Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Java Memory Consistency Model
@LAFK_pl
Consultant @
Tomasz Borek
GeeCON 2015 Kraków!
@LAFK_pl
Consultant @
http://2015.geecon.org/register/
I'll take ALL feedback I can get.
@LAFK_pl, #JMCM
@LAFK_pl
Consultant @
Symentis
@LAFK_pl
Symentis
@LAFK_pl
Jarosław Pałka Kuba Marchwicki
Today...
● A little bragging
● Fallacy correction
● Memory model
● The Java one mostly
● Short advice what about it
● Lot ...
Who knows (heard of)?
● Gene Amdahl?
● Gordon Moore?
● Leslie Lamporte?
● Bill Pugh?
● Sarita Adve?
● Hans Boehm?
● Martin...
Hands up, who...
● Doesn't program
● Knows Moore's law?
● Knows Amdahl's law?
● Can explain concurrency vs parallelism?
● ...
Fallacy #0
@LAFK_pl
Java Memory Model is about GC, memory
management, Eden, Survivors etc.
Not true at all!
@LAFK_pl
Java Memory Model is about GC, memory
management, Eden, Survivors, etc.
@LAFK_pl
MANAGEMENT!!
@LAFK_pl
Written in 2012...
@LAFK_pl
… and still there
@LAFK_pl
Correction!
@LAFK_pl
Consultant @
JLS, section 17.4:
A memory model describes, given a program and
an execution trace of that program, whether the
execution...
JLS, section 17.4:
A memory model describes, given a program and
an execution trace of that program, whether the
execution...
Sarita Adve: memory consistency
model
@LAFK_pl
Consultant @
Aleksey Shipilev:
@LAFK_pl
Consultant @
Memory model answers one simple question:
What values can a particular read in a p...
Bill Pugh, Jeremy Manson
● Most cores have many cache layers
● What if 2 cores look at same value?
● Memory model defines ...
Bill Pugh, Jeremy Manson:
What is a memory model, anyway?
At the processor level, a memory model defines
necessary and suf...
So?
● Memory CONSISTENCY
● Allowed optimisations
● Possible executions of a (possibly
multithreaded!) program
● Which core...
Fallacy #1
I don't need to worry about JMCM since
REALLY smart engineers crafted it.
@LAFK_pl
Consultant @
Half-true
I don't need to worry about JMCM since
REALLY smart engineers crafted it
@LAFK_pl
Consultant @
Smart, sure! But still:
● Smart people are still people
● JMCM is damn hard! Yeah, they botched it.
● Java <> JVM
● JMCM i...
JSR-133?
● Messed up final
● Spec not for humans
● Messed up double-locking
● Messed up volatile
● Each implementation on ...
More?
@LAFK_pl
Consultant @
JEPS-188 and JMM9?
Fallacy #2
JMCM is irrelevant for me.
@LAFK_pl
Consultant @
Depends!
● What you write and with what
● Java? Not?
● Need performance?
● What platforms?
● Multicore era...
@LAFK_pl
Con...
Fallacy #3
JMCM is for fanatics. I have frameworks for that.
@LAFK_pl
Consultant @
Moore's ”law”?
Consultant @Consultant @
@LAFK_pl
Moore's ”law”
I see Moore’s law dying here in the next decade
or so.
– Gordon Moore, 2015
Consultant @
@LAFK_pl
Consultant...
Amdahl's law?
@LAFK_pl
Consultant @
Amdahl's law
@LAFK_pl
The speedup of a program using multiple
