From: "Bruno Luong" <b.luong@fogale.findmycountry>
Newsgroups: comp.soft-sys.matlab
Subject: Re: Derangement: efficient full permutation
Date: Fri, 18 Feb 2011 12:03:03 +0000 (UTC)
Organization: FOGALE nanotech
Lines: 9
Message-ID: <ijln5n$o67$>
References: <ii8ut8$3fb$> <ii90lg$qmd$> <> <iin0v4$ic2$> <iin5l5$h5n$> <iin77n$rje$> <iio9b0$m0i$> <ijk9se$gcb$> <ijlldg$181$>
Reply-To: "Bruno Luong" <b.luong@fogale.findmycountry>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
X-Trace: 1298030583 24775 (18 Feb 2011 12:03:03 GMT)
NNTP-Posting-Date: Fri, 18 Feb 2011 12:03:03 +0000 (UTC)
X-Newsreader: MATLAB Central Newsreader 390839
Xref: comp.soft-sys.matlab:710874

"Derek O'Connor" wrote in message <ijlldg$181$>...

> Also, remember that Roger's method uses at least twice as much memory as GRDrej and GRDsim, which becomes a problem for n > 10^8, with 16GB ram.

I just take a closer look of Roger's algorithm, and it looks like it is doable to work with a single array (in a C Mex coding). It is also possible to plug a O(N) shuffle engine for it.