[daip] [forwarded from Jeff Pedelty] My recent RH 7.0 experience and preliminary AM(00) for Pentium
Patrick P. Murphy
pmurphy at NRAO.EDU
Mon Feb 26 12:32:22 EST 2001
I think Eric got this, but wanted others to see it. Some impressive
results!
------- start of forwarded message (RFC 934 encapsulation) -------
Content-Length: 2237
From: Jeff Pedelty <pedelty at ltpmail.gsfc.nasa.gov>
To: pmurphy at NRAO.EDU
Subject: My recent RH 7.0 experience and preliminary AM(00) for Pentium
Date: Mon, 26 Feb 2001 12:14:13 -0500
Message-Id: <3.0.3.32.20010226121413.0379a370 at ltpmail.gsfc.nasa.gov>
Pat - used baboon on my first sending, so am resending to pmurphy at nrao.edu.
>Eric, Pat,
>
>I wanted to send this just to you first as I had a couple of uncertainties.
> I just got a Dell Precision Workstation 330 with a 1.5GHz Pentium 4
>processor. I have 384MB memory and 3 of the 15K RPM 18GB drives. I
>striped 2 of the drives to make /usr and the other is /data. I have AIPS
>disk 1 on /usr, disk 2 on /data, disk 3 back on /usr, and disk 4 back on
>/data. The GSA price was ~$4800 (no monitor).
>
>I installed RH 7.0 out of the box, but had to get the latest openssh (as
>listed in bugzilla) to work in our environment. I then did 'rpm -e' on all
>the packages that involved gcc-2.96 (don't have the list in my notes). I
>then installed the gcc-2.95 packages from the Red Hat support pages. All
>of 31DEC00 then built on /usr/local/aips without reported error, and I
>think the Y2K results are OK, but I wanted you to look them over:
>
>'IMAGE ' 12.891 15.5342
>'IMBEAM' 14.299 16.5649
>'APCLN ' 8.6961 16.0761
>'APRES ' 15.3931 22.8872
>'IMCLN ' 8.6849 16.226
>'VTESS ' 10.3478 17.9358
>'VTESSC' 11.8384 17.1775
>'IMAGE ' 0.00012 -0.000061 -0.000086
>'IMBEAM' 0 0 0.00000006
>'APCLN ' 0.00012 0.00037 0.0176
>'APRES ' -0.0000057 0 -2.98023E-08
>'IMCLN ' 0.00055 -0.00061 0.0173
>'VTESS ' -0.00049 0.00031 0.000036
>'VTESSC' 0.000061 0 -0.00002
>
>I left FDEFAULT.SH as is. I set the AP memory size to 80MB. BTW, the
>build took a bit over half an hour. I didn't take notes, but the timestamp
>on AIPSUB.LIS is 20:37, INSTEP2.LOG is 20:54, and INSTEP4.LOG is 21:09. I
>seem to recall it took a bit longer in ~'83 for our first 11/780
>installation at Minnesota...
>
>Y2K runs in the range 724 to 727 seconds, or AM(00)=55.0 to 55.2. However,
>DDT takes 126 seconds, or AM(93)=31.7. Is this consistent with your
>expectations? I'd be happy to try other tests, AIPS data layouts, etc., or
>could send you any rpm -q listings. I could also set up an NRAO account as
>I did years ago, if you want to play.
>
>Jeff
>
------- end -------
More information about the Daip
mailing list