Database or store to handle 30 Mb/sec and 40,000 inserts/sec

uspe...@gmail.com

unread,

Feb 7, 2006, 10:02:05 AM2/7/06

to

We have a need for some sort of database package that can handle the
following specs:

Average data rate - 30 megabits/sec.
Max # inserts - 40,000 records/sec.

What sort of database or store could handle this load? I'm looking to
price out a repository, basically capturing the best bid/ask and last
information from all 6 option exchanges, and then have a retrieval
application query this data in a timely fashion.

suggestions (with links to benchmarks/info) are much appreciated

thanks much

paul c

unread,

Feb 7, 2006, 10:50:02 AM2/7/06

to

given those rates, the flip answer would be a 'memory-based' db except
for the fact that 40,000 recs/sec might approach a billion records in 8
hours!

are you sure those rates are right? (i find it hard to believe that a
single option exchange, let alone six is doing 40,000/6*3,600 =
something like 21 million inserts in an hour!)

the data rate would also seem to imply a dedicated network.

p

Volker Hetzer

unread,

Feb 7, 2006, 11:46:37 AM2/7/06

to

uspe...@gmail.com schrieb:

Have you looked at Oracle TimesTen?
This is a in memory database that can also be coupled to a standard
oracle database. The division of labour typically goes like, TimesTen
does the data reception and event handling (and can do processing too)
and then bulk transfers it to the oracle server, in case you have need
to store the data.

Other than that, Oracle Real Application clusters can probably take
quite a bit of load.

But, really, what kind of hardware are you talking about? You sure
can't do that on a standard PC or workstation.

Btw, I'm not recommending oracle because I get paid by them but because
I run a small oracle db in my company and that's the only database system
I know, apart from mysql.

Lots of Greetings!
Volker

Roy Hann

unread,

Feb 7, 2006, 11:51:05 AM2/7/06

to

<uspe...@gmail.com> wrote in message
news:1139324525.8...@g14g2000cwa.googlegroups.com...

> We have a need for some sort of database package that can handle the
> following specs:
>
> Average data rate - 30 megabits/sec.
> Max # inserts - 40,000 records/sec.
>
> What sort of database or store could handle this load?

There are probably seismic data recorders that can handle data at those
rates ;-)

Maybe www.streambase.com has something for you.

Roy

Jags

unread,

Feb 7, 2006, 12:53:06 PM2/7/06

to

You probably have to go with some main-memory data management solution
to achieve this high rate. GemStone provides a very high speed main
memory based distributed data management platform that is specifically
geared to meets the extreme performance requirements in capital
markets.
Take a look at this white paper ->
http://www.gemstone.com/pdf/GemFireEnterprise_Technicalwhitepaper.pdf

Here you would find the concepts described along with a use case that
describes the use of the technology in a program trading environment.
The following link provides more information on the applicability of
the technology in several areas in capital markets...
http://www.gemstone.com/solutions/capital.php

Best regards,
-- Jags Ramnarayan
(jram...@gemstone.com)

Tony Rogerson

unread,

Feb 7, 2006, 1:18:54 PM2/7/06

to

Hi,

Its more getting the hardware sorted than the database, you need gear
capable of writing to disk at that speed and more, and with todays hardware
that isn't a problem.

Even a pair of mirrored SATA disks on Windows 2003 Server will write at
around 50MBytes / second.

Once you have your hardware set, personally i'd be looking at something with
a couple of GB of RAM (nice big cache) and a couple of mirrored SCSI 15Krpm
disks for a dedicated transaction log and a set of 4 disks scsi 15Krpm disks
in a RAID 10 array.

Microsoft SQL Server 2005 is another option, workground edition which is all
you'll need for that is around £2K per physical CPU if you have a lot of
users or just go for CAL version which is cheaper.

You can use the new committed snapshot isolation so that writers aren't
blocking readers so you won't have a contention problem.

Check out http://microsoft.com/sql for more information.

What sort of DR / high availability do you want on it? Clustering? Database
mirroring??

Tony.

--
Tony Rogerson
SQL Server MVP
http://sqlserverfaq.com - free video tutorials

<uspe...@gmail.com> wrote in message
news:1139324525.8...@g14g2000cwa.googlegroups.com...

Chan

unread,

Feb 7, 2006, 2:17:41 PM2/7/06

to

How much history you are going to keep for how long and how many
concurent users going to access data.

here is math:
1 see 30mb
1 min 1.8gb
1 hour 108 gb
1 day 2.592 TB

axw...@gmail.com

unread,

Feb 7, 2006, 4:06:29 PM2/7/06

to

Upensky...give me a shout. We could probably help you with this
problem.

(latimerp)

unread,

Feb 7, 2006, 4:38:47 PM2/7/06

to

This should handle it with enough Iron under it
http://www.northgate-is.com/reality/v12.php?PHPSESSID=188c11e5381229e8b272c130c4715faf

felix...@gmail.com

unread,

Feb 8, 2006, 4:58:43 AM2/8/06

to

> Average data rate - 30 megabits/sec.
> Max # inserts - 40,000 records/sec.

that's not much. these is what we do on daily basis with kdb+
(http://kx.com).

felix

Noons

unread,

Feb 8, 2006, 6:57:39 AM2/8/06

to

Chan wrote:

Folks, the OP said:
30Mbits/sec

That's "bits", not "Bytes"!

In bytes, that's around 3.333 MB/s.

Which is fine with most current hardware.
40000 records/sec means nothing: what we
need to know is how many bytes/rec on average,
and how many rows can be batched into a single
write operation.

David Cressey

unread,

Feb 8, 2006, 7:00:24 AM2/8/06

to

"paul c" <toledob...@oohay.ac> wrote in message
news:Kk3Gf.580621$ki.6168@pd7tw2no...
> uspe...@gmail.com wrote:

> > Average data rate - 30 megabits/sec.
> > Max # inserts - 40,000 records/sec.

> are you sure those rates are right? (i find it hard to believe that a

> single option exchange, let alone six is doing 40,000/6*3,600 =
> something like 21 million inserts in an hour!)

