JVM Anatomy Park

by r4um • 9 years ago

356 32 9 years ago

kondor6c• 9 years ago

I really enjoy the completeness of this publication, additionally the ability for me to read it later from an epub is great.

This reminds me when the Java Pub House did a few podcasts on Garbage Collections and performance (http://www.javapubhouse.com/2012/04/episode-22-garbage-man-i...)

jonahx• 9 years ago

Agreed. I would love to see this become the standard for longer blog posts.

Would also love details on the author's workflow and automation behind this.

SamPutnam3• 9 years ago

7k views, 155 upvotes, 1 comment...

yeukhon• 9 years ago

My take: not everyone has time to read this right now. I use upvote to "bookmark", and I expect not everyone here use Java. I am not, but I intend to give it a round at some point.

gok• 9 years ago

Half the upvotes are probably for the Rick and Morty reference.

johnfn• 9 years ago

How do you know it has 7K views?

SamPutnam3• 9 years ago

https://raw.githubusercontent.com/SamDeepLearning/The-Terrib...

bdcravens2• 9 years ago

I peek at articles all the time without upvoting, and upvote articles without commenting all the time.

baxtr2• 9 years ago

I was skeptical at first as well, but your own stats show a different picture: the site gets a lot of views, which correlates with the upvotes. No comments is probably due to the fact that people use upvoting as a mechanism to save links to material (which might be of use later)

SamPutnam3• 9 years ago

To clarify, I don't think this is improper use of the upvote mechanism, I was just hoping for more engagement with the comment feature of Hacker News, given the 7k click thrus.

zerr• 9 years ago

Why PDF is so large?

zanker_swe5• 9 years ago

Regarding the question in the 1st topic: (https://shipilev.net/jvm-anatomy-park/1-lock-coarsening-for-...)

" for (...) { synchronized (obj) { // something } }

…could it optimize into this?

synchronized (this) { for (...) { <----XXX---- // something } } "

The answer would be no in general, I think, since it is unsafe.

Moving the entire loop into the synchronized block would make the statement marked XXX above, i.e. "for (...)" execute inside the synchronized block, which has the potential to change the semantics (for example, the statement may include an rpc, and we don't want to make that rpc under the lock).

The earlier optimization of

synchronized (this) { statement1; }

synchronized (this) { statement2; }

synchronized (this) { statement1; statement2; }

is safe.

chrisseaton4• 9 years ago

Moving the entire loop into the synchronized block would make the statement marked XXX above, i.e. "for (...)" execute inside the synchronized block, which has the potential to change the semantics (for example, the statement may include an rpc, and we don't want to make that rpc under the lock).

That's not change in semantics. The JVM doesn't provide any guarantees about parallelism or scheduling semantics in the first place so how can you possibly argue about the semantics of holding a lock too long?

And it's unsafe? For what definition of safety?

But in practice - yes you obviously wouldn't do this if the code in the loop control code had any side effects.

zanker_swe5• 9 years ago

And it's unsafe? For what definition of safety?

Let me try to answer with an example. Let us say, we have original code like this:

  synchronized(this) {
   a();
   b();
 }
 c();
 synchronized(this) {
   d();
   e();
 }

It would be unsafe (in general) to transform the above code to

  synchronized(this) {
    a();
    b();
    c();
    d();
    e();
  }

Simply because Compiler does not know (again, in general) what may happen during the execution of c(). However, the following transformation is safe (the timing behavior changes, but as you point out, that is not a guarantee programmers should expect).

  synchronized(this) {
    p();
    q();
  }
  synchronized(this) {
    r();
    s();
  }

  synchronized(this) {
    p();
    q();
    r();
    s();
  }

since there is nothing happening between the two synchronized sections.

For the for loop, an example of code where pulling the synchronized statement out of the loop is problematic:

  for(a = AcquireLock(), c = 0; c < 100; c++) {
    synchronized(this) {
      f();
    }
  }
  ReleaseLock(a);

I don't think it is safe to transform to

  a = AcquireLock();
  synchronized(this) {
  for(c = 0; c < 100; c++) {
      f();
    }
  }
  ReleaseLock(a);

Perhaps you (and the original blog post) assume we are only talking about movement of code after ensuring that such movement is safe, but it was not clear from the document.

thomaslee• 9 years ago

The answer would be no in general, I think, since it is unsafe. ... for example, the statement may include an rpc, and we don't want to make that rpc under the lock

I do agree that it's not a "safe" optimization in the extreme general case (so don't go rewriting your code assuming it's equivalent!), but in the case where the loop is a candidate for unrolling it works just fine. Imagine you had a more CPU- or memory-bound workload, and these benchmarks are a whole lot more interesting.

Put another way: if there's an RPC call in the for loop, the time spent in the RPC will dwarf the work involved in executing loop itself so ... odds are good it's not going to be a candidate for unrolling anyways. :)

endoplasm2• 9 years ago

Neither of those optimizations looks safe to me. Couldn't they potentially introduce deadlocks?

adrianmonk3• 9 years ago

I had the same reaction, but now I'm starting to suspect it actually is safe, for a reason that's not immediately obvious.

Consider the unoptimized version. It's possible that after you release the lock, another thread will take it. But I'm pretty confident there is no guarantee of this, and your own thread might immediately take the lock back. (Probably often, especially if there's just one CPU.) If taking the lock back immediately causes a deadlock, then your code is already broken as written.

So you already must account for the possibility that nothing changes between the time you release the lock and take it again. All the optimization does is take away the _other_ possibility, which was something you couldn't rely on anyway.

adrianmonk3• 9 years ago

This is of no real consequence except curiosity, but I'm at a loss to understand what "park" means in the title.

I don't even know what sense of the word is intended. Is it "park" like a city or amusement park that you visit, and the idea is that you enjoy a variety of different activities while you're there? Is it "park" like a ballpark, and the analogy is that sporting events take place there and each one of these topics is an event? Is it somehow "park" like a parking lot? Maybe "park" is a too-literal translation from another language where the corresponding word has a meaning that the English one doesn't?

I'm trying to guess based on context, but none of these really seems to fit.

leodeid• 9 years ago

Maybe it is a reference to the Rick and Morty episode "Anatomy Park". http://rickandmorty.wikia.com/wiki/Anatomy_Park_(episode)

adrianmonk3• 9 years ago

Oh, that has to be it! So it's just a reference I didn't understand.

Flenser• 9 years ago

Could be as in wildlife park.

tveita• 9 years ago

Besides being interesting in itself, this is the best collection of examples I've seen for JMH. When I looked last the official documentation was a bit thin, and most articles on it just included an unmotivated hello world example.

JVM Anatomy Park

Related Stories