Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Reply for "stata9 vs stata8 memory issues"

From   Alan Neustadtl <[email protected]>
To   [email protected]
Subject   Re: st: Reply for "stata9 vs stata8 memory issues"
Date   Thu, 14 Jul 2005 11:48:35 -0400

I just upgraded from V8 Intercooled to V9 SE and was hoping to have
better control with datasets with a large number of variables.  But,
because I moved in two directions (V8->v9 and intercooled->SE) it is
hard to be specific about the nature of my problem.  My computer has
1g memory and the dataset is about 500m (if I can read it and comptess
it should end up around 250m).  I can't allocate enough memory to read
the original dataset.  After the hotfix I am unable to allocate as
much as before.  I use WinXP and set a restore point so I was able to
rollback the system.

My solution was to break the dataset into three smaller datasets,
compress each one, and then append them.

Regardless, after spending lots of $s to upgrade to SE and about 1.5
hours on the phone with MS, I was not thrilled.


On 7/14/05, Richard Williams <[email protected]> wrote:
> At 10:06 AM 7/14/2005 -0400, Alan Neustadtl wrote:
> >I went through a lot of rigmarole with MS to get this hotfix last
> >week.   It seemed to make memory management issues with Stata worse on
> >my computer (e.g. less memory available through "set mem" and
> >noticeably slower run times).  Hopefully, SP3 will be somewhat better.
> OK, now you've got me worried, and since I'm getting a new computer with
> 4GB ram this issue suddenly has relevance to me.  If I have a monster file,
> am I better off analyzing it with 8.2 if I don't need 9.0's features for
> it?  I'm reluctant to install a patch if, overall, it is going to make
> things worse and/or isn't really necessary.  It also seems a bit peculiar
> that MS withdrew it because it is going to be part of SP3 - I am sure there
> is all sorts of stuff that will be part of SP3 that is available as
> individual patches in the meantime.  Maybe they are hiding it precisely
> because it is sub-optimal in its current form.
> -------------------------------------------
> Richard Williams, Notre Dame Dept of Sociology
> OFFICE: (574)631-6668, (574)631-6463
> FAX:    (574)288-4373
> HOME:   (574)289-5227
> EMAIL:  [email protected]
> WWW (personal):
> WWW (department):
> *
> *   For searches and help try:
> *
> *
> *

*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index