English version
Accueil     À propos     Téléchargement     Ressources     Contactez-nous    

Ce site est rarement mis à jour. Pour les informations les plus récentes, rendez-vous sur le nouveau site OCaml à l'adresse ocaml.org.

Browse thread
Re: Why OCaml sucks
[ Home ] [ Index: by date | by threads ]
[ Search: ]

[ Message by date: previous | next ] [ Message in thread: previous | next ] [ Thread: previous | next ]
Date: 2008-05-09 (13:01)
From: Ulf Wiger (TN/EAB) <ulf.wiger@e...>
Subject: Re: [Caml-list] Re: Why OCaml sucks
Jon Harrop skrev:
> . Parallelism is for performance and performance
 >   requires mutable data structures.

I disagree. SMP Erlang has no mutable data structures,
but achieves very good scalability anyway.


Performance implies different things in different
applications, of course. In some cases, I'm sure mutable
data structures might seem unavoidable. But you can't
equate them with performance.

The following quote from a recent post on the ErlyWeb list
is an example of both multicore scalability and some
wicked performance - without mutable data structures:

"Under some pretty extreme loads - around 20,000 open
mainframe connections - I was blowing up Erlyweb/Yaws.
As a sanity check, this was when Erlyweb/Yaws was
consuming ~90% of all 4 CPUs on a dedicated newish
Dell server running Ubuntu 7.10 server; there was
probably smoke coming out of the box at the time ;->

On occasion, I was also blowing up CICS on the mainframe
(a Z9 series, so pretty new gear) with workload supplied
by this single Erlyweb box on occasion, so maybe we were
getting into some weird sort of temporal, black-hole type
situation where the universe was starting to fold in on
itself ;->  There's nothing like applying extreme workloads
to show up some really strange problems. " (David Mitchell)


Ulf W