HN Books @HNBooksMonth

> 2nd edition but I can't find it

Perhaps you're referring to this one [1]?

[1] https://www.amazon.co.uk/Garbage-Collection-Handbook-Managem...

⬐ tempguy9999

Thanks, I think I was confusing that with the 2nd ed, which doesn't seem to exist (in fact there is a 1999 2nd printing of the 1996 book with a different cover but with the same ISBN, just to make things worse, https://www.cs.kent.ac.uk/people/staff/rej/gc.html, so if you want to buy the later one which incorporate the errata from the first printing, check his website first. Publishers do make a mess of things sometimes).

Jones' website https://www.cs.kent.ac.uk/people/staff/rej/gc.html is excellent too.

One indication is that Azul, after 3 generations of ccNUMA systems with zillions of custom chips and a memory infrastructure that gives each one "mediocre" access speed to all the system's memory for running Java with gonzo GC ("generic" 64 bit RISC chips with things like a custom read barrier instruction), has given up and is doing their thing on x86-64 systems with their Zing product, albeit at least initially with tricks like kernel extensions to do bulk MMU operations before a single TLB invalidation. Look up their papers on the Pauseless and C4 GCs. The former was done in time to make it into the 2nd edition of sorts of the book on GC: http://www.amazon.com/Garbage-Collection-Handbook-Management...

Or to put it another way, without exhausting my bank account I could build from parts I can purchase today on Newegg a many CPUs 3/4ths TiB DRAM Supermicro system. Supermicro has standard boards with more memory, and has a monster you can only buy complete that'll hold 4 CPU chips and up to 6 TiB DRAM on daughter boards; I think based on some Googling that has a starting price of less than $35K.

Moore's Law is our friend. But its economics is not the friend of custom CPUs in competition with commodity ones.

The bibliography stops in 2002.

Here's a 2011 book that's sort of an advanced, vs. updated version of it's 1996 predecessor: http://www.amazon.com/The-Garbage-Collection-Handbook-Manage... It covers the recent state of the art up to it's publication (e.g. Azul's Pauseless but not C4).

No, the older version of their GC used bulk VM operations, but not virtualization features, and there was still a penalty reported ... errr, I can't find it now. Probably in a Clifford Click blog posting. I just skimmed the new edition of Jones' GC book (http://www.amazon.com/gp/product/1420082795/), published before it could consider the newer one; it talks about the changes needed on stock hardware but I didn't see any estimation of costs while glancing through it. (I'm not searching any more right now because it's obsolete.)

The base papers are:

Pauseless GC, uses a read barrier instruction in their custom 64 bit RISC chips: https://www.usenix.org/legacy/events/vee05/full_papers/p46-c...

And the newer one that they're using in that old hardware and the software on commodity hardware Zing JVM, the Continuously Concurrent Compacting Collector (C4), which I have not studied (the paper was published 2 weeks after the Joplin tornado trashed my apartment and rather disrupted my life): http://www.azulsystems.com/sites/default/files/images/c4_pap...

It's possible they figured out how to minimize or eliminate the penalties of the original software read barrier they applied to the Pauseless system in C4 (or perhaps in relation to their custom vs. newer commodity hardware); I just did a quick skim of the relevant part of the C4 paper and a few keywords and couldn't tell.

This is all great stuff that I hope to get back to soon....

Which I would add is a production garbage collector, an updated version of the Pauseless one that's got its own paper and is covered in the 2nd edition of the book on garbage collection, http://www.amazon.com/The-Garbage-Collection-Handbook-Manage...

Does require custom hardware or kernel cooperation for speed (e.g. it needs to do batched MMU operations without clearing the TBL on each 2MB page). Looks like it's got a better read barrier than the Pauseless one; that does of course cost extra on stock hardware.

Wilson's highly readable "Uniprocessor Garbage Collection Techniques" is a far better overview: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.138.... You could write a decent GC with just the info in that paper.

It doesn't cover parallel or real-time garbage collection (which both get far more complicated); for those, you want Jones et. al's _The Garbage Collection Handbook_ (http://www.amazon.com/The-Garbage-Collection-Handbook-Manage...) and plenty of time to explore its bibliography. (His older book is also good, but doesn't cover those topics.)

⬐ mattyb

That is a good paper. Thanks for the pointer!