1. Introduction Companion slides for The Art of Multiprocessor Programming by Maurice Herlihy & Nir Shavit TexPoint fonts used in EMF. Read the TexPoint manual before you delete this box.: A A AA

2. Art of Multiprocessor Programming 2 Moores Law Clock speed flattening sharply Transistor count still rising

3. Moores Law (in practice) Art of Multiprocessor Programming3

4. 4 Nearly Extinct: the Uniprocesor memory cpu

5. Art of Multiprocessor Programming 5 Endangered: The Shared Memory Multiprocessor (SMP) cache Bus shared memory cache

6. Art of Multiprocessor Programming 6 The New Boss: The Multicore Processor (CMP) cache Bus shared memory cache All on the same chip Sun T2000 Niagara

7. Art of Multiprocessor Programming 7 From the 2008 press… …Intel has announced a press conference in San Francisco on November 17th, where it will officially launch the Core i7 Nehalem processor… …Suns next generation Enterprise T5140 and T5240 servers, based on the 3rd Generation UltraSPARC T2 Plus processor, were released two days ago…

8. Art of Multiprocessor Programming 8 Why is Kunle Smiling? Niagara 1

9. Art of Multiprocessor Programming 9 Why do we care? Time no longer cures software bloat –The free ride is over When you double your programs path length –You cant just wait 6 months –Your software must somehow exploit twice as much concurrency

10. Art of Multiprocessor Programming 10 Traditional Scaling Process User code Traditional Uniprocessor Speedup 1.8x 7x 3.6x Time: Moores law

11. Ideal Scaling Process Art of Multiprocessor Programming 11 User code Multicore Speedup 1.8x7x3.6x Unfortunately, not so simple…

12. Actual Scaling Process Art of Multiprocessor Programming 12 1.8x 2x 2.9x User code Multicore Speedup Parallelization and Synchronization require great care…

13. Art of Multiprocessor Programming 13 Multicore Programming: Course Overview Fundamentals –Models, algorithms, impossibility Real-World programming –Architectures –Techniques

14. Art of Multiprocessor Programming 14 Sequential Computation memory object thread

15. Art of Multiprocessor Programming 15 Concurrent Computation memory object threads

16. Art of Multiprocessor Programming 16 Asynchrony Sudden unpredictable delays –Cache misses (short) –Page faults (long) –Scheduling quantum used up (really long)

17. Art of Multiprocessor Programming 17 Model Summary Multiple threads –Sometimes called processes Single shared memory Objects live in memory Unpredictable asynchronous delays

18. 18 Road Map We are going to focus on principles first, then practice –Start with idealized models –Look at simplistic problems –Emphasize correctness over pragmatism –Correctness may be theoretical, but incorrectness has practical impact Art of Multiprocessor Programming

19. 19 Concurrency Jargon Hardware –Processors Software –Threads, processes Sometimes OK to confuse them, sometimes not. Art of Multiprocessor Programming

20. 20 Parallel Primality Testing Challenge –Print primes from 1 to 10 10 Given –Ten-processor multiprocessor –One thread per processor Goal –Get ten-fold speedup (or close) Art of Multiprocessor Programming

21. 21 Load Balancing Split the work evenly Each thread tests range of 10 9 … …10 910 2·10 9 1 P0P0 P1P1 P9P9

22. 22 Procedure for Thread i void primePrint { int i = ThreadID.get(); // IDs in {0..9} for (j = i*10 9 +1, j<(i+1)*10 9 ; j++) { if (isPrime(j)) print(j); } Art of Multiprocessor Programming

23. 23 Issues Higher ranges have fewer primes Yet larger numbers harder to test Thread workloads –Uneven –Hard to predict Art of Multiprocessor Programming

24. 24 Issues Higher ranges have fewer primes Yet larger numbers harder to test Thread workloads –Uneven –Hard to predict Need dynamic load balancing rejected

