[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <2137066743.909185.1375862468657.open-xchange@email.1and1.com>
Date: Wed, 7 Aug 2013 03:01:08 -0500 (CDT)
From: Steve Thomas <steve@...tu.com>
To: discussions@...sword-hashing.net
Subject: Re: [PHC] A (naively?) simple PHC submission using hash chains
>
I should say I'm mostly thinking of the case of a website/server login.
> On August 7, 2013 at 2:44 AM Tony Arcieri <tony.arcieri@...il.com> wrote:
> On Wed, Aug 7, 2013 at 12:39 AM, Steve Thomas <steve@...tu.com
> <mailto:steve@...tu.com> >wrote:
> > > You shouldn't rely on threading as it can cause problems
> > >
> What problems?
>
> > > and extra resources.
> > >
> Like multiple CPU cores?
>
Creating/destroying threads is costly (time and memory). You can avoid some of
this overhead if you have a service/daemon running and take requests.
>
> > > If anything it should be nice for vector instructions (SSE2, AVX,
> > > AVX2, AVX512)
> > >
> How would these help leverage multicore CPUs?
>
It won't but the multiple concurrent requests to login will.
>
> > > Anyone know if the AES-NI run in constant time?
> > >
> It'd be pretty facepalm if it wasn't. That said AES-NI would be an
> interesting way to get speedups on x86 CPUs at least
>
I agree but I don't like to assume.
Content of type "text/html" skipped
Powered by blists - more mailing lists