If an infinite number of monkeys were clicking on an infinite number of
hyperlinks, somebody would be trying to capture all the clicks in a
database. The internet isn't quite there yet, but we're close!

paul c

unread,

Feb 8, 2006, 11:42:38 AM2/8/06

to

Okay, I think I'm starting to get it - part of the angle for handling
such systems is that much of the stream is redundant. If the NYSE
trades a billion shares a day, the day might max-out with only 10
million trades and probably many fewer. If quotes for 1000 issues arrive
every second, a five-hour trading day would produce less than 20 million
quotes, the majority of them differing only in their time stamps.

pc

Jon

unread,

Feb 9, 2006, 10:30:16 AM2/9/06

to

I can send you a white paper which outlines how a benchmark we did with
Caché exceeds this by a large factor.

If you are interested, drop me a line.

Regards

Jon

unread,

Feb 9, 2006, 11:58:03 AM2/9/06

to

To Paul,

I tried to email you but your domain is unreachable.

Jon

Joel Garry

unread,

Feb 9, 2006, 6:21:06 PM2/9/06

to

Noons wrote:

>That's "bits", not "Bytes"!

>In bytes, that's around 3.333 MB/s.

So that's an average of 2TB/week. Do you really defend what Tony said:
Windows/pair of SATA drives can do that?

jg
--
@home.com is bogus.
700HP biodiesel hotrod Willys pickup: "It's cool, it's country and
it fits."
http://www.signonsandiego.com/uniontrib/20060209/news_1b9biodiesel.html

Noons

unread,

Feb 9, 2006, 7:21:00 PM2/9/06

to

Joel Garry wrote:
>
> >That's "bits", not "Bytes"!
>
> >In bytes, that's around 3.333 MB/s.
>
> So that's an average of 2TB/week. Do you really defend what Tony said:
> Windows/pair of SATA drives can do that?
>

Of course not. They can cope with the data capture volumes.
The OP said nothing about retention periods nor can anyone
extrapolate how long they will be.

If the OP wants to keep that volume of data online for a
relatively long period of time, then partitioning and some form
of pruning will be required.

The same applies to any database or hardware you'd care to mention.
I'd love to see how a memory-only db architecture would cope with
2TB/week non-stop without any purges.

Joel Garry

unread,

Feb 9, 2006, 8:14:19 PM2/9/06

to

Noons said:

>The OP said nothing about retention periods nor can anyone
>extrapolate how long they will be.

The OP said:

>What sort of database or store could handle this load? I'm looking to
>price out a repository, basically capturing the best bid/ask and last
>information from all 6 option exchanges, and then have a retrieval
>application query this data in a timely fashion.

I see "repository" and "option exchanges," sounds to me like data one
needs to keep for a long time. Unless it's some unregulated
penny-stock exchange, or someone trying to sell a day-trading advice
service...

I didn't mean to pick on you, I meant to pick on the Windows guy.

jg
--
@home.com is bogus.

I want a blue one!
http://www.autoweek.com/apps/pbcs.dll/article?AID=/20060206/FREE/60127003/1008/CoverStory

Tony Rogerson

unread,

Feb 10, 2006, 9:41:57 AM2/10/06

to

Right then Joel, lets have a go then.

Whats your argument?

Even on a build your own box costing around £500 can deliver over 50MBytes /
second write and read speeds using Windows Server.

Go for the 64 bit version and you can get quite a few GBytes of memory, most
entry level boards <£100 take 4GB of DDR.

REmember the poster said 30megabits (which I read as MBytes) and 40,000 rows
per second; SQL Server will do that without problem.

SQL Server will handle TB's too, if like with ALL vendor databases, you
design it properly.

--
Tony Rogerson
SQL Server MVP
http://sqlserverfaq.com - free video tutorials

"Joel Garry" <joel-...@home.com> wrote in message
news:1139534059....@g47g2000cwa.googlegroups.com...

Frank van Bortel

unread,

Feb 10, 2006, 1:54:42 PM2/10/06

to

Tony Rogerson wrote:
> Right then Joel, lets have a go then.
>
> Whats your argument?
>
> Even on a build your own box costing around £500 can deliver over 50MBytes /
> second write and read speeds using Windows Server.
>
> Go for the 64 bit version and you can get quite a few GBytes of memory, most
> entry level boards <£100 take 4GB of DDR.
>

Don't top post - can I suggest you use a decent news reader?
As you see, your ugly format/utility cuts off the rest of the thread.

Point is: a decent Linux will deliver better throughput on the *same*
hardware. Changed 'twice the' into 'better'. Can't substantiate twice,
can better.

And personally, I'd love to see you stop promoting non-oracle stuff
here. It's annoying; most people here hate Microsoft anyway.
Just a little further, and you're spamming.
--
Regards,
Frank van Bortel

Top-posting is one way to shut me up...

murthi

unread,

Feb 10, 2006, 3:54:30 PM2/10/06

to

Well, then , Frank, I guess you won't reply to this one, eh?

No I don't use a decent new reader myself and I would appreciate it if all
WOULD top post. Many's the time I don't bother to read something because its
too frigging long way down.

So thanks, Tony, for making it easy to read...

Chandru Murthi

"Frank van Bortel" <frank.va...@gmail.com> wrote in message
news:dsimqj$1tp$1...@news2.zwoll1.ov.home.nl...

Excalibur

unread,

Feb 10, 2006, 5:40:48 PM2/10/06

to

Hi Murthi
Thanks for the response. People not top posting drive me nuts. I have
followed a thread and want to sea the latest response not piles of pointy
brackets. Who dreamed up this so-called etiquette anyway?
Peter McMurray
"murthi" <c_xyz_murthi@seeing_xyz_green.net> wrote in message
news:a47Hf.21$Hn.19@trndny05...

Ed Sheehan

unread,

Feb 10, 2006, 6:48:57 PM2/10/06

to

"Excalibur" <excal...@bigpond.com> wrote in message
news:QD8Hf.4873$yK1....@news-server.bigpond.net.au...

>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>

I agree.

Ed

Joel Garry

unread,

Feb 10, 2006, 7:23:11 PM2/10/06

to

Tony Rogerson wrote:
> Right then Joel, lets have a go then.
>
> Whats your argument?
>
> Even on a build your own box costing around £500 can deliver over 50MBytes /
> second write and read speeds using Windows Server.
>
> Go for the 64 bit version and you can get quite a few GBytes of memory, most
> entry level boards <£100 take 4GB of DDR.
>
> REmember the poster said 30megabits (which I read as MBytes) and 40,000 rows
> per second; SQL Server will do that without problem.
>
> SQL Server will handle TB's too, if like with ALL vendor databases, you
> design it properly.

