful model  s  ::= skip empty statement |  x  =  y  variable-variable binding |  x  =  v  variable-value binding |  s 1   s 2  sequential composition |local  x  in  s 1  end declaration | proc {  x   y 1  …  y n  }  s 1  end procedure creation |if  x  then  s 1  else  s 2  end conditional |{  x   y 1  …  y n  } procedure application |case  x  of  pattern  then  s 1  else  s 2  end pattern matching |{NewName  x  } name creation | thread  s  end thread creation |{ByNeed  x   y  } trigger creation |try  s 1  catch  x  then  s 2  end exception context |raise  x  end raise exception | {NewCell  x   y  } cell creation | {Exchange  x   y   z  } cell exchange

65. C. Varela; Adapted with permission from S. Haridi and P. Van Roy65 Concurrency and state are tough when used together Execution consists of multiple threads, all executing independently and all using shared cells A thread’s execution is a sequence of Access and Assign operations (or Exchange operations) Because of interleaving semantics, execution happens as if there was one global order of operations Assume two threads and each thread does k operations. Then the total number of possible interleavings is This is exponential in k. One can program by reasoning on all possible interleavings, but this is extremely hard. What do we do? 2k k ()

66. C. Varela; Adapted with permission from S. Haridi and P. Van Roy66 Programming with concurrency and state Programming with concurrency and state is largely a matter of reducing the number of interleavings, so that we can reason about programs in a simpler way. There are two basic approaches: message passing and atomic actions. Message passing with active objects: Programs consist of threads that send asynchronous messages to each other. Each thread only receives a message when it is ready, which reduces the number of interleavings. Atomic actions on shared state: Programs consist of passive objects that are called by threads. We build large atomic actions (e.g., with locks, monitors, or transactions) to reduce the number of interleavings.

67. C. Varela; Adapted with permission from S. Haridi and P. Van Roy67 When to use each approach Message passing: useful for multi-agent applications, i.e., programs that consist of autonomous entities (« agents », « actors » or « active objects ») that communicate with each other. Atomic actions: useful for data-centered applications, i.e., programs that consist of a large repository of data (« database » or « shared state ») that is accessed and updated concurrently. Both approaches can be used together in the same application, for different parts

68. C. Varela; Adapted with permission from S. Haridi and P. Van Roy68 Ports and cells We have seen cells, the basic unit of encapsulated state, as a primitive concept underlying stateful and object-oriented programming. Cells are like variables in imperative languages. Cells are the natural concept for programming with shared state There is another way to add state to a language, which we call a port. A port is an asynchronous FIFO communication channel. Ports are a natural concept for programming with active objects Cells and ports are duals of each other –Each can be implemented with the other, so they are equal in expressiveness –Each is more natural in some circumstances –They are equivalent because each allows many-to-one communication (cell shared by threads, port shared by threads)

69. C. Varela; Adapted with permission from S. Haridi and P. Van Roy69 Ports A port is an ADT with two operations: –{NewPort S P}: create a new port P with a new stream S. The stream is a list with unbound tail, used to model the FIFO nature of the communications channel. –{Send P X}: send message X on port P. The message is appended to the stream S and can be read by threads reading S. Example: declare P S in {NewPort S P} {Browse S} thread{Send P 1}end thread{Send P 2}end

70. C. Varela; Adapted with permission from S. Haridi and P. Van Roy70 Building locks with cells The basic way to program with shared state is by using locks A lock is a region of the program that can only be occupied by one thread at a time. If a second thread attempts to enter, it will suspend until the first thread exits. More sophisticated versions of locks are monitors and transactions: –Monitors: locks with a gating mechanism (e.g., wait/notify in Java) to control which threads enter and exit and when. Monitors are the standard primitive for concurrent programming in Java. –Transactions: locks that have two exits, a normal and abnormal exit. Upon abnormal exit (called « abort »), all operations performed in the lock are undone, as if they were never done. Normal exit is called « commit ». Locks can be built with cells. The idea is simple: the cell contains a token. A thread attempting to enter the lock takes the token. A thread that finds no token will wait until the token is put back.

71. C. Varela; Adapted with permission from S. Haridi and P. Van Roy71 Building active objects with ports Here is a simple active object: declare P in local Xs in {NewPort Xs P} thread {ForAll Xs proc {$ X} {Browse X} end} end end {Send P foo(1)} thread {Send P bar(2)} end

72. C. Varela; Adapted with permission from S. Haridi and P. Van Roy72 Defining ports with cells A port is an unbundled stateful ADT: proc {NewPort S P} C={NewCell S} in P={Wrap C} end proc {Send P X} C={Unwrap P} Old in {Exchange C X|Old Old} end Anyone can do a send because anyone can do an exchange

73. C. Varela; Adapted with permission from S. Haridi and P. Van Roy73 Active objects with classes An active object’s behavior can be defined by a class The class is used to create a (passive) object, which is invoked by one thread that reads from a port’s stream Anyone can send a message to the object asynchronously, and the object will execute them one after the other, in sequential fashion: declare ActObj in local Obj Xs P in Obj={New Class init} {NewPort Xs P} thread {ForAll Xs proc {$ M} {Obj M} end} end proc {ActObj M} {Send P M} end end {ActObj msg(1)} Note that {Obj M} is synchronous and {ActObj M} is asynchronous!

74. C. Varela; Adapted with permission from S. Haridi and P. Van Roy74 Creating active objects with NewActive We can create a function NewActive that behaves like New except that it creates an active object: fun {NewActive Class Init} Obj Xs P in Obj={New Class Init} {NewPort Xs P} thread {ForAll Xs proc {$ M} {Obj M} end} end proc {$ M} {Send P M} end end ActObj = {NewActive Class init}

75. C. Varela; Adapted with permission from S. Haridi and P. Van Roy75 Making active objects synchronous We can make an active object synchronous by using a dataflow variable to store a result, and waiting for the result before continuing fun {NewSynchronousActive Class Init} Obj Xs P in Obj={New Class Init} {NewPort Xs P} thread {ForAll Xs proc {$ msg(M X)} {Obj M} X=unit end} end proc {$ M} X in {Send P msg(M X)} {Wait X} end end This can be modified to handle when the active object raises an exception, to pass the exception back to the caller

76. C. Varela; Adapted with permission from S. Haridi and P. Van Roy76 Playing catch class Bounce attr other count:0 meth init(Other) other:=Other end meth ball count:=@count+1 {@other ball} end meth get(X) X=@count end end declare B1 B2 in B1={NewActive Bounce init(B2)} B2={NewActive Bounce init(B1)} % Get the ball bouncing {B1 ball} % Follow the bounces {Browse {B1 get($)}} B1B2 ball

77. C. Varela; Adapted with permission from S. Haridi and P. Van Roy77 An area server class AreaServer meth init skip end meth square(X A) A=X*X end meth circle(R A) A=3.14*R*R end end declare S in S={NewActive AreaServer init} % Query the server declare A in {S square(10 A)} {Browse A} declare A in {S circle(20 A)} {Browse A}

78. C. Varela; Adapted with permission from S. Haridi and P. Van Roy78 Event manager with active objects An event manager contains a set of event handlers Each handler is a triple Id#F#S where Id identifies it, F is the state update function, and S is the state Reception of an event causes all triples to be replaced by Id#F#{F E S} (transition from F to {F E S}) The manager EM is an active object with four methods: –{EM init} initializes the event manager –{EM event(E)} posts event E at the manager –{EM add(F S Id)} adds new handler with F, S, and returns Id –{EM delete(Id S)} removed handler Id, returns state This example taken from real use in Erlang

79. C. Varela; Adapted with permission from S. Haridi and P. Van Roy79 Defining the event manager Mix of functional and object-oriented style class EventManager attr handlers meth init handlers:=nil end meth event(E) handlers:= {Map @handlers fun {$ Id#F#S} Id#F#{F E S} end} end meth add(F S Id) Id={NewName} handlers:=Id#F#S|@handlers end meth delete(DId DS) handlers:={List.partition @handlers fun {$ Id#F#S} DId==Id end [_#_#DS]} end State transition done using functional programming

80. C. Varela; Adapted with permission from S. Haridi and P. Van Roy80 Using the event manager Simple memory-based handler keeps list of events declare EM MemH Id in EM={NewActive EventManager init} MemH=fun {$ E Buf} E|Buf end {EM add(MemH nil Id)} {EM event(a1)} {EM event(a2)}... An event handler is purely functional, yet when put in the event manager, the latter is a concurrent imperative program. This is an example of separation of concerns by using multiple paradigms.

81. C. Varela; Adapted with permission from S. Haridi and P. Van Roy81 Exercises 1.VRH Exercise 4.11.3 (page 339) Compare the sequential vs concurrent execution performance of equivalent SALSA programs. 2.VRH Exercise 4.11.5 (page 339) 3.SALSA asynchronous message passing enables to tag messages with properties: priority, delay, and waitfor. Compare these mechanisms with Oz thread priorities, time delays and alarms, and futures. 4.How do SALSA tokens relate to Oz dataflow variables and futures? 5.What is the difference between multiple thread termination detection in Oz and join blocks in SALSA?

82. C. Varela; Adapted with permission from S. Haridi and P. Van Roy82 Exercises 6.Do Python, Java and C++ provide a linguistic abstraction for active objects? If so, which? If not, how would you go about implementing this abstraction? 7.Exercise VRH 7.9.6(a) (pg 568) 8.Write a Producer-Consumer program in SALSA (see VRH Section 4.3.1. for specification.)