processors in parallel computing is limited by the
sequentia...
Rok 1967, Gene Amdahl states:
@LAFK_pl
For over a decade prophets have voiced the
contention that the organization of a si...
Rok 1967, Gene Amdahl states:
@LAFK_pl
For over a decade prophets have voiced the
contention that the organization of a si...
Cores... MOAR!!
@LAFK_pl
Consultant @
Data distance
● http://i.imgur.com/k0t1e.png
So, data distance varies
@LAFK_pl
Know thyne cache lines!
Consultant @
@LAFK_pl
Consultant @
So if it matters, what then?
Where it matters
● Javac / Jython / ...
● JIT
● Hardware, duh!
● Each time: another
team
@LAFK_pl
Consultant @
Hardware
● Various ISA CPUs
● Number of registers
● Caches size or type, buses implementations
● Cache protocols (MESI, AM...
Program / code optimizations?
● Reorder
● (e.g. prescient store)
● Remove what's unnecessary
● (e.g. synchronize)
● Replac...
Exemplary CPU
@LAFK_pl
Consultant @
Barriers / fences
„once memory has been pushed to the cache
then a protocol of messages will occur to
ensure all caches ar...
Barriers / Fences
● CPU instruction
● Means ”flush BUFFER now!”
● CMPXCHG (may be
lacking!)
● Forces update
● Starts cache...
@LAFK_pl
Consultant @
Words of summary and gratitude
Doug Lea says:
@LAFK_pl
Consultant @
The best way is to build up a small repertoire of
constructions that you know the ans...
Doug Lea advice:
@LAFK_pl
Consultant @
The best way is to build up a small repertoire
of constructions that you know the a...
Mechanical sympathy:
@LAFK_pl
Consultant @
● Cache lines misses hurt
● Going to main memory hurts
● Cycles are important
●...
Gordon Moore
● Fairchild Semi-
conductors co-
founder
● ”Law author”
● Intel co-founder
@LAFK_pl
Gene Amdahl
● IBM fellow
● IBM & Amdahl
mainframes
● Coined law in 1967
@LAFK_pl
Leslie Lamport
● Distributed system
clocks
● Happens before
● Sequential
consistency
@LAFK_pl
Bill Pugh
● FindBugs
● Java Memory Model
is broken
● Final - Volatile
● Double-checked
locking
● ”New” JMM
@LAFK_pl
Sarita Adve
● Java Memory Model
is broken
● Great many MCM
papers
● Best MCM def I found
@LAFK_pl
Martin Thompson
● Mechanical sympathy
blog & mailing list
● Aeron protocol
● Mechanical sympathy
proponent
@LAFK_pl
This wouldn't have happened if not
● Jarek Pałka, who kicked me out here some
time ago
● Those folks, who said ”make more”...
Not possible without:
● Leslie Lamport's works on distributed sistems
● Bill Pugh's work on JSR-133!
http://www.cs.umd.edu...
Terrific – and tough - reading
● Martin Thompson: Mechanical Sympathy (mailing list & blog)
● JEPS 188: http://openjdk.jav...
Laws and related:
● Moore's ”law”:
http://www.cs.utexas.edu/~fussell/courses/cs352h/papers/moore.pdf
● Rock's law: http://...
Special thanks
● Konrad Malawski and Tomek Kowalczewski,
these guys really dig that stuff
● Bartosz Milewski who helped me...
YOU! You persevered through!
@LAFK_pl
Consultant @
Upcoming SlideShare
Loading in …5
×