Well, here's where I disagree. The design necessary to get SQL Server
to handle TB's of data along with random transactional queries puts the
onus on programmers to do it right. All you need is one newbie and you
are screwed. Unless you use the new feature that makes it work like
Oracle. So let's see: New unproven feature or risk of manual error.
New unproven feature or risk of manual error. New unproven feature
that probably has bugs (like with ALL vendor databases new features),
or near-100% chance of manual error.

I'll pass. Oracle handles MVCC right by default, Oracle environments
have more problems with SQL-Server people who haven't unlearned doing
it wrong than with the actual native environment.

Then there's recovery. May be a lot simpler in your way, but simpler
isn't necessarily better when you get to TB. Depends. There is no
such thing as "without problem."

A few years ago I would have said Rdb, by the way, but Oracle has blown
by, even with some historical baggage.

Every couple of years I think to myself "self, MS has a new generation
of stuff, let's give it a try." And every couple of years I discover
all the things wrong, the hard way. And then I become an Oracle/unix
bigot all over again. Perhaps it's because I started on similar
hardware as Bill Gates and can't understand why he allows things to be
so bad, he ought to know better. I'd think "Maybe he just forked off
too early," but considering the previous experience of the NT team,
that doesn't work.

jg
--
@home.com is bogus.

Culture Clash:
http://www.businessweek.com/technology/content/feb2006/tc20060209_810527.htm

Tony Rogerson

unread,

Feb 11, 2006, 6:07:51 AM2/11/06

to

Hi Joel,

I read a lot of words but no substance there, you don't mention features,
you don't go into detail, you don't specify links to back up your case.

For a fully documented reference implementation on how they did Barnes and
Noble go here: http://www.microsoft.com/sql/solutions/bi/projectreal.mspx -
it contains a ton of white papers and architecture documents, a very good
source of material if you are starting to develop a big system, if you need
more just ask.

A design necessary to get SQL Server to handles TB's puts the onus on the
database designer to get the database design and physical implementation of
it right and also of your system engineer to get the hardware right; that is
the same for ALL vendors. The days when you brush SQL Server away as a
developer tool are long gone.

DB2 was my foundation, 5 years DB2 and the past 13 years Microsoft SQL
Server, I have done a bit of Oracle but not worth mentioning.

For ALL vendor databases all you need is one newbie writing a cross join
between big tables and you suddenly have a load on your system.

I here the MS biggotry often, my take is that you should use the right tool
for the right job and at the right cost and not blindly lead yourselves down
alleys you can't get out of, I recall the Oracle pricing model and lock-into
their software model.

--
Tony Rogerson
SQL Server MVP
http://sqlserverfaq.com - free video tutorials

"Joel Garry" <joel-...@home.com> wrote in message

news:1139617391.0...@g14g2000cwa.googlegroups.com...

Sybrand Bakker

unread,

Feb 11, 2006, 6:14:49 AM2/11/06

to

On Fri, 10 Feb 2006 22:40:48 GMT, "Excalibur"
<excal...@bigpond.com> wrote:

>Hi Murthi
>Thanks for the response. People not top posting drive me nuts. I have
>followed a thread and want to sea the latest response not piles of pointy
>brackets. Who dreamed up this so-called etiquette anyway?
>Peter McMurray

In this group it's exactly the opposite: many posters just hate morons
like you insisting on top-posting. The human species happens to read
from top to bottom: that is apart from the people brainwashed by the
Evil Empire driven by Bill Gates, who - you are a specific example-
demonstrate over and over and over again they are just too damn lazy
to scroll down.
--
Sybrand Bakker, Senior Oracle DBA

Sybrand Bakker

unread,

Feb 11, 2006, 6:15:45 AM2/11/06

to

On Fri, 10 Feb 2006 16:48:57 -0700, "Ed Sheehan"
<NOed...@xmission.com> wrote:

>
>I agree.
>
>Ed

Yet another moron.

Sybrand Bakker

unread,

Feb 11, 2006, 6:17:29 AM2/11/06

to

On Tue, 07 Feb 2006 16:38:47 -0500, "(latimerp)"
<"(latimerp)"@comcast.net> wrote:

>This should handle it with enough Iron under it
>http://www.northgate-is.com/reality/v12.php?PHPSESSID=188c11e5381229e8b272c130c4715faf

Please stop top-posting

Sybrand Bakker

unread,

Feb 11, 2006, 6:17:03 AM2/11/06

to

On 7 Feb 2006 13:06:29 -0800, axw...@gmail.com wrote:

>Upensky...give me a shout. We could probably help you with this
>problem.

Please stop top-posting

Leif B. Kristensen

unread,

Feb 11, 2006, 6:21:46 AM2/11/06

to

Tony Rogerson skrev:

> For a fully documented reference implementation on how they did Barnes
> and Noble go here:
> http://www.microsoft.com/sql/solutions/bi/projectreal.mspx

Aha. That explains why the Barnes & Noble web site insists that my
Mozilla on Linux doesn't support cookies. I rather buy my books from
Amazon, who have understood how to write browser-independent web apps.
--
Leif Biberg Kristensen
http://solumslekt.org/

Tony Rogerson

unread,

Feb 11, 2006, 8:42:04 AM2/11/06

to

Some of us use a nice threaded interface so can see what the previous
message, suggestion is to get a better news reader, try google groups if you
only have a browser.

Unless you are replying to a specific comment of the previous thread then
why quote it?

In life, you do not repeat a sentance back to the person you have just
listened to in order to reply or give your opinion.

Perhaps you have too much time on your hands, does your boss know?

--
Tony Rogerson
SQL Server MVP
http://sqlserverfaq.com - free video tutorials

"Sybrand Bakker" <pos...@sybrandb.demon.nl> wrote in message
news:bjhru11q4tcg6s4hn...@4ax.com...

Tony Rogerson

unread,

Feb 11, 2006, 8:49:39 AM2/11/06

to

> Aha. That explains why the Barnes & Noble web site insists that my
> Mozilla on Linux doesn't support cookies. I rather buy my books from
> Amazon, who have understood how to write browser-independent web apps.

Probably a bug on the website, have you done the professional thing and
reported it?

