[erlang-questions] ring benchmark results: Is Erlang crazy fast?

Johannes dajo.mail@REDACTED
Tue Jul 24 01:02:55 CEST 2007

i checkd it with joes code and got this results:
Setup : 0.181000 s (11.3132 us per spawn) (15999 spawns)
Run   : 1.35300 s (1.34226 us per msg) (1007999 msgs)

and i tried with ecc/elink:
j_schn14@REDACTED:~/tmp/erlang/ring$ ecc zog.erl
j_schn14@REDACTED:~/tmp/erlang/ring$ elink zog.beam
j_schn14@REDACTED:~/tmp/erlang/ring$ !.
real    0m0.386s
user    0m0.164s
sys     0m0.020s

why does this happen ?? why is there a different result if i use
ecc&elink or if i use erl ??


Peter Donald schrieb:
> Hi,
> On 7/22/07, David Cabana <dcabana@REDACTED> wrote:
>> Joe Armstrong proposed an exercise in Programming Erlang :
>> "Write a ring benchmark. Create N processes in a ring. Send a message
>> round the ring M times. So that a total of N * M messages get sent.
>> Time how long this takes for different values of N and M."
> This is probably the same benchmark as described in the following
> report (that also has source code).
> http://www.sics.se/~joe/ericsson/du98024.html

More information about the erlang-questions mailing list