s-news
[Top] [All Lists]

Re: slowness on linux

To: "S-PLUS Newsgroup \(E-mail\)" <s-news@lists.biostat.wustl.edu>
Subject: Re: slowness on linux
From: "David Smith" <dsmith@insightful.com>
Date: Wed, 21 Nov 2001 12:50:05 -0800
Cc: "Winifred Lambert" <lambert.winifred@ensco.com>
Importance: Normal
In-reply-to: <D7E1E71C0658D4118C9D00104B6D756D13A76C@amu01>
I'd guess HAL here is operating as a server machine, and you're using S-PLUS
at a remote workstation (say, Windows with an X emulator, or a separate
Linux workstation with XFree86).  If you use the S-PLUS 6 Unix/Linux GUI on
a remote system, the responsiveness of the GUI may be poor.  This is
discussed in detail in the release notes, but it's basically because the
entire GUI image is rendered on the server and then needs to be transferred
to the desktop, and moving what is essentially a large image over a network
takes time. This is a feature of Java that we don't have any control over.
The problem is mitigated somewhat by running S-PLUS in a smaller window or
at lower resolution (especially fewer color planes), but you may prefer
simply to use the command line version (better yet, with ESS).  Another
alternative is to use S-PLUS Analytic Server instead, which has all the
features of S-PLUS 6 plus local GUI clients -- with a local client the GUI
is rendered locally and is therefore much much faster.

The performance of the S computation engine -- i.e. the time to run
expressions using the command line version -- should be about the same on an
Intel machine whether it's running Windows or Linux.  If you're observing
otherwise, it may be because your .Data directory is on a remote-mounted
disk (remote from the server machine where S-PLUS is running, that is).
There can be a lot of random access to files in that folder which can slow
down S-PLUS on some networks.  I'd suggest running S-PLUS with a
locally-mounted .Data folder and see if that makes a difference.

# David Smith

--
David M Smith <dsmith@insightful.com>
S-PLUS Product Marketing Manager, Insightful Corp, Seattle WA
Tel: +1 (206) 283 8802 x360
Fax: +1 (206) 283 0347

Insightful Corporation provides analytical solutions leveraging
S-PLUS, StatServer, S-PLUS Analytic Server and consulting services.
See www.insightful.com for details.

> -----Original Message-----
> From: s-news-owner@lists.biostat.wustl.edu
> [mailto:s-news-owner@lists.biostat.wustl.edu]On Behalf Of Winifred
> Lambert
> Sent: Wednesday, November 21, 2001 08:38
> To: S-PLUS Newsgroup (E-mail)
> Subject: [S] slowness on linux
>
>
> All,
>
> I am running a trial version of S-PLUS 6 on Red Hat Linux release 7.0
> (Guinness)
> Kernel 2.2.16-35lfssmp on a 2-processor i686.  The processors are
> Pentium III 933 MHz/ 512 RAM with 2 per node, 16 nodes in all (i.e. 32
> processors).  It's a big, tall, black machine with lots of flashing
> green lights that we nicknamed 'HAL', but its system name is Thor to
> match our workstation nameing convention.  But, I digress.  Suffice to
> say, its a powerful and fast machine.  All tests of our other software
> packages indicate much faster performance.
>
> We recieved a demo version of S-PLUS 6 for Linux from Insightful because
> their HP version would not work on our UNIX workstations.  The S-PLUS
> GUI and command lines seem to work on the Linux system, but are
> *excrutiatingly* slow.  I am starting a project in which I will be
> working with 7 years of data collected every 5 minutes at 52 locations
> and at several heights at each location.  In other words, the data files
> are huge.
>
> As a test, I imported one 5 MB file into a data frame consisting of 15
> columns and 62,900 rows.  It took over 2 minutes to import, and when I
> tried to look at it through the new data viewer S-Plus locked up and I
> had to end the program through the Task Manager.  It typed the data
> frame name on the command line to view it that way, and it took over 3
> minutes to output it.  I can do calculations and other manipulations to
> the data frame, but they take a long time.
>
> I cross-checked this with MS Excel and S-PLUS 2000 on my PC (400 MHz,
> Windows 2000, 384 RAM - less powerful than even 1 processor on our Linux
> system).  It took less than a minute to import the file into an Excel
> spreadsheet, and even less time to import it into an S-PLUS dataframe.
> The data were displayed *very* quickly when I typed the name of the
> dataframe at the command line, and manipulation of the data in both
> packages was quicker than on the Linux system.
>
> I checked the object.size on the Linux version and it is set to Inf.
> Are there any other memory things I should check?  Has anyone else
> experienced this on Linux?  ANY advice is appreciated.  The Linux
> version *should* be running circles around the PC version given all the
> computing power available to it.
>
> Win
>
> *************************************************
> Winifred C. Lambert            Senior Scientist/Meteorologist
> ENSCO, Inc.
> Aerospace Sciences and Engineering Division
> 1980 N Atlantic Ave, Suite 230
> Cocoa Beach, FL  32931
> VOICE:  321.853.8130           FAX:  321.853.8415
> lambert.winifred@ensco.com
>
> AMU Quarterly Reports are available online:
> http://science.ksc.nasa.gov/amu/home.html
> *************************************************
>
> ---------------------------------------------------------------------
> This message was distributed by s-news@lists.biostat.wustl.edu.  To
> unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu with
> the BODY of the message:  unsubscribe s-news


<Prev in Thread] Current Thread [Next in Thread>