vgnet

There's a very interesting comparison between libumem (mtmalloc) and Hoard at http://www.oracle.com/technetwork/articles/servers-storage-d... (HN link: https://news.ycombinator.com/item?id=3755621). How do you view this class of allocators to tcmalloc, jemalloc, ptmalloc3, etc., specially in the concurrent multi-cpu scenario?

Here's what I'll be reading on allocators tonight, any further suggestions (following all the links in comments is part of the deal)?

http://en.wikipedia.org/wiki/C_dynamic_memory_allocation

http://www.facebook.com/notes/facebook-engineering/scalable-...

http://locklessinc.com/benchmarks_allocator.shtml

http://blog.reverberate.org/2009/02/20/one-malloc-to-rule-th...

See also: http://weblogs.mozillazine.org/asa/archives/2012/03/the-end-...

In the general case, Dropbox benefits from unused (but paid for) capacity, while users want a fixed price, even if they're overpaying (compared to use). For some products, pay-as-you-use seems better and more attractive, but Dropbox wants simplicity and ubiquity, so I don't think that model would be better for them.

See http://blog.dreamhost.com/2006/05/18/the-truth-about-oversel... for a related discussion.

TSA has a recent blog post[1] about how that is a myth, where people keep mentioning that it has happened before (maybe not to armed soldiers).

[1] http://blog.tsa.gov/2012/03/tsa-taking-nail-clippers-from-so...

Great way to sum it up, by that measure I'm a clear introvert. That doesn't mean shy, quiet in gatherings or even unhappy in company of others, but group time is exhausting and alone time is, as you say, recharge time.

Should I edit to "Ubuntu may break virtual terminals on some graphic cards" to better reflect content or leave the original title?

I suggest you check the Extended Support Release, tailored for your usecase: http://www.mozilla.org/en-US/firefox/organizations/

For some context:

https://blog.mozilla.com/blog/2012/01/10/delivering-a-mozill...

https://wiki.mozilla.org/Enterprise

AFAIK, Unladen was a 20% project for a couple (less than 5) Googlers, running without any corporate support beyond the 20% system.

See http://webtide.intalio.com/2012/03/jetty-spdy-is-joining-the... for a hyperbolic endorsement.

To sum it up, clients (SPDY enabled by default in FF13 and already in Chrome) and servers (Apache, nginx, Jetty, node, Google sites) are becoming available, some niches (mobile, high latency) are bound to benefit a lot and it's a good (better?) solution for the general case.

I would guess they did and that they were probably selected. It's just that the list is still incomplete (180 orgs selected, displaying less than 100 as of right now).

For the first hours, not all organizations will be listed there, as listing depends on having an up-to-date profile sent by admins. Don't fret if you don't see your org there yet: there are 180 selected organizations, compare to the number shown at the bottom of the page (88 as I write this).

Great video linked in the article (the author's PyCon keynote about PyPy).

An email thread started by the author gives some more information on targets and characteristics: https://groups.google.com/group/numpy/t/ad48cbce6002da9e

AFAIUI, what you're missing is that PyPy can (and often does) interface directly to C libraries, from RPython. So the prospect of re-implementing those specialized codebases isn't a real issue: only the CPython-API based wrappers would need re-implementing. I believe those are a small part of the total code you mention.

It suggests that there aren't many production users, true. But it does prove that there are some.

Yes, there are.

Recently Quora announced it was running on PyPy[1]. Some other disclosures were made (a Django project[2], LWN internal processing[3], tweets about speedups in production, etc.), but the PyPy team is thinking about officially asking for success stories in the near future[4].

[1]: http://www.quora.com/Alex-Gaynor/Quora-product/Quora-is-now-...

[2]: https://convore.com/python/whos-using-pypy-in-production/

[3]: http://lwn.net/Articles/442268/

[4]: https://bitbucket.org/pypy/extradoc/src/tip/blog/draft/succe...

> I remember that GSoC! Any idea as to what happened to it, > if it was scrapped, or whatever?

I know it made a lot of progress in Cython-land (updates at http://rguillebert.blogspot.com/) but no real overview or high level status update has been given yet (AFAIK). Since it's from GSoC 2011, it's about time for that to happen... and for other people to start contributing code <hint, hint> :).

It's "real" support, using your criteria. However, it's slower in PyPy than in CPython and than what the same thing based on ctypes would be on PyPy.

If there's a pure-Python version of the C extension, it might be faster on PyPy than the C extension support that cpyext (PyPy C-API compatibility module) provides.

Cython-based code is currently incompatible (it goes well beyond the public C-API), but a GSoC project to generate ctypes-based pure-Python code from Cython is (was?) going on.

PyPy also has a plan to get rid of the GIL: http://morepypy.blogspot.com/2011/06/global-interpreter-lock...

I hope PyPy pings all those sponsors. As a wise lady once said, 'connections never hurt'.