[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: MD5 vs. SHA-1, Performance & Pedigree

To: Bart Preneel <Bart.Preneel@esat.kuleuven.ac.be>
Subject: Re: MD5 vs. SHA-1, Performance & Pedigree
From: Masataka Ohta <mohta@necom830.hpcl.titech.ac.jp>
Date: Thu, 13 Jun 96 12:24:26 JST
Cc: baldwin@rsa.com, ipsec@TIS.COM
In-Reply-To: <Pine.HPP.3.91.960611125847.19679b-100000@domus.esat.kuleuven.ac.be>; from "Bart Preneel" at Jun 11, 96 1:08 pm
Sender: ipsec-approval@neptune.tis.com

> >                 Digest Performance in MegaBytes per Second
> > 
> >           Pentium P5     Power Mac    SPARC 4     DEC Alpha
> >             90 MHz        80 MHz      110 MHz      200 MHz
> > 
> > MD5         13.1          3.1         5.1          8.5
> > 
> > SHA1         2.5          1.2         2.0          3.3

>   Performance in Megabytes per Second on a 90 MHz Pentium
> 
>      MD4    MD5   SHA-1   RIPEMD RIPEMD-128 RIPEMD-160
> 
>     20.9   14.2    6.1     10.3     8.0       5.0

A problem is that a lot of computation time of a generic
hash functions on byte streams is comsumed for adjusting word
alignment, which can be avoided on hosts and routers by adjusting
alignment of packet buffers.

Considering Amdahl's law, the actual performance difference
between MD5 and SHA-1 is unnegligibly large.

Note that Pentium derived from 8080, is good at handling byte
streams and DEC Alpha can access memory only word-by-word.

					Masataka Ohta

References:

Re: MD5 vs. SHA-1, Performance & Pedigree

From: Bart Preneel <Bart.Preneel@esat.kuleuven.ac.be>

Prev by Date: I-D ACTION:draft-ietf-ipsec-payload-00.txt
Next by Date: I-D ACTION:draft-ietf-ipsec-cdp-01.txt
Prev by thread: Re: MD5 vs. SHA-1, Performance & Pedigree
Next by thread: I-D ACTION:draft-ietf-ipsec-auth-header-00.txt
Index(es):
- Main
- Thread