Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: 90GB table on Windows 2000

Re: 90GB table on Windows 2000

From: Jim Kennedy <kennedy-family_at_attbi.com>
Date: Fri, 11 Oct 2002 11:59:16 GMT
Message-ID: <o2zp9.961$gr6.1638@rwcrnsc53>


I remember a local hospital system that moved their data center. So naturally they had to turn the Unix (hpux) machines off to do the move. They had trouble getting all their services and so forth running again because some of those boxes had been running for years and people had kicked off demons and forgotten to put them in the start up scripts. (and those people had left.)

It really depends upon how much stability you need. Most people don't need stability like the system that tracks the NASDQ (in the US the National Association of Securities Dealers Quotation system). Consists of parallel server at 2 locations (Connecticut and Maryland) that replicate to each other every 5 seconds. (Unix)

But personally, given the different software philosophies of MS on the one hand and Sun (for example) on the other. MS is more graphically, consumer oriented, less concerned with stability and uptime. An MS network admin is quicker to reboot a server if something funny is happening (and it usually fixes it) than a UNIX admin. Think of it another way. No one calls (or at least not with any success) MS if their MS Server goes down. On the other hand, if your mainframe goes down for a couple of minutes (unplanned) you are on the phone asking the vendor why. Different philosophies. Jim

"Howard J. Rogers" <howardjr2000_at_yahoo.com.au> wrote in message news:bBvp9.50692$g9.147169_at_newsfeeds.bigpond.com...
>
> "Niall Litchfield" <n-litchfield_at_audit-commission.gov.uk> wrote in message
> news:3da68457$0$1292$ed9e5944_at_reading.news.pipex.net...
> > "Jim Stern" <jdstern_at_k2services.com> wrote in message
> > news:ao52s5$fdl$1_at_news.utelfla.com...
> > > Damn right... NT is definitely a no no for real critical applications,
> too
> > > many variables for the O/S to be stable.
> >
> > Are you sure this isn't just a knee jerk response. We have 2000 servers
> > running line of business apps that have been up for (checks uptime) 118
> > days. The downtime was for power work. Setup correctly windows 2000 *is*
a
> > stable OS. Now I'm not suggesting that we move apps supporting 10,000
> users
> > across to win2k from Solaris or HPUX or whatever, but windows is now a
> > perfectly satisfactory server operating system. Moreover in this
> particular
> > case the application appears to be a single table app storing logs or
> stats
> > or some such.
> >
>
> Yup. Frankly, I'm getting bored with the "Micro$oft" and "Windoze" crap.
>
> My home server (2000) has been up for 221 days. It would have been more,
> but the previous 119 days were interrupted by a thunderstorm knocking out
> the power. One UPS later, and all is well. Even my XP desktop, which
gets
> punishment aplenty and all manner of freeware installs, deinstalls and
> anything in between, has been up for 54 days.
>
> This myth that Windows is unstable is just that: myth. These days. But
> even so, I remember my NT4 servers at a large Insurance company running
for
> three months at a time without interruption, except when I wanted to
bounce
> them for service pack or backup issues.
>
> I blame the installers. Not the software.
>
> Regards
> HJR
>
>
>
>
> >
> > --
> > Niall Litchfield
> > Oracle DBA
> > Audit Commission UK
> > *****************************************
> > Please include version and platform
> > and SQL where applicable
> > It makes life easier and increases the
> > likelihood of a good answer
> >
> > ******************************************
> >
> >
>
>
Received on Fri Oct 11 2002 - 06:59:16 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US