Concurrency for dummies
1 like1,561 views
1) Concurrency allows multiple computations to execute simultaneously and potentially interact with each other. 2) Concurrent applications are important for business processes that can benefit from parallel processing like e-commerce and logistics. 3) GPars is a Groovy concurrency library that supports models like actors and dataflow.
Concurrency for dummies
- 1. Concurrency for dummies and for real people too By Azrul Hasni Mad Isa (azrul.madisa@my.experian.com)
- 2. About me • Senior Software Architect at Experian Decision Analytics • 10 years of experience in IT and such • Graduated from France (INSA)
- 3. About Experian • In Cyberjaya – Other R&D Centres in Nottingham, Monaco, Washington • Financial analytics, market analytics • 200+ people and growing • We use Scrum • We are looking for people – Java, PHP, Perl, C# – Dev, QA, Architects, Team lead
- 4. Introduction • What is concurrency Concurrency is a property of systems in which several computations are executing simultaneously, and potentially interacting with each other -- Wikipedia
- 5. Introduction • Why concurrent applications are important
- 6. Introduction • Our business processes are getting more complex!
- 7. Business case for concurrency • E-commerce Buy Product Online User Parallel Split Charge Credit Card “Your request is being process” Check availability Reserve product Email confirmation
- 8. Business case for concurrency • Financial – credit verification CCRIS Apply Credit Card Credential Verification Equifax User Decision making Experian Approve/Not
- 9. Business case for concurrency • Logistics Arrival of merchandises For every merchandise Parallel Split Custom & Port Warehousing Quality check … authority Transportation
- 10. Business case for concurrency • Many others …
- 11. Model for concurrency management • Actors • Agents • Dataflow • Stm • Data paralellism – Concurrent collections – Asynchronous functions (closure) – Fork/Join
- 12. Model for concurrency management • Actors • Agents • Dataflow • Stm • Data paralellism – Concurrent collections – Asynchronous functions (closure) – Fork/Join
- 13. GPARS
- 14. GPars • Gpars is a concurrency management library for Groovy and Java • Allow running models presented above • http://gpars.codehaus.org/
- 15. ACTORS
- 16. Actors • Why couldn’t they call ‘em actresses instead? • A body of code that: – Receive a message asynchronously – React to a message if it can – Queue a received message if it’s busy – Asynchronous messaging (JMS)
- 17. Actor • Synchronous messaging = immediate reply The Nora Elena actress is soo cute! Why You !! You !!... I’ll kill you!!
- 18. Actor • Asynchronous messaging = non-immediate reply – We can do other things while waiting Could you please make me a drink? … I’m going to ponder about the meaning of life in the meantime
- 19. Actor • Asynchronous messaging = non-immediate reply – We can do other things while waiting Could you please make me a drink? … I’m going to ponder about the meaning of 2 hours later life in the meantime I was watching TV. Go make the drink yourself!
- 20. Actors Queue Queue “sending a OneActor message” AnotherActor
- 21. Actors Queue “reply to a Queue message” OneActor AnotherActor
- 22. Actors • Example: DefaultActor { class MyActor extends Actor anotherActor void act() { loop { anotherActor.send “Hello” //send message to someone react { String message -> //message could be any POGO …//do something reply someReply //some reply is any POGO … terminate() //kill the actor *gasp!* } } } } }
- 23. Actors • Two types – Stateless : DynamicDispatchActor – Stateful: DefaultActor • Actors are can share a common threadpool • Threadpool + statelessness = SCALABILITY • No remote actors yet
- 24. Actor • Some methods – loop • Infinite loop on the actor – react • React on message received – send • Send a message – reply • Reply to a message
- 25. Actor • Cool stuff – Loop counting – Code in actors are guaranteed to be thread safe – Life cycle methods – Dealing with unsent messages • Business Process Compensation – Remote actors – coming
- 26. DATAFLOW
- 27. Dataflow • First introduced in a language called Oz – Composed of tasks – Each task has input and output – A task is run as soon as all its input are available
- 28. Dataflow Not executed Task print z Task z = a+b Task b=32 Task a=20
- 29. Dataflow Not executed First round Executed Task Task print z print z Task Task z = a+b z = a+b Task Task b=32 b=32 No input needed Task Task a=20 a=20
- 30. Dataflow Not executed First round Second round Executed Task Task Task print z print z print z Task Task Task z = a+b z = a+b z = a+b Input, a and b, Task Task Task becomes b=32 b=32 b=32 available Task Task Task a=20 a=20 a=20
- 31. Dataflow Not executed First round Second round Third round Task Task Task Task print z print z Input,print z z, print z becomes Task Task available Task Task z = a+b z = a+b z = a+b z = a+b Task Task Task Task b=32 b=32 b=32 b=32 Task Task Task Task a=20 a=20 a=20 a=20
- 32. Dataflow: code final def x = new DataFlowVariable() final def y = new DataFlowVariable() final def z = new DataFlowVariable() task { z << x.val + y.val } task { x << 10 } task { y << 5 } println "Result: ${z.val}"
- 33. Dataflow: principles • Create a dataflow variable – The variable can be written once only • Read it – If it has no value yet, the program will wait for the new value while executing other tasks • Write to it – The moment a variable is written (or bounded) any task waiting for this variable will be reactivated
- 34. Dataflow: beauty • “Declarative” concurrency • Clean code – Business logic is much more obvious – Business logic can be sequential • Flow is deterministic • If no deadlock in unit test, it is guaranteed that there will be no deadlocks in production.
- 35. Dataflow in business process Apply Credit Card Parallel split CCRIS Equifax Experian Parallel join Decision making Approve/Not
- 36. Dataflow in business process Apply Credit Card Do business Parallel split users really think in CCRIS Equifax Experian Parallel executions? Parallel join Decision making Approve/Not
- 37. Dataflow in business process I want to verify user credentials against CCRIS, Experian and You need parallel Equifax processing Business Solution architect stakeholder
- 38. Dataflow in business process Suggestion by the solution architect I want to verify user credentials against CCRIS, Experian and You need parallel Equifax processing Business Solution architect stakeholder
- 39. Dataflow in business process • Meanwhile, in a parallel universe … pun intended ☺ I want to verify user Dataflow Tasks OK? credentials against Apply Credit Card CCRIS, Experian and Equifax CCRIS Equifax Experian Decision making Approve/Not Solution Business architect stakeholder
- 40. Dataflow in business process • Meanwhile, in a parallel universe Parallel execution is done Hey, I totally get that! automatically I totally want to give by GPars you more business! Dataflow Would you like a pony too? Business Solution architect stakeholder
- 41. Dataflow in business process • Meanwhile, in a parallel universe Yes please Solution architect
- 42. Conclusion • Actor – Asynchronous messaging • Dataflow – “Declarative” concurrency
- 43. Conclusion • What we see is merely the tip of the iceberg • Explore yourself – http://gpars.codehaus.org/ – http://groovy.codehaus.org/ – http://en.wikipedia.org/wiki/Acto r_model – http://en.wikipedia.org/wiki/Data flow • My contact: azrul.madisa@my.experian.com