Also, thats very little if anything to do with the underlying database
platform; the web code / middle tier is a different business unit from the
SQL team.

Instead of 'bitching' why not comment on the discussion, perhaps thats a
trait from the linux community - they are just too busy bashing microsoft
and feeding their egos to consider what the business wants (or needs).

--
Tony Rogerson
SQL Server MVP
http://sqlserverfaq.com - free video tutorials

"Leif B. Kristensen" <junk...@solumslekt.org> wrote in message
news:SoSdnctkyYd...@telenor.com...

Sybrand Bakker

unread,

Feb 11, 2006, 1:59:18 PM2/11/06

to

On Sat, 11 Feb 2006 13:42:04 -0000, "Tony Rogerson"
<tonyro...@sqlserverfaq.com> wrote:

>In life, you do not repeat a sentance back to the person you have just
>listened to in order to reply or give your opinion.
>
>Perhaps you have too much time on your hands, does your boss know?
>
>--
>Tony Rogerson
>SQL Server MVP
>http://sqlserverfaq.com - free video tutorials

It appears you insist on being rude, and ignore netiquette.Am I right
in asssuming you mother forgot to civilize you?

Sybrand Bakker

unread,

Feb 11, 2006, 2:02:38 PM2/11/06

to

On Sat, 11 Feb 2006 11:07:51 -0000, "Tony Rogerson"
<tonyro...@sqlserverfaq.com> wrote:

>I here the MS biggotry often, my take is that you should use the right tool
>for the right job and at the right cost and not blindly lead yourselves down
>alleys you can't get out of, I recall the Oracle pricing model and lock-into
>their software model.
>
>--
>Tony Rogerson
>SQL Server MVP
>http://sqlserverfaq.com - free video tutorials
>

I agree there are people who are blind and can't see sqlserver will
not have the same set of features as Oracle in even a hundred years
from now. Those people also prefer to be locked in by the Evil Empire,
aka Microsoft, and show their slavery to this Empire by insisting on
being rude and ignore Netiquette.

hpuxrac

unread,

Feb 11, 2006, 2:51:42 PM2/11/06

to

Sybrand Bakker wrote:
# and show their slavery to this Empire by insisting on being rude and
ignore Netiquette.

Coming from you?

Is this humor?

Mark Townsend

unread,

Feb 11, 2006, 2:59:47 PM2/11/06

to

Tony Rogerson wrote:

> For a fully documented reference implementation on how they did Barnes and
> Noble go here: http://www.microsoft.com/sql/solutions/bi/projectreal.mspx -
> it contains a ton of white papers and architecture documents, a very good
> source of material if you are starting to develop a big system, if you need
> more just ask.

Hmm - I couldn't find any reference to B&N in the link you provided. I
did not explore the sub links that much - could you perhaps provide a
direct reference to the B&N documented reference implementation ?

I work with B&N, and they have some Microsoft. They have a great deal of
Oracle. Even Microsoft's own reference material cites this -
http://members.microsoft.com/customerevidence/Common/FileOpen.aspx?FileName=21672_Barnes_Noble_SQL2005_final.doc

"The company, which runs Oracle databases for operational systems ...."

Tony Rogerson

unread,

Feb 11, 2006, 3:01:48 PM2/11/06

to

> I agree there are people who are blind and can't see sqlserver will
> not have the same set of features as Oracle in even a hundred years

Does it need it? SQL Server has CLR which means it can be expanded
infinitely.

> from now. Those people also prefer to be locked in by the Evil Empire,
> aka Microsoft, and show their slavery to this Empire by insisting on
> being rude and ignore Netiquette.

I've been in the industry over 19 years now starting on IBM mainframe of
which I served 7 years.

Talk about 'evil empire'.

I am not a blind Microsoft advocate, remember right tool right job; but I
will say one thing for Microsoft - they have completely opened up the field
of IT and that to me is a good thing - just look at what's happend the BI,
its now affordable at the 2 man company level; you get BI for free in the
standard edition (£2K).

The adolecent bitching of the other parts of your post show you to be
somebody in their infancy within IT with little broad experience so I'll
comment no further.

--
Tony Rogerson
SQL Server MVP
http://sqlserverfaq.com - free video tutorials

"Sybrand Bakker" <pos...@sybrandb.demon.nl> wrote in message

news:d1dsu1t2722j9cka0...@4ax.com...

Mark Townsend

unread,

Feb 11, 2006, 3:16:25 PM2/11/06

to Tony Rogerson

Tony Rogerson wrote:
>>I agree there are people who are blind and can't see sqlserver will
>>not have the same set of features as Oracle in even a hundred years
>
>
> Does it need it? SQL Server has CLR which means it can be expanded
> infinitely.
>

Well - Oracle also supports CLR in the database as well. However the
ability to extend a database with stored procedures etc written in CLR,
Java or even a database's own data processing language does not equate
to the ability to add new features to the database.

But I suspect that you knew that and are just being ornery

murthi

unread,

Feb 11, 2006, 3:28:20 PM2/11/06

to

I suppose this is from another of those people who inhabit those other
database forums. Amazing how rude people can be so blind to their own
rudenness. Time to tune up the spam filter.

Chandru Murthi

"Sybrand Bakker" <pos...@sybrandb.demon.nl> wrote in message

news:bjhru11q4tcg6s4hn...@4ax.com...

csig...@hotmail.com

unread,

Feb 11, 2006, 3:46:01 PM2/11/06

to

I don't believe this thread. Doesn't this forum identify itself as
com.databases.pick?

Is it not reasonable to assume that the same request has been posted to
other forums and that those forums will respond with their solution
while the poster expects this forum to respond with a Pick solution?

Henry Keultjes
Microdyne Company
Mansfield Ohio USA

paul c

unread,

Feb 11, 2006, 3:49:19 PM2/11/06

to

Mark Townsend wrote:
> Tony Rogerson wrote:
>
>>> I agree there are people who are blind and can't see sqlserver will
>>> not have the same set of features as Oracle in even a hundred years
>>
>>
>>
>> Does it need it? SQL Server has CLR which means it can be expanded
>> infinitely.
>>
>
> Well - Oracle also supports CLR in the database as well. However the
> ability to extend a database with stored procedures etc written in CLR,
> Java or even a database's own data processing language does not equate
> to the ability to add new features to the database.
>

> ...

