Re: Postvayler

Hi Hakan, 

This effort looks promising, congratulations! :)

You could also use the databen.ch project to test the Postvayler consistency/scalability. I'm really interested on the results.

Cheers, 

-- 
Flavio W. Brasil


On Friday, December 6, 2013 at 7:02 PM, hakan eryargi wrote:

> I believe the project has reached the proof of concept state:)
> 
> I've extended the bank sample with accounts and money transfers and made another bank sample which emulates instrumentation (ie, manually added injected code) that helped a lot to visualize what is going on. 
> 
> adding objects to pool with a reliable id was a bit tricky but I guess I've managed it with a special transaction which behaves different at regular run and recovery time. see ConstructorTransaction for some details. 
> 
> allowing garbage collection with a WeakValueMap is also done. there is also a trick here to trick GC not to garbage collect target objects before we're done. see GCPreventingPrevayler for some details.** 
> 
> I've made two tests: 
> 1. run the same sequence of thousands of operations on both persisted and non-persisted instances and checked the results for equality. this also includes closing prevayler, recovering and going on. all seems good.
> 2. run thousands of operations with many threads. no objects missed. I wish to have a multi-threaded check final result test similar to first one but I couldnt figure how.
> 
> next things:
> * scan all reachable @Persistent classes. at the moment generic parameters (like collection types) is not parsed. I'm planning to scan all packages of all reached @Persistent classes. this still does not guarantee all required classes is instrumented (some one can extend a @Persistent class in a totally different package), but there is a run time check to catch that condition
> * proper encapsulation. at the moment lots of things is public and anyone can crash the system with a simple cast and java call
> * more tests
> * maybe a better trick to prevent GC
> * cleanup and optimize compiler code
> * ant build.xml (or maven?)
> * an ant task for compiler
> * support for spring dependency injection (at least injecting root as a bean to other spring beans)
> * some documentation
> * adapt prevayler samples to postvayler
> * what else?
> 
> any feedback is welcome. please have a look at the samples (especially the emulated sample) and speak up:)
> 
> cheers, 
> r a f t
> 
> ** after very very long multi-threaded runs sometimes an object is missed (garbage collected). this happens very seldom and I suspect this is related to JVM's runtime optimization. It may be detecting that our trick to prevent GC is a noop and removing the call. 
> 
> 
> On Mon, Dec 2, 2013 at 11:57 PM, hakan eryargi <[email protected] (mailto:[email protected])> wrote:
> > runtime bytecode injection isnt a current objective. it may be possible later or not, doesnt really matter in this stage. pre-instrumented classes is good enough for now, like jibx. 
> > come on, you should have some faith in flying spagetti monster man:)
> > seriously, if someone sees a flaw, which cant be done, please speak up, so i wont spend unnecessary hours. to me all seems good for now.
> > btw, details are constantly changing and evolving. i've shared the project in a very early stage for discussion
> > r a f t
> > On Dec 2, 2013 10:52 PM, "Naveen Chawla" <[email protected] (mailto:[email protected])> wrote:
> > > Hakan, Congratulations for carving out the time to attempt this.
> > > 
> > > How easy do you think it would be to instrument the object, when it is passed in, to simply journal all changes to non-transient fields, and batch them into single journal entries when they occur during a synchronized lock over the prevalent object? 
> > > 
> > > Can it all be done with run-time instrumentation from start to finish, such that no annotations are necessary?
> > > 
> > > 
> > > On 1 December 2013 23:12, hakan eryargi <[email protected] (mailto:[email protected])> wrote:
> > > > i'm experimenting with the idea. so far not bad. just put what I've done to GitHub. any suggestions, critics and contributions are welcome:)
> > > > https://github.com/raftAtGit/Postvayler
> > > > 
> > > > apologies for the name, it sounded like a nice word play:)
> > > > 
> > > > to run the sample, run the Compiler class with raft.postvayler.samples.bank.Bank argument and then run raft.postvayler.samples.bank.Main 
> > > > 
> > > > how it works:
> > > > compiler injects bytecode to root class and all @Persistent classes that can be accesible* from root. 
> > > > 
> > > > root class contains an injected object pool, where each @Persistent class is assigned a Long id and put into this object pool when created. this pool will be* a WeakValueMap so objects which are not attached to root will be garbage collected at some time. 
> > > > 
> > > > each @Persist method in a @Persistent class is instrumented like this:
> > > > @Persist
> > > > ReturnType doSomething(someParams) {..}
> > > > 
> > > > doSomehing is copied to __postvayler__doSomething(..). the original doSomehing becomes:
> > > > 
> > > > ReturnType doSomething(someParams) {
> > > >     if (!there is postvayler context) 
> > > >         return __postvayler__doSomething(..);
> > > >     if (we are in a prevayler transaction) 
> > > >         return __postvayler__doSomething(..);
> > > >     return prevayler.execute(new MethodTransaction(..));
> > > > }
> > > > 
> > > > the arguments to doSomehing(..) is transformed in MethodTransaction such that references to @Persistent objects are replaced with Reference's, which will later be restored from object pool. seems as this also solves the Baptism problem.  