25. Art of Multiprocessor Programming 25 17 18 19 Shared Counter each thread takes a number

26. 26 Procedure for Thread i int counter = new Counter(1); void primePrint { long j = 0; while (j < 10 10 ) { j = counter.getAndIncrement(); if (isPrime(j)) print(j); } Art of Multiprocessor Programming

27. 27 Counter counter = new Counter(1); void primePrint { long j = 0; while (j < 10 10 ) { j = counter.getAndIncrement(); if (isPrime(j)) print(j); } Procedure for Thread i Shared counter object

28. Art of Multiprocessor Programming 28 Where Things Reside cache Bus cache 1 shared counter shared memory void primePrint { int i = ThreadID.get(); // IDs in {0..9} for (j = i*10 9 +1, j<(i+1)*10 9 ; j++) { if (isPrime(j)) print(j); } code Local variables

29. Art of Multiprocessor Programming 29 Procedure for Thread i Counter counter = new Counter(1); void primePrint { long j = 0; while (j < 10 10 ) { j = counter.getAndIncrement(); if (isPrime(j)) print(j); } Stop when every value taken

30. Art of Multiprocessor Programming 30 Counter counter = new Counter(1); void primePrint { long j = 0; while (j < 10 10 ) { j = counter.getAndIncrement(); if (isPrime(j)) print(j); } Procedure for Thread i Increment & return each new value

31. 31 Counter Implementation public class Counter { private long value; public long getAndIncrement() { return value++; } Art of Multiprocessor Programming

32. 32 Counter Implementation public class Counter { private long value; public long getAndIncrement() { return value++; } OK for single thread, not for concurrent threads

33. Art of Multiprocessor Programming 33 What It Means public class Counter { private long value; public long getAndIncrement() { return value++; }

34. Art of Multiprocessor Programming 34 What It Means public class Counter { private long value; public long getAndIncrement() { return value++; } temp = value; value = temp + 1; return temp;

35. Art of Multiprocessor Programming 35 time Not so good… Value… 1 read 1 read 1 write 2 read 2 write 3 write 2 232

36. Art of Multiprocessor Programming 36 Is this problem inherent? If we could only glue reads and writes together… read write read write !!

37. 37 Challenge public class Counter { private long value; public long getAndIncrement() { temp = value; value = temp + 1; return temp; } Art of Multiprocessor Programming

38. 38 Challenge public class Counter { private long value; public long getAndIncrement() { temp = value; value = temp + 1; return temp; } Make these steps atomic (indivisible)

39. Art of Multiprocessor Programming 39 Hardware Solution public class Counter { private long value; public long getAndIncrement() { temp = value; value = temp + 1; return temp; } ReadModifyWrite() instruction

40. Art of Multiprocessor Programming 40 An Aside: Java public class Counter { private long value; public long getAndIncrement() { synchronized { temp = value; value = temp + 1; } return temp; }

41. Art of Multiprocessor Programming 41 An Aside: Java public class Counter { private long value; public long getAndIncrement() { synchronized { temp = value; value = temp + 1; } return temp; } Synchronized block

42. Art of Multiprocessor Programming 42 An Aside: Java public class Counter { private long value; public long getAndIncrement() { synchronized { temp = value; value = temp + 1; } return temp; } Mutual Exclusion

43. 43 Mutual Exclusion, or Alice & Bob share a pond AB Art of Multiprocessor Programming

44. 44 Alice has a pet AB Art of Multiprocessor Programming

45. 45 Bob has a pet AB Art of Multiprocessor Programming

46. 46 The Problem AB The pets dont get along Art of Multiprocessor Programming

47. 47 Formalizing the Problem Two types of formal properties in asynchronous computation: Safety Properties –Nothing bad happens ever Liveness Properties –Something good happens eventually Art of Multiprocessor Programming

48. 48 Formalizing our Problem Mutual Exclusion –Both pets never in pond simultaneously –This is a safety property No Deadlock –if only one wants in, it gets in –if both want in, one gets in. –This is a liveness property Art of Multiprocessor Programming