I'm curious how CLR per se 'expands' (advances?) the db. I noticed
something that is maybe related called Qlink or similar by one of the
CLR authors but it seemed like bumpf to me. Is CLR anything more than a
copycat of the JVM (the JVM seeming to me to be nothing more an
interpreted version of some arbitrary low-level assembly language with a
few complex operators added for an object/typing machine of some sort,
which isn't necessarily relevant to database theory)?

cheers,
pc

Tony Rogerson

unread,

Feb 11, 2006, 3:56:22 PM2/11/06

to

> Well - Oracle also supports CLR in the database as well. However the
> ability to extend a database with stored procedures etc written in CLR,
> Java or even a database's own data processing language does not equate to
> the ability to add new features to the database.

In SQL Server its in-process so performs better, in Oracle and DB2 its out
of process and they both have their limitations.

In SQL Server you can write, User Definied - Types, Functions, Aggregates,
Stored Procedures, Triggers within CLR, so its not just stored procedures;
so, yes - you can add new features to the database; if I want to write a
regurlar expression function that can be used anywhere within SQL Server I
can write it in CLR and call it anywhere I can use a built-in system
function.

--
Tony Rogerson
SQL Server MVP
http://sqlserverfaq.com - free video tutorials

"Mark Townsend" <markbt...@comcast.net> wrote in message
news:43EE4619...@comcast.net...

Tony Rogerson

unread,

Feb 11, 2006, 4:00:15 PM2/11/06

to

> Hmm - I couldn't find any reference to B&N in the link you provided. I did
> not explore the sub links that much - could you perhaps provide a direct
> reference to the B&N documented reference implementation ?
>
> I work with B&N, and they have some Microsoft. They have a great deal of
> Oracle. Even Microsoft's own reference material cites this -
> http://members.microsoft.com/customerevidence/Common/FileOpen.aspx?FileName=21672_Barnes_Noble_SQL2005_final.doc
>
> "The company, which runs Oracle databases for operational systems ...."

Check the project technical overview here
http://www.microsoft.com/technet/prodtechnol/sql/2005/projreal.mspx

Also here
http://members.microsoft.com/customerevidence/Search/EvidenceDetails.aspx?EvidenceID=10013&LanguageID=1

Also, watch the SQL 2005 launch videos - google that.

Do you want more?

--
Tony Rogerson
SQL Server MVP
http://sqlserverfaq.com - free video tutorials

"Mark Townsend" <markbt...@comcast.net> wrote in message

news:kYadnQDRPZS...@comcast.com...

Tony Rogerson

unread,

Feb 11, 2006, 4:06:40 PM2/11/06

to

> Hmm - I couldn't find any reference to B&N in the link you provided. I did
> not explore the sub links that much - could you perhaps provide a direct
> reference to the B&N documented reference implementation ?
>
> I work with B&N, and they have some Microsoft. They have a great deal of
> Oracle. Even Microsoft's own reference material cites this -
> http://members.microsoft.com/customerevidence/Common/FileOpen.aspx?FileName=21672_Barnes_Noble_SQL2005_final.doc

Check the project technical overview here
http://www.microsoft.com/technet/prodtechnol/sql/2005/projreal.mspx

Also here
http://members.microsoft.com/customerevidence/Search/EvidenceDetails.aspx?EvidenceID=10013&LanguageID=1

Also, watch the SQL 2005 launch videos - google that.

Do you want more?

--

Tony Rogerson
SQL Server MVP
http://sqlserverfaq.com - free video tutorials

"Mark Townsend" <markbt...@comcast.net> wrote in message
news:kYadnQDRPZS...@comcast.com...

Tony Rogerson

unread,

Feb 11, 2006, 4:10:20 PM2/11/06

to

Take SQL Server implementation of CLR.

You can write a User Definied scalar function in say C# that uses the
regular expression library for instance to validate the EmailAddress column
contains a valid email, now you can use that CLR function inside a regular
CHECK constraint on your table which is very powerful. People might say,
well do the validation in the application, thats true - but you could really
do with integrity at the heart, what's to stop your DBA loading rubbish
data?

Within SQL Server, CLR runs in-process, alongside SQL Server; the main
reason for it late shipping - its so bolted into the product; it doesn't
cause instability unlike using extended stored procedures because of the way
.NET works.

Tony.

--
Tony Rogerson
SQL Server MVP
http://sqlserverfaq.com - free video tutorials

"paul c" <toledob...@oohay.ac> wrote in message
news:j5sHf.480197$2k.314310@pd7tw1no...

Bill H

unread,

Feb 11, 2006, 5:13:28 PM2/11/06

to

Come on, Henry. It's one heck-of-a grin, and worth it too. :-)

Bill

<csig...@hotmail.com> wrote in message
news:1139690761.2...@f14g2000cwb.googlegroups.com...

Leif B. Kristensen

unread,

Feb 11, 2006, 5:38:21 PM2/11/06

to

Tony Rogerson skrev:

> Probably a bug on the website, have you done the professional thing
> and reported it?

Yes of course, several weeks ago. But I still haven't got an answer.

> Instead of 'bitching' why not comment on the discussion, perhaps thats
> a trait from the linux community - they are just too busy bashing
> microsoft and feeding their egos to consider what the business wants
> (or needs).

It's also a very common trait among the MS fanboys to just write off
any critic of the MS raping of open industry standards as "Microsoft
bashing".

Frank Hamersley

unread,

Feb 11, 2006, 6:53:49 PM2/11/06

to

Tony Rogerson wrote:
>>Well - Oracle also supports CLR in the database as well. However the
>>ability to extend a database with stored procedures etc written in CLR,
>>Java or even a database's own data processing language does not equate to
>>the ability to add new features to the database.
>
> In SQL Server its in-process so performs better, in Oracle and DB2 its out
> of process and they both have their limitations.

I am curious about the in-process aspect - is that akin to loading
device drivers in ring 0? What are the memory protection aspects of the
various offerings?

> In SQL Server you can write, User Definied - Types, Functions, Aggregates,
> Stored Procedures, Triggers within CLR, so its not just stored procedures;
> so, yes - you can add new features to the database; if I want to write a
> regurlar expression function that can be used anywhere within SQL Server I
> can write it in CLR and call it anywhere I can use a built-in system
> function.

