lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  PHC 
Open Source and information security mailing list archives
 
Hash Suite for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Wed, 12 Feb 2014 01:23:39 +0000
From: Marsh Ray <maray@...rosoft.com>
To: "discussions@...sword-hashing.net" <discussions@...sword-hashing.net>
Subject: RE: [PHC] Is bandwidth all that counts?

From: Peter Maxwell [mailto:peter@...icient.co.uk]
> My problem is understanding why memory bandwidth is the critical
> factor - is memory bandwidth inherently more expensive than the actual DRAM?

It’s something that depends on MACROscopic hardware, which imposes physical limits to growth. It grows much slower that Moore’s law.

Yet commodity hardware comes very highly optimized for memory bandwidth, so in effect defenders collectively participate in a big economy of scale. An attacker may be able to do get a better price/performance with custom hardware, but probably not by much (on the log2 scale).

In short, we can expect that an attacker could fit 512 parallel instances of MD5 on a custom silicon chip. But it’s probably also a reasonable assumption that no one has an external RAM subsystem that’s 512 times more efficient than the latest DDR chips.

- Marsh
----------------------------
My own opinions, usual disclaimers apply.

Content of type "text/html" skipped

Powered by blists - more mailing lists