49. 49 Simple Protocol Idea –Just look at the pond Gotcha –Not atomic –Trees obscure the view Art of Multiprocessor Programming

50. 50 Interpretation Threads cant see what other threads are doing Explicit communication required for coordination Art of Multiprocessor Programming

51. 51 Cell Phone Protocol Idea –Bob calls Alice (or vice-versa) Gotcha –Bob takes shower –Alice recharges battery –Bob out shopping for pet food … Art of Multiprocessor Programming

52. 52 Interpretation Message-passing doesnt work Recipient might not be –Listening –There at all Communication must be –Persistent (like writing) –Not transient (like speaking) Art of Multiprocessor Programming

53. 53 Can Protocol cola Art of Multiprocessor Programming

54. 54 Bob conveys a bit AB cola Art of Multiprocessor Programming

55. 55 Bob conveys a bit AB cola Art of Multiprocessor Programming

56. 56 Can Protocol Idea –Cans on Alices windowsill –Strings lead to Bobs house –Bob pulls strings, knocks over cans Gotcha –Cans cannot be reused –Bob runs out of cans Art of Multiprocessor Programming

57. 57 Interpretation Cannot solve mutual exclusion with interrupts –Sender sets fixed bit in receivers space –Receiver resets bit when ready –Requires unbounded number of interrupt bits Art of Multiprocessor Programming

58. 58 Flag Protocol AB Art of Multiprocessor Programming

59. 59 Alices Protocol (sort of) AB Art of Multiprocessor Programming

60. 60 Bobs Protocol (sort of) AB Art of Multiprocessor Programming

61. 61 Alices Protocol Raise flag Wait until Bobs flag is down Unleash pet Lower flag when pet returns Art of Multiprocessor Programming

62. 62 Bobs Protocol Raise flag Wait until Alices flag is down Unleash pet Lower flag when pet returns danger!

63. 63 Bobs Protocol (2 nd try) Raise flag While Alices flag is up –Lower flag –Wait for Alices flag to go down –Raise flag Unleash pet Lower flag when pet returns Art of Multiprocessor Programming

64. 64 Bobs Protocol Raise flag While Alices flag is up –Lower flag –Wait for Alices flag to go down –Raise flag Unleash pet Lower flag when pet returns Bob defers to Alice

65. 65 The Flag Principle Raise the flag Look at others flag Flag Principle: –If each raises and looks, then –Last to look must see both flags up Art of Multiprocessor Programming

66. 66 Proof of Mutual Exclusion Assume both pets in pond –Derive a contradiction –By reasoning backwards Consider the last time Alice and Bob each looked before letting the pets in Without loss of generality assume Alice was the last to look… Art of Multiprocessor Programming

67. 67 Proof time Alices last look Alice last raised her flag Bobs last look QED Alice must have seen Bobs Flag. A Contradiction Bob last raised flag

68. 68 Proof of No Deadlock If only one pet wants in, it gets in. Art of Multiprocessor Programming

69. 69 Proof of No Deadlock If only one pet wants in, it gets in. Deadlock requires both continually trying to get in. Art of Multiprocessor Programming

70. 70 Proof of No Deadlock If only one pet wants in, it gets in. Deadlock requires both continually trying to get in. If Bob sees Alices flag, he gives her priority (a gentleman…) QED

71. 71 Remarks Protocol is unfair –Bobs pet might never get in Protocol uses waiting –If Bob is eaten by his pet, Alices pet might never get in Art of Multiprocessor Programming

72. 72 Moral of Story Mutual Exclusion cannot be solved by –transient communication (cell phones) –interrupts (cans) It can be solved by – one-bit shared variables – that can be read or written Art of Multiprocessor Programming

73. 73 The Arbiter Problem (an aside) Pick a point

74. 74 The Fable Continues Alice and Bob fall in love & marry Art of Multiprocessor Programming

75. 75 The Fable Continues Alice and Bob fall in love & marry Then they fall out of love & divorce –She gets the pets –He has to feed them Art of Multiprocessor Programming

76. 76 The Fable Continues Alice and Bob fall in love & marry Then they fall out of love & divorce –She gets the pets –He has to feed them Leading to a new coordination problem: Producer-Consumer Art of Multiprocessor Programming

77. 77 Bob Puts Food in the Pond A Art of Multiprocessor Programming

78. 78 mmm… Alice releases her pets to Feed B mmm… Art of Multiprocessor Programming

79. 79 Producer/Consumer Alice and Bob cant meet –Each has restraining order on other –So he puts food in the pond –And later, she releases the pets Avoid –Releasing pets when theres no food –Putting out food if uneaten food remains Art of Multiprocessor Programming

80. 80 Producer/Consumer Need a mechanism so that –Bob lets Alice know when food has been put out –Alice lets Bob know when to put out more food Art of Multiprocessor Programming

81. 81 Surprise Solution AB cola Art of Multiprocessor Programming

82. 82 Bob puts food in Pond AB cola Art of Multiprocessor Programming

83. 83 Bob knocks over Can AB cola Art of Multiprocessor Programming

84. 84 Alice Releases Pets AB cola yum… B Art of Multiprocessor Programming

85. 85 Alice Resets Can when Pets are Fed AB cola Art of Multiprocessor Programming

86. 86 Pseudocode while (true) { while (can.isUp()){}; pet.release(); pet.recapture(); can.reset(); } Alices code

87. Art of Multiprocessor Programming 87 Pseudocode while (true) { while (can.isUp()){}; pet.release(); pet.recapture(); can.reset(); } Alices code while (true) { while (can.isDown()){}; pond.stockWithFood(); can.knockOver(); } Bobs code

88. 88 Correctness Mutual Exclusion –Pets and Bob never together in pond Art of Multiprocessor Programming

89. 89 Correctness Mutual Exclusion –Pets and Bob never together in pond No Starvation if Bob always willing to feed, and pets always famished, then pets eat infinitely often. Art of Multiprocessor Programming

90. 90 Correctness Mutual Exclusion –Pets and Bob never together in pond No Starvation if Bob always willing to feed, and pets always famished, then pets eat infinitely often. Producer/Consumer The pets never enter pond unless there is food, and Bob never provides food if there is unconsumed food. safety liveness safety

91. 91 Could Also Solve Using Flags AB Art of Multiprocessor Programming

92. 92 Waiting Both solutions use waiting –while(mumble){} In some cases waiting is problematic –If one participant is delayed –So is everyone else –But delays are common & unpredictable Art of Multiprocessor Programming

93. 93 The Fable drags on … Bob and Alice still have issues Art of Multiprocessor Programming

94. 94 The Fable drags on … Bob and Alice still have issues So they need to communicate Art of Multiprocessor Programming

95. 95 The Fable drags on … Bob and Alice still have issues So they need to communicate They agree to use billboards … Art of Multiprocessor Programming

96. 96 E 1 D 2 C 3 Billboards are Large B 3 A 1 Letter Tiles From Scrabble box Art of Multiprocessor Programming

97. 97 E 1 D 2 C 3 Write One Letter at a Time … B 3 A 1 W 4 A 1 S 1 H 4 Art of Multiprocessor Programming

98. 98 To post a message W 4 A 1 S 1 H 4 A 1 C 3 R 1 T 1 H 4 E 1 whew Art of Multiprocessor Programming

99. 99 S 1 Lets send another message S 1 E 1 L 1 L 1 L 1 V 4 L 1 A 1 M 3 A 1 A 1 P 3 Art of Multiprocessor Programming

100. 100 Uh-Oh A 1 C 3 R 1 T 1 H 4 E 1 S 1 E 1 L 1 L 1 L 1 OK Art of Multiprocessor Programming

101. 101 Readers/Writers Devise a protocol so that –Writer writes one letter at a time –Reader reads one letter at a time –Reader sees snapshot Old message or new message No mixed messages Art of Multiprocessor Programming

102. 102 Readers/Writers (continued) Easy with mutual exclusion But mutual exclusion requires waiting –One waits for the other –Everyone executes sequentially Remarkably –We can solve R/W without mutual exclusion Art of Multiprocessor Programming

103. 103 Esoteric? Java container size() method Single shared counter? –incremented with each add() and –decremented with each remove() Threads wait to exclusively access counter performance bottleneck

104. 104 Readers/Writers Solution Each thread i has size[i] counter –only it increments or decrements. To get objects size, a thread reads a snapshot of all counters This eliminates the bottleneck Art of Multiprocessor Programming

105. 105 Why do we care? We want as much of the code as possible to execute concurrently (in parallel) A larger sequential part implies reduced performance Amdahls law: this relation is not linear… Art of Multiprocessor Programming

106. 106 Amdahls Law Speedup= 1-thread execution time n-thread execution time

107. Art of Multiprocessor Programming 107 Amdahls Law Speedup=

108. Art of Multiprocessor Programming 108 Amdahls Law Speedup= Parallel fraction

109. Art of Multiprocessor Programming 109 Amdahls Law Speedup= Parallel fraction Sequential fraction

110. Art of Multiprocessor Programming 110 Amdahls Law Speedup= Parallel fraction Sequential fraction Number of threads

111. Amdahls Law (in practice) Art of Multiprocessor Programming111

112. 112 Example Ten processors 60% concurrent, 40% sequential How close to 10-fold speedup? Art of Multiprocessor Programming

113. 113 Example Ten processors 60% concurrent, 40% sequential How close to 10-fold speedup? Speedup = 2.17= Art of Multiprocessor Programming

114. 114 Example Ten processors 80% concurrent, 20% sequential How close to 10-fold speedup? Art of Multiprocessor Programming

115. 115 Example Ten processors 80% concurrent, 20% sequential How close to 10-fold speedup? Speedup = 3.57 = Art of Multiprocessor Programming

116. 116 Example Ten processors 90% concurrent, 10% sequential How close to 10-fold speedup? Art of Multiprocessor Programming

117. 117 Example Ten processors 90% concurrent, 10% sequential How close to 10-fold speedup? Speedup = 5.26=

118. 118 Example Ten processors 99% concurrent, 01% sequential How close to 10-fold speedup? Art of Multiprocessor Programming

119. 119 Example Ten processors 99% concurrent, 01% sequential How close to 10-fold speedup? Speedup = 9.17=

120. Back to Real-World Multicore Scaling 1.8x 2x 2.9x User code Multicore Speedup Not reducing sequential % of code Art of Multiprocessor Programming

121. Shared Data Structures 75% Unshared 25% Shared Coarse Grained Fine Grained 75% Unshared 25% Shared

122. Shared Data Structures 75% Unshared 25% Shared Coarse Grained Fine Grained Why only 2.9 speedup 75% Unshared 25% Shared Honk!

123. Shared Data Structures 75% Unshared 25% Shared Coarse Grained Fine Grained Why fine-grained parallelism maters 75% Unshared 25% Shared Honk!

124. Art of Multiprocessor Programming Diminishing Returns This course is about the parts that are hard to make concurrent … but still have a big influence on speedup!

125. Art of Multiprocessor Programming 125 This work is licensed under a Creative Commons Attribution- ShareAlike 2.5 License.Creative Commons Attribution- ShareAlike 2.5 License You are free: –to Share to copy, distribute and transmit the work –to Remix to adapt the work Under the following conditions: –Attribution. You must attribute the work to The Art of Multiprocessor Programming (but not in any way that suggests that the authors endorse you or your use of the work). –Share Alike. If you alter, transform, or build upon this work, you may distribute the resulting work only under the same, similar or a compatible license. For any reuse or distribution, you must make clear to others the license terms of this work. The best way to do this is with a link to –http://creativecommons.org/licenses/by-sa/3.0/. Any of the above conditions can be waived if you get permission from the copyright holder. Nothing in this license impairs or restricts the author's moral rights.