Re: Switch from SBCL to Erlang backend due to scalability issues(GC).



Matthew Swank wrote:
http://lambda-the-ultimate.org/node/2102
This seems to making the rounds; though Lisp is still used as a source
language then compiled to Erlang.


From the developers comments ....

<quote>
The "benchmark code" is difficult to isolate. It's not like we're computing fibbonaci sequences here... there has to be a stream of events coming in from real-world users. SBCL starts to use an insane amount of memory, and the footprint grows even though the set of reachable objects (essentially we use the interal functions (room) calls to walk the objects, and the output of (room) itself) doesn't get much bigger. It's been a while since we've run any numbers on this problem so I don't have any handy.

Clearly something changed with the garbage collector in more recent versions of SBCL -- now SBCL uses 100% CPU all the time in production and lasts much longer before finally bombing due to exhaustion of its pre-committed space.

And we aren't using OS-level threads at all.
By a1k0n at Mon, 2007-03-05 17:39 | login or register to post comments
</quote>

The question is... what are the SBCL group going to do about it? Sounds like
they have messed up (maybe by tinkering too much?). Screwed the pooch. Of course
the Vendetta developers should have known better then use beta software. It is
getting tiring that computer languages are getting bad reps because of human
screw-ups. Just Fix It!!!

Wade
.



Relevant Pages

  • Re: Switch from SBCL to Erlang backend due to scalability issues(GC).
    ... language then compiled to Erlang. ... SBCL starts to use an insane amount of memory, and the footprint grows even though the set of reachable objects calls to walk the objects, and the output of itself) doesn't get much bigger. ... the Vendetta developers should have known better then use beta software. ...
    (comp.lang.lisp)
  • Re: Python and Lisp Test
    ... > For the record, your python version: ... > This is SBCL 0.9.4, an implementation of ANSI Common Lisp. ... If so, then get a language ...
    (comp.lang.lisp)
  • Re: scripting in CL -- please comment
    ... got a significant amount of SBCL code working in our Legato Networker ... are much more daunting in Perl than CL. ... denominator approach doesn't work very well in ANY language. ...
    (comp.lang.lisp)
  • Re: nyc lisp conf: WAY too expensive! (4me) How to *turn-away* lisp-interest!
    ... I'm pretty sure that one or more of CLISP, CMUCL, SBCL will run on ... It's interesting that you've picked up the impression that CMUCL has ... > interested in the language, its new directions, clever ... http://www.cliki.net/ - Link farm for free CL-on-Unix resources ...
    (comp.lang.lisp)