4Developers 2015: Java Memory Consistency Model or intro to multithreaded programming - Tomasz Borek, Jacek Jagieła

194 views

Published on

Tomasz Borek, Jacek Jagieła

Language: English

JMM is often (incorrectly) treated as Java Memory Management, which it's NOT. It's NOT about Eden, Survivors, GC, heap memory and all that. It's about Memory Consistency, or making things visible to threads, and making things understandable for humans. I'll tell you what is memory model, why should it be consistent, will explain memory barriers, volatile, final, synchronization and from there will delve in lock-free, wait-free and prepping this with examples all along.

Published in: Software
  • Be the first to comment

  • Be the first to like this

4Developers 2015: Java Memory Consistency Model or intro to multithreaded programming - Tomasz Borek, Jacek Jagieła

  1. 1. Java Memory Consistency Model @LAFK_pl Consultant @ Tomasz Borek
  2. 2. GeeCON 2015 Kraków! @LAFK_pl Consultant @ http://2015.geecon.org/register/
  3. 3. I'll take ALL feedback I can get. @LAFK_pl, #JMCM @LAFK_pl Consultant @
  4. 4. Symentis @LAFK_pl
  5. 5. Symentis @LAFK_pl Jarosław Pałka Kuba Marchwicki
  6. 6. Today... ● A little bragging ● Fallacy correction ● Memory model ● The Java one mostly ● Short advice what about it ● Lot of links for later reading ● Yeah, 45 minutes only :P @LAFK_pl Consultant @
  7. 7. Who knows (heard of)? ● Gene Amdahl? ● Gordon Moore? ● Leslie Lamporte? ● Bill Pugh? ● Sarita Adve? ● Hans Boehm? ● Martin Thompson? ● Aleksey Shipilev? @LAFK_pl
  8. 8. Hands up, who... ● Doesn't program ● Knows Moore's law? ● Knows Amdahl's law? ● Can explain concurrency vs parallelism? ● Codes with mechanical sympathy? ● Tries to? ● Knows what mechanical sympathy is? @LAFK_pl
  9. 9. Fallacy #0 @LAFK_pl Java Memory Model is about GC, memory management, Eden, Survivors etc.
  10. 10. Not true at all! @LAFK_pl Java Memory Model is about GC, memory management, Eden, Survivors, etc.
  11. 11. @LAFK_pl
  12. 12. MANAGEMENT!! @LAFK_pl
  13. 13. Written in 2012... @LAFK_pl
  14. 14. … and still there @LAFK_pl
  15. 15. Correction! @LAFK_pl Consultant @
  16. 16. JLS, section 17.4: A memory model describes, given a program and an execution trace of that program, whether the execution trace is a legal execution of the program. The Java programming language memory model works by examining each read in an execution trace and checking that the write observed by that read is valid according to certain rules. @LAFK_pl Consultant @
  17. 17. JLS, section 17.4: A memory model describes, given a program and an execution trace of that program, whether the execution trace is a legal execution of the program. The Java programming language memory model works by examining each read in an execution trace and checking that the write observed by that read is valid according to certain rules. @LAFK_pl Consultant @
  18. 18. Sarita Adve: memory consistency model @LAFK_pl Consultant @
  19. 19. Aleksey Shipilev: @LAFK_pl Consultant @ Memory model answers one simple question: What values can a particular read in a program return?
  20. 20. Bill Pugh, Jeremy Manson ● Most cores have many cache layers ● What if 2 cores look at same value? ● Memory model defines when and who sees what ● There're strong and weak models ● Strong guarantee seeing same things across whole system ● Weak only sometimes, via barriers / fences @LAFK_pl Consultant @
  21. 21. Bill Pugh, Jeremy Manson: What is a memory model, anyway? At the processor level, a memory model defines necessary and sufficient conditions for knowing that writes to memory by other processors are visible to the current processor, and writes by the current processor are visible to other processors. @LAFK_pl Consultant @
  22. 22. So? ● Memory CONSISTENCY ● Allowed optimisations ● Possible executions of a (possibly multithreaded!) program ● Which cores / threads see which values ● How to make it consistent for programmers ● What you're allowed to assume @LAFK_pl Consultant @
  23. 23. Fallacy #1 I don't need to worry about JMCM since REALLY smart engineers crafted it. @LAFK_pl Consultant @
  24. 24. Half-true I don't need to worry about JMCM since REALLY smart engineers crafted it @LAFK_pl Consultant @
  25. 25. Smart, sure! But still: ● Smart people are still people ● JMCM is damn hard! Yeah, they botched it. ● Java <> JVM ● JMCM is for JVM... but with Java in mind ● NO tech is a talisman of functionality! @LAFK_pl Consultant @
  26. 26. JSR-133? ● Messed up final ● Spec not for humans ● Messed up double-locking ● Messed up volatile ● Each implementation on it's own @LAFK_pl Consultant @
  27. 27. More? @LAFK_pl Consultant @ JEPS-188 and JMM9?
  28. 28. Fallacy #2 JMCM is irrelevant for me. @LAFK_pl Consultant @
  29. 29. Depends! ● What you write and with what ● Java? Not? ● Need performance? ● What platforms? ● Multicore era... @LAFK_pl Consultant @
  30. 30. Fallacy #3 JMCM is for fanatics. I have frameworks for that. @LAFK_pl Consultant @
  31. 31. Moore's ”law”? Consultant @Consultant @ @LAFK_pl
  32. 32. Moore's ”law” I see Moore’s law dying here in the next decade or so. – Gordon Moore, 2015 Consultant @ @LAFK_pl Consultant @
  33. 33. Amdahl's law? @LAFK_pl Consultant @
  34. 34. Amdahl's law @LAFK_pl The speedup of a program using multiple processors in parallel computing is limited by the sequential fraction of the program. For example, if 95% of the program can be parallelized, the theoretical maximum speedup using parallel computing would be 20× as shown in the diagram, no matter how many processors are used. Consultant @
  35. 35. Rok 1967, Gene Amdahl states: @LAFK_pl For over a decade prophets have voiced the contention that the organization of a single computer has reached its limits and that truly significant advances can be made only by interconnection of a multiplicity of computers in such a manner as to permit cooperative solution. Consultant @
  36. 36. Rok 1967, Gene Amdahl states: @LAFK_pl For over a decade prophets have voiced the contention that the organization of a single computer has reached its limits and that truly significant advances can be made only by interconnection of a multiplicity of computers in such a manner as to permit cooperative solution. Consultant @
  37. 37. Cores... MOAR!! @LAFK_pl Consultant @
  38. 38. Data distance ● http://i.imgur.com/k0t1e.png
  39. 39. So, data distance varies @LAFK_pl Know thyne cache lines! Consultant @
  40. 40. @LAFK_pl Consultant @ So if it matters, what then?
  41. 41. Where it matters ● Javac / Jython / ... ● JIT ● Hardware, duh! ● Each time: another team @LAFK_pl Consultant @
  42. 42. Hardware ● Various ISA CPUs ● Number of registers ● Caches size or type, buses implementations ● Cache protocols (MESI, AMD's MOESI, Intel's...) ● How many functional units per CPU ● How many CPUs ● Pipeline: ● Instruction decode > address decode > memory fetch > register fetch > compute ... @LAFK_pl Consultant @
  43. 43. Program / code optimizations? ● Reorder ● (e.g. prescient store) ● Remove what's unnecessary ● (e.g. synchronize) ● Replace instructions / shorten machine code ● Function optimizations ● (e.g. Inlining) ● ... @LAFK_pl Consultant @
  44. 44. Exemplary CPU @LAFK_pl Consultant @
  45. 45. Barriers / fences „once memory has been pushed to the cache then a protocol of messages will occur to ensure all caches are coherent for any shared data. The techniques for making memory visible from a processor core are known as memory barriers or fences. – Martin Thompson, Mechanical Sympathy differs per architecture / CPU / cache type! @LAFK_pl Consultant @
  46. 46. Barriers / Fences ● CPU instruction ● Means ”flush BUFFER now!” ● CMPXCHG (may be lacking!) ● Forces update ● Starts cache coherency protocols ● Read / Write / Full @LAFK_pl Consultant @
  47. 47. @LAFK_pl Consultant @ Words of summary and gratitude
  48. 48. Doug Lea says: @LAFK_pl Consultant @ The best way is to build up a small repertoire of constructions that you know the answers for and then never think about the JMM rules again unless you are forced to do so! Literally nobody likes figuring things out of JMM rules as stated, or can even routinely do so correctly. This is one of the many reasons we need to overhaul JMM someday.
  49. 49. Doug Lea advice: @LAFK_pl Consultant @ The best way is to build up a small repertoire of constructions that you know the answers for and then never think about the JMM rules again unless you are forced to do so! Literally nobody likes figuring things out of JMM rules as stated, or can even routinely do so correctly. This is one of the many reasons we need to overhaul JMM someday.
  50. 50. Mechanical sympathy: @LAFK_pl Consultant @ ● Cache lines misses hurt ● Going to main memory hurts ● Cycles are important ● L1, L2 caches are cheap but require cache coherency protocols and memory barriers ● Not every hardware has all barriers ●
  51. 51. Gordon Moore ● Fairchild Semi- conductors co- founder ● ”Law author” ● Intel co-founder @LAFK_pl
  52. 52. Gene Amdahl ● IBM fellow ● IBM & Amdahl mainframes ● Coined law in 1967 @LAFK_pl
  53. 53. Leslie Lamport ● Distributed system clocks ● Happens before ● Sequential consistency @LAFK_pl
  54. 54. Bill Pugh ● FindBugs ● Java Memory Model is broken ● Final - Volatile ● Double-checked locking ● ”New” JMM @LAFK_pl
  55. 55. Sarita Adve ● Java Memory Model is broken ● Great many MCM papers ● Best MCM def I found @LAFK_pl
  56. 56. Martin Thompson ● Mechanical sympathy blog & mailing list ● Aeron protocol ● Mechanical sympathy proponent @LAFK_pl
  57. 57. This wouldn't have happened if not ● Jarek Pałka, who kicked me out here some time ago ● Those folks, who said ”make more” after the lightning talk I've done ● Java Day Kiev 2014 @LAFK_pl Consultant @
  58. 58. Not possible without: ● Leslie Lamport's works on distributed sistems ● Bill Pugh's work on JSR-133! http://www.cs.umd.edu/~pugh/java/memoryModel/jsr-133-faq.html ● Sarita Adve's paperts, especially shared MCM tutorial: http://www.hpl.hp.com/techreports/Compaq-DEC/WRL-95-7.pdf @LAFK_pl Consultant @
  59. 59. Terrific – and tough - reading ● Martin Thompson: Mechanical Sympathy (mailing list & blog) ● JEPS 188: http://openjdk.java.net/jeps/188 ● Goetz et al: "Java Concurrency in Practice" ● Herilhy, Shavit, "The Art of Multiprocessor Programming" ● Adve, "Shared Memory Consistency Models: A Tutorial" ● Manson, "Special PoPL Issue: The Java Memory Model" ● Huisman, Petri, "JMM: A Formal Explanation" ● Aleksey Shipilev blog post: http://shipilev.net/blog/2014/jmm-pragmatics/ @LAFK_pl Consultant @
  60. 60. Laws and related: ● Moore's ”law”: http://www.cs.utexas.edu/~fussell/courses/cs352h/papers/moore.pdf ● Rock's law: http://en.wikipedia.org/wiki/Rock's_law ● Amdahl's law: ● http://en.wikipedia.org/wiki/Amdahl%27s_law ● Validity of the Single-Processor Approach to Achieving Large-Scale Computing Capabilities AFIPS Press, 1967 ● J.L. Gustafson, “Reevaluating Amdahl’s Law,” Comm. ACM, May 1988 ● Pleasantly parallel problems: http://en.wikipedia.org/wiki/Embarrassingly_parallel @LAFK_pl Consultant @
  61. 61. Special thanks ● Konrad Malawski and Tomek Kowalczewski, these guys really dig that stuff ● Bartosz Milewski who helped me rediscover Hans Boehm @LAFK_pl Consultant @
  62. 62. YOU! You persevered through! @LAFK_pl Consultant @

×