> > > > 
> > > > looks good to me for now:)
> > > > 
> > > > cheers,
> > > > r a f t
> > > > 
> > > > * not implemented yet
> > > > 
> > > > On Wed, Nov 27, 2013 at 11:07 AM, Naveen Chawla <[email protected] (mailto:[email protected])> wrote:
> > > > > Sure, here: http://docs.oracle.com/javase/tutorial/java/generics/methods.html
> > > > > 
> > > > > As for consistency, any multi-changes explicitly synchronized over prevalentObject can be auto-batched. 
> > > > > 
> > > > > 
> > > > > On 27 November 2013 00:25, Klaus Wuestefeld <[email protected] (mailto:[email protected])> wrote:
> > > > > > > It is ordinary generic method syntax.
> > > > > > 
> > > > > > Could you point me to some doc on this syntax? I couldnt find it here:
> > > > > > http://docs.oracle.com/javase/tutorial/extra/generics/methods.html
> > > > > > 
> > > > > > > The beginning and end of a transaction would simply be any change to any
> > > > > > > non-transient field.
> > > > > > 
> > > > > > Suppose a transaction needs to update two fields to be correct but the
> > > > > > system crashes after the first. Will the system not be inconsistent on
> > > > > > recovery?
> > > > > > 
> > > > > > Klaus
> > > > > > 
> > > > > > ------------------------------------------------------------------------------
> > > > > > Rapidly troubleshoot problems before they affect your business. Most IT
> > > > > > organizations don't have a clear picture of how application performance
> > > > > > affects their revenue. With AppDynamics, you get 100% visibility into your
> > > > > > Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
> > > > > > http://pubads.g.doubleclick.net/gampad/clk?id=84349351&iu=/4140/ostg.clktrk
> > > > > > _______________________________________________
> > > > > > To unsubscribe go to the end of this page: http://lists.sourceforge.net/lists/listinfo/prevayler-discussion
> > > > > > _______________________________________________
> > > > > > "Databases in Memoriam" -- http://www.prevayler.org
> > > > > 
> > > > > 
> > > > > ------------------------------------------------------------------------------
> > > > > Rapidly troubleshoot problems before they affect your business. Most IT
> > > > > organizations don't have a clear picture of how application performance
> > > > > affects their revenue. With AppDynamics, you get 100% visibility into your
> > > > > Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
> > > > > http://pubads.g.doubleclick.net/gampad/clk?id=84349351&iu=/4140/ostg.clktrk
> > > > > _______________________________________________
> > > > > To unsubscribe go to the end of this page: http://lists.sourceforge.net/lists/listinfo/prevayler-discussion
> > > > > _______________________________________________
> > > > > "Databases in Memoriam" -- http://www.prevayler.org
> > > > > 
> > > > 
> > > > 
> > > > ------------------------------------------------------------------------------
> > > > Rapidly troubleshoot problems before they affect your business. Most IT
> > > > organizations don't have a clear picture of how application performance
> > > > affects their revenue. With AppDynamics, you get 100% visibility into your
> > > > Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
> > > > http://pubads.g.doubleclick.net/gampad/clk?id=84349351&iu=/4140/ostg.clktrk
> > > > _______________________________________________
> > > > To unsubscribe go to the end of this page: http://lists.sourceforge.net/lists/listinfo/prevayler-discussion
> > > > _______________________________________________
> > > > "Databases in Memoriam" -- http://www.prevayler.org
> > > > 
> > > 
> > > 
> > > ------------------------------------------------------------------------------
> > > Rapidly troubleshoot problems before they affect your business. Most IT
> > > organizations don't have a clear picture of how application performance
> > > affects their revenue. With AppDynamics, you get 100% visibility into your
> > > Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
> > > http://pubads.g.doubleclick.net/gampad/clk?id=84349351&iu=/4140/ostg.clktrk
> > > _______________________________________________
> > > To unsubscribe go to the end of this page: http://lists.sourceforge.net/lists/listinfo/prevayler-discussion
> > > _______________________________________________
> > > "Databases in Memoriam" -- http://www.prevayler.org
> > > 
> 
> ------------------------------------------------------------------------------
> Sponsored by Intel(R) XDK 
> Develop, test and display web and hybrid apps with a single code base.
> Download it for free now!
> http://pubads.g.doubleclick.net/gampad/clk?id=111408631&iu=/4140/ostg.clktrk
> 
> _______________________________________________
> To unsubscribe go to the end of this page: http://lists.sourceforge.net/lists/listinfo/prevayler-discussion
> _______________________________________________
> "Databases in Memoriam" -- http://www.prevayler.org
> 
>

------------------------------------------------------------------------------
Sponsored by Intel(R) XDK 
Develop, test and display web and hybrid apps with a single code base.
Download it for free now!
http://pubads.g.doubleclick.net/gampad/clk?id=111408631&iu=/4140/ostg.clktrk

_______________________________________________
To unsubscribe go to the end of this page: http://lists.sourceforge.net/lists/listinfo/prevayler-discussion
_______________________________________________
"Databases in Memoriam" -- http://www.prevayler.org