hotspot heap and L1 and L2 cache misses

[Vitaly Davidovich]

Andy,

You probably want to move this thread to hotspot-gc-use and remove compiler.

TLAB is thread local at a given time (but can be multiplexed over time).
You can view TLAB stats via XX:+PrintTLAB.  There are other flags that let
you tune TLAB sizing that you can play with.

It's not clear what exactly you mean by "link and unlink" but if you're
updating references of objects with different lifetime, you may want to
turn on conditional card marking for multi threaded scenarios if you're on
a recent enough hotspot version --
http://mail.openjdk.java.net/pipermail/hotspot-compiler-dev/2011-March/005056.html

Sent from my phone
On Sep 26, 2012 12:40 PM, "Andy Nuss" <andrew_nuss at yahoo.com> wrote:

> I tested TLAB allocations in single threaded microbenchmark, and when no
> GC was involved, it seems like it was about 5 nanos overhead to create a
> small object.  That is plenty fast enough.
>
> However, now I'm wondering about my chained objects.  My long running
> execution function unlinks and relinks many types of chains.  The question
> is, how strong is the guarantee of co-location with a thread, i.e. when
> many Java threads are calling this execution function that iteratively
> creates small objects per thread.  (NOTE: simultaneous calls of the
> execution function do not share objects in any way).  I.e. is TLAB a
> threadlocal approach that uses a reasonable sized block of known free
> memory for each thread?
>
>   ------------------------------
> *From:* Christian Thalinger <christian.thalinger at oracle.com>
> *To:* Andy Nuss <andrew_nuss at yahoo.com>
> *Cc:* hotspot <hotspot-compiler-dev at openjdk.java.net>
> *Sent:* Monday, September 17, 2012 11:39 AM
> *Subject:* Re: hotspot heap and L1 and L2 cache misses
>
>
> On Sep 15, 2012, at 12:03 PM, Andy Nuss <andrew_nuss at yahoo.com> wrote:
>
> > Hi,
> >
> > Lets say I have a function which mutates a finite automata.  It creates
> lots of small objects (my own link and double-link structures).  It also
> does a lot of puts in my own maps.  The objects and maps in turn have
> references to arrays and some immutable objects.
> >
> > My question is, all these arrays and objects created in one function
> that has to do a ton of construction, are there any things to watchout for
> so that hotspot will try to create all the objects in this one
> function/thread colocated on the heap so that L1/L2 cache misses are
> reduced when the finite automata is executed against data?
> >
> > Ideally, someone could tell me that when my class constructors in turn
> creates new instances of other various size other objects and arrays, they
> are all colocated on the heap.
> >
> > Ideally, someone could tell me that when I have a looping function that
> creates alot of very small Linked List objects in succession, again they
> are colocated.
> >
> > In general, how does hotspot try with creating new objects to help the
> L1/L2 caches?
> >
> > By the way, I did a test port of my automata to C++ where for objects
> like the above, I had big memory chunks that my inplace constructors just
> subdivided the memory chunk that it owned so that all the subobjects were
> absolutely as colocated as possible.
> >
> > This C++ ported automata out-performed my java version by 5x in
> execution against data.  And in cases where I tested the performance of
> construction-time cost of the automata where the comparison is between the
> hotspot new, versus my simple inplace C++ member functions which basically
> just return the current chunk cursor, after calculating the size of the
> object, and updating the chunk cursor to point beyond the new size, in
> those cases I saw 25x performance differences (5 yrs ago).
>
> TLAB allocations do the same pointer-bump in HotSpot.  Do the 5x really
> come from co-located data?  Did you measure it?  And maybe you should redo
> your 25x experiment.  5 years is a long time...
>
> -- Chris
>
> >
> > Andy
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/hotspot-compiler-dev/attachments/20120926/c7f4462a/attachment.html