IMO in the CLR option vendors are moving from providing a developer from
having a shotgun to blow their left foot off to providing a 155mm
howitzer! Of course we are all responsible adults here so there is no
risk of it ever happening - yeah right!

Cheers, Frank.

Frank Hamersley

unread,

Feb 11, 2006, 7:03:52 PM2/11/06

to

Tony Rogerson wrote:
>>Hmm - I couldn't find any reference to B&N in the link you provided. I did
>>not explore the sub links that much - could you perhaps provide a direct
>>reference to the B&N documented reference implementation ?
>>
>>I work with B&N, and they have some Microsoft. They have a great deal of
>>Oracle. Even Microsoft's own reference material cites this -
>>http://members.microsoft.com/customerevidence/Common/FileOpen.aspx?FileName=21672_Barnes_Noble_SQL2005_final.doc
>
> Check the project technical overview here
> http://www.microsoft.com/technet/prodtechnol/sql/2005/projreal.mspx
>
> Also here
> http://members.microsoft.com/customerevidence/Search/EvidenceDetails.aspx?EvidenceID=10013&LanguageID=1
>
> Also, watch the SQL 2005 launch videos - google that.
>
> Do you want more?
>

Its just a DW - not a TP solution. Sure it outperforms its predecessor
- as you would hope if you threw new hardware at the problem !

Therefore I hold it says nothing about SQL5 in respect of the OP's question.

Cheers, Frank.

HansF

unread,

Feb 12, 2006, 12:07:14 AM2/12/06

to

On Fri, 10 Feb 2006 16:48:57 -0700, Ed Sheehan wrote:

>
>
> I agree.
>
> Ed

Actually, the important thing with bottom posting is to remove the crap
that is totally irrelevant.

(Yet another reason why I no longer use Outlook <g>)

--
Hans Forbrich
Canada-wide Oracle training and consulting
mailto: Fuzzy.GreyBeard_at_gmail.com
*** Top posting [replies] guarantees I won't respond. ***

Sybrand Bakker

unread,

Feb 12, 2006, 6:29:55 AM2/12/06

to

On Sat, 11 Feb 2006 20:28:20 GMT, "murthi"
<c_xyz_murthi@seeing_xyz_green.net> wrote:

>I suppose this is from another of those people who inhabit those other
>database forums. Amazing how rude people can be so blind to their own
>rudenness. Time to tune up the spam filter.
>
>Chandru Murthi

POT-KETTLE-BLACK

PLOINK!!!!!!!!

murthi

unread,

Feb 12, 2006, 12:18:08 PM2/12/06

to

I forwarded this sequence to my 8-year old as an example of how not to be
childish. He says thank you, Mr Bakker, for the lesson, what does ploink
mean in this context, and if you really are a "senior oracle", could he
forward some questions he has been pondering lately?

CDPers: I remember getting into similar exchanges two years ago with members
of this group. Perhaps they work too hard?

10-4, Chandru Murthi

"Sybrand Bakker" <pos...@sybrandb.demon.nl> wrote in message

news:a07uu11oi9bqe702g...@4ax.com...

Bob Hairgrove

unread,

Feb 12, 2006, 12:22:03 PM2/12/06

to

On Sun, 12 Feb 2006 05:07:14 GMT, HansF <News...@telus.net> wrote:

>(Yet another reason why I no longer use Outlook <g>)

You mean ... LookOut?? <g>

--
Bob Hairgrove
NoSpam...@Home.com

Bill Karwin

unread,

Feb 12, 2006, 2:37:32 PM2/12/06

to

"HansF" <News...@telus.net> wrote

> Actually, the important thing with bottom posting is to remove the crap
> that is totally irrelevant.

Hm, interesting. If that's true, then why hasn't it eliminated this entire
thread? <g>

Bill

Luke Webber

unread,

Feb 12, 2006, 4:07:40 PM2/12/06

to

Frank van Bortel wrote:

> Don't top post - can I suggest you use a decent news reader?
> As you see, your ugly format/utility cuts off the rest of the thread.

You're a touchy bastard, aren't you? I have no problem with top posting.
I'd far rather read a top post than a single-line reply quoting a week
of context.

> Point is: a decent Linux will deliver better throughput on the *same*
> hardware. Changed 'twice the' into 'better'. Can't substantiate twice,
> can better.
>
> And personally, I'd love to see you stop promoting non-oracle stuff
> here. It's annoying; most people here hate Microsoft anyway.
> Just a little further, and you're spamming.

Check the headers and apologise to Tony. This is cross-posted to three
groups other than your Oracle hangout.

Luke

Luke Webber

unread,

Feb 12, 2006, 4:11:30 PM2/12/06

to

Sybrand Bakker wrote:

> In this group it's exactly the opposite: many posters just hate morons
> like you insisting on top-posting. The human species happens to read
> from top to bottom: that is apart from the people brainwashed by the
> Evil Empire driven by Bill Gates, who - you are a specific example-
> demonstrate over and over and over again they are just too damn lazy
> to scroll down.

"This group" is actually four separate groups, not all of which are as
precious as cdos apparently is. Check the headers.

Luke

Thomas Kellerer

unread,

Feb 12, 2006, 4:55:32 PM2/12/06

to

Luke Webber wrote on 12.02.2006 22:07:
> You're a touchy bastard, aren't you? I have no problem with top posting.
> I'd far rather read a top post than a single-line reply quoting a week
> of context.
>

I couldn't agree more!

Since the time I started using a thread newsreader, I *never* understood the
advantages of bottom posting where I have to scroll all the way down just to
read the reply that would repeat everything that is displayed conventiently for
me in a line above.

But then, when in Rome...

Thomas

(latimerp)

unread,

Feb 12, 2006, 5:51:48 PM2/12/06

to

Luke Webber wrote:
> Frank van Bortel wrote:
>
>> Don't top post - can I suggest you use a decent news reader?
>> As you see, your ugly format/utility cuts off the rest of the thread.
>
>
> You're a touchy bastard, aren't you?

Astute observation Luke ;)

<snip>,psss This cuts off the thread too.
>

>> Point is: a decent Linux will deliver better throughput on the *same*
>> hardware. Changed 'twice the' into 'better'. Can't substantiate twice,
>> can better.

Frank, try using smaller words that you have a better command of.
If I was you I'd start with 2 syllables and build on that.

<;=), Patrick

>
> Luke

hpuxrac

unread,

Feb 12, 2006, 8:06:47 PM2/12/06

to

Sybrand Bakker wrote:
> On Sat, 11 Feb 2006 20:28:20 GMT, "murthi"
> <c_xyz_murthi@seeing_xyz_green.net> wrote:
>
> >I suppose this is from another of those people who inhabit those other
> >database forums. Amazing how rude people can be so blind to their own
> >rudenness. Time to tune up the spam filter.
> >
> >Chandru Murthi
>
> POT-KETTLE-BLACK
>
> PLOINK!!!!!!!!

Hows the rehab going?

HansF

unread,

Feb 12, 2006, 11:17:41 PM2/12/06

to

Because there are bigots on both sides who still haven't figured out what
(and why) appropriate posting involves.

Any time you let religion [for example top vs bottom posting mantra] do
your thinking for you and you can create a war based on irrelevancies any
time you want. May the Lilliputians prevail forever.

DA Morgan

unread,

Feb 13, 2006, 12:23:24 PM2/13/06

to

murthi wrote:
> I suppose this is from another of those people who inhabit those other
> database forums. Amazing how rude people can be so blind to their own
> rudenness. Time to tune up the spam filter.
>
> Chandru Murthi

Let me see if I've got this correctly.

1. You called someone rude and blind.
2. You top posted the statement.
3. You called someone rude and blind.
4. You cross posted to three different groups.
5. 2. You called someone rude and blind.

Don't they make mirrors where you live?
--
Daniel A. Morgan
http://www.psoug.org
damo...@x.washington.edu
(replace x with u to respond)

Bruce A. Holt

unread,

Feb 13, 2006, 12:29:06 PM2/13/06

to

Enough of this insanity! My inbox is in flames!

Please read the following Slashdot article and let's all be friends.

http://tinyurl.com/dogs6

murthi

unread,

Feb 13, 2006, 1:45:15 PM2/13/06

to

>Enough of this insanity! My inbox is in flames!
>Please read the following Slashdot article and let's all be friends.
http://tinyurl.com/dogs6

Yeah, you're right Bruce, good article. I just deleted what I thought was an
apt reply to the "insanity" below (which no doubt Mr Morgan will misconstrue
as being the same as my calling *him* insane) so this doesn't have to go on
ad nauseum. I await further developments without interest.

Chandru Murthi

"DA Morgan" <damo...@psoug.org> wrote in message
news:11398514...@jetspin.drizzle.com...

joel-...@home.com

unread,

Feb 13, 2006, 4:35:42 PM2/13/06

to

Tony Rogerson wrote:

> In life, you do not repeat a sentance back to the person you have just
> listened to in order to reply or give your opinion.

You mean, you haven't heard of "active listening?" You should try it.
It is based on the precept that people don't listen to most of what is
said, and alter what they do hear. So, if you repeat back a concept in
your own words, you give feedback to the other how well you understood
what they said.

I didn't bother to give more specific examples of SQL-Servers'
shortcomings because it has already been cross-flamed to death (and
even discussed rationally in Tom Kyte's books). I was just posting the
executive summary. Which, as Noons alluded, is probably beyond what
can be inferred from the OP.

The point about having a high volume AND read consistency being
difficult is an important one. If you have two people sitting next to
each other and one is quoted at $1.15/share and the other at
$1.08/share, you have big, big problems.

jg
--
@home.com is bogus.
"Bob Costas makes sweet love to Grandma Luge." - heard on radio.

HansF

unread,

Feb 13, 2006, 5:08:08 PM2/13/06

to

On Sat, 11 Feb 2006 13:42:04 +0000, Tony Rogerson wrote:

>
> In life, you do not repeat a sentance back to the person you have just
> listened to in order to reply or give your opinion.

You never do?

Even if there are several other conversations going on concurrently that
may involve the same person. Even if there are time delays of,
perhaps days, between the parts of the conversation?

Do you never, ever, verify that your reply will be relevant to only one
part of the conversation?

Do you always jump into conversations assuming your comments are relevant
and any all parts of the conversation?

I bow to the amazing communication skill you have. I could not do that.

--
Hans Forbrich (CTM)

Volker Hetzer

unread,

Feb 14, 2006, 7:43:53 AM2/14/06

to

Tony Rogerson schrieb:

> Unless you are replying to a specific comment of the previous thread then
> why quote it?
So that others can see which part I'm replying to.

>
> In life, you do not repeat a sentance back to the person you have just
> listened to in order to reply or give your opinion.

Actually, in scientific discourse it often happens. In normal oral
talks it doesn't and is the reason why written discussions often are
more productive.

Volker

Jeff Clausius

unread,

Feb 14, 2006, 10:00:21 AM2/14/06

to

Frank:

Actually, please avoid bottom posting. Its difficult to follow a thread
when users have to scroll to the bottom to see the content. If you want
to jump in mid-stream, then work bottom up.

Yes, a good news reader lets you choose to top post or bottom post. I
personally like top posting.

Should I even get into the cross posting done here?

Jeff

Frank van Bortel <frank.va...@gmail.com> wrote in
news:dsimqj$1tp$1...@news2.zwoll1.ov.home.nl:

> Tony Rogerson wrote:
>> Right then Joel, lets have a go then.
>>
>> Whats your argument?
>>
>> Even on a build your own box costing around £500 can deliver over
>> 50MBytes / second write and read speeds using Windows Server.
>>
>> Go for the 64 bit version and you can get quite a few GBytes of
>> memory, most entry level boards <£100 take 4GB of DDR.

>>
>
> Don't top post - can I suggest you use a decent news reader?
> As you see, your ugly format/utility cuts off the rest of the thread.
>

> Point is: a decent Linux will deliver better throughput on the *same*
> hardware. Changed 'twice the' into 'better'. Can't substantiate twice,
> can better.
>

mAsterdam

unread,

Feb 14, 2006, 11:01:18 AM2/14/06

to

Jeff Clausius wrote:
> Actually, please avoid bottom posting.

Agreed. Please avoid top posting as well.
When you really address the whole content
of a small post it doesn't matter.

Otherwise, for the sake of the reader,
please just quote what is relevant to your answer,
and put your answer just below the quote.

Christopher Browne

unread,

Feb 14, 2006, 8:44:24 PM2/14/06

to

Jeff Clausius <nucla...@tendlrow.tta.ten> wrote:
> Frank:
>
> Actually, please avoid bottom posting. Its difficult to follow a thread
> when users have to scroll to the bottom to see the content. If you want
> to jump in mid-stream, then work bottom up.

Bottom posting is as heinous as top posting. Neither should be done.

> Yes, a good news reader lets you choose to top post or bottom post. I
> personally like top posting.

By my private definition, you are a pine tree. That doesn't mean top
posting is a good thing.

> Should I even get into the cross posting done here?

Probably not.
--
select 'cbbrowne' || '@' || 'gmail.com';
http://linuxfinances.info/info/lsf.html
"By your private definition, they were poor, but who cares?. By my
private definition, you are an elm tree."
-- Mike Coffin <m...@Eng.Sun.COM>

Jeff Clausius

unread,

Feb 15, 2006, 9:41:34 AM2/15/06

to

Yes, I agree. Inline answering should be used for quick question/answers.
There is no real absolute is there?

However, for larger discussions where there are paragraphs of text in the
post, top-posting still rules.

To each their own.

Christopher Browne <cbbr...@acm.org> wrote in
news:878xsdj...@wolfe.cbbrowne.com:

> Jeff Clausius <nucla...@tendlrow.tta.ten> wrote:
>> Frank:
>>
>

JEDIDIAH

unread,

Feb 16, 2006, 3:46:12 PM2/16/06

to

On 2006-02-11, Tony Rogerson <tonyro...@sqlserverfaq.com> wrote:
> Hi Joel,
>
> I read a lot of words but no substance there, you don't mention features,
> you don't go into detail, you don't specify links to back up your case.
>
> For a fully documented reference implementation on how they did Barnes and
> Noble go here: http://www.microsoft.com/sql/solutions/bi/projectreal.mspx -
> it contains a ton of white papers and architecture documents, a very good
> source of material if you are starting to develop a big system, if you need
> more just ask.

That's not a big system. That wasn't even a big system 5 years ago.

[deletia]

--
NO! There are no CODICILES of Fight Club! |||
/ | \
That way leads to lawyers and business megacorps and credit cards!

Posted Via Usenet.com Premium Usenet Newsgroup Services
----------------------------------------------------------
** SPEED ** RETENTION ** COMPLETION ** ANONYMITY **
----------------------------------------------------------
http://www.usenet.com

JEDIDIAH

unread,

Feb 16, 2006, 3:36:11 PM2/16/06

to

On 2006-02-11, Sybrand Bakker <pos...@sybrandb.demon.nl> wrote:
> On Fri, 10 Feb 2006 22:40:48 GMT, "Excalibur"
><excal...@bigpond.com> wrote:
>
>>Hi Murthi
>>Thanks for the response. People not top posting drive me nuts. I have
>>followed a thread and want to sea the latest response not piles of pointy
>>brackets. Who dreamed up this so-called etiquette anyway?
>>Peter McMurray

>
> In this group it's exactly the opposite: many posters just hate morons
> like you insisting on top-posting. The human species happens to read
> from top to bottom: that is apart from the people brainwashed by the

Don't be a moron and make gross generalizations about subjects
you obviously don't have any expertise in.

Joel Garry

unread,

Feb 16, 2006, 7:39:19 PM2/16/06

to

JEDIDIAH wrote:
> On 2006-02-11, Tony Rogerson <tonyro...@sqlserverfaq.com> wrote:
> > Hi Joel,
> >
> > I read a lot of words but no substance there, you don't mention features,
> > you don't go into detail, you don't specify links to back up your case.
> >
> > For a fully documented reference implementation on how they did Barnes and
> > Noble go here: http://www.microsoft.com/sql/solutions/bi/projectreal.mspx -
> > it contains a ton of white papers and architecture documents, a very good
> > source of material if you are starting to develop a big system, if you need
> > more just ask.
>
> That's not a big system. That wasn't even a big system 5 years ago.
>

Not only that, but my wife recently used the B&N web system for the
first time, and it screwed up my wife's discount. They seem to have
some disconnect between ordering and membership.

PUT REFERENTIAL INTEGRITY IN THE DATABASE, FOLKS!

jg
--
@home.com is bogus.

http://www.eusprig.org/stories.htm

Joel Garry

unread,

Feb 22, 2006, 2:13:55 PM2/22/06

to

Tony Rogerson wrote:
> Guys, I simple corrected Joel or whomever it was on their original post and
> its gone from there, for each myth posted about SQL Server 2005 I've
> responded and kept responding, its not me driving this - its you guys!

You mean where I said "So that's an average of 2TB/week. Do you really
defend what Tony said:
Windows/pair of SATA drives can do that? "

Or where I said "Well, here's where I disagree. The design necessary
to get SQL Server
to handle TB's of data along with random transactional queries puts the

onus on programmers to do it right. All you need is one newbie and you

are screwed. Unless you use the new feature that makes it work like
Oracle. So let's see: New unproven feature or risk of manual error.
New unproven feature or risk of manual error. New unproven feature
that probably has bugs (like with ALL vendor databases new features),
or near-100% chance of manual error.

I'll pass. Oracle handles MVCC right by default, Oracle environments
have more problems with SQL-Server people who haven't unlearned doing
it wrong than with the actual native environment. "

Or where I said "I didn't bother to give more specific examples of

SQL-Servers'
shortcomings because it has already been cross-flamed to death (and
even discussed rationally in Tom Kyte's books). I was just posting the

executive summary. Which, as Noons alluded, is probably beyond what
can be inferred from the OP. "

Frankly, I haven't seen where you've disproved any of that, or where
you've properly responded to Galen et al once again hashing over what I
didn't even bother to rehash.

You just don't seem to get that SS has done it wrong, most all the
people that use it have gotten it wrong, and they aren't going to
change any time soon. You don't even see there is a problem with
people waiting for blocked reads! You don't even see there is a
problem with combining different transaction isolations in the same
system!

>
> I've tried my best to have a technical discussion but with you and gaven its
> just impossible, you only ever post myths, hearsay and blatant lies (like
> the sap thing).
>
> I am not the troll here, remember - this message started across a wide
> number of groups and not just oracle.

So are you blaming the OP or Daniel or me?

>
> I'm done here anyway, better things to do with my time; