Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 07 Feb 2011 13:38:45 -0800
From:      Chuck Swiger <cswiger@mac.com>
To:        kellyremo <kellyremo@zoho.com>
Cc:        FreeBSD <freebsd-questions@freebsd.org>
Subject:   Re: "https is faster on amd64?"
Message-ID:  <D2810ED9-73BA-4ADC-ABA2-C9E776C3843E@mac.com>
In-Reply-To: <12dfa77447c.7163284093726025667.-8541699583891767164@zoho.com>
References:  <12dfa77447c.7163284093726025667.-8541699583891767164@zoho.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Feb 6, 2011, at 2:14 AM, kellyremo wrote:
> http://devcentral.f5.com/weblogs/macvittie/archive/2011/01/31/dispelling-the-new-ssl-myth.aspx
> 
> according to the "SSL Performance" table it says that the transactions per second is 2-3 times better using 64bit kernels opposite to 32bit kernels?
> 
> is this true, or i am just misunderstanding something

Sort of, although you're partially misunderstanding something.

SSL cryptography, like most highly math-intensive workloads, can be performed more efficiently in 64-bit architectures than in 32-bit architectures.  However, SSL is almost always done in userland processes and not in the kernel.  You don't have to run a 64-bit kernel to run 64-bit userland apps (on some systems, anyway), much as 64-bit kernels support running 32-bit processes as well as running 64-bit userland processes.

Regards,
-- 
-Chuck




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?D2810ED9-73BA-4ADC-ABA2-C9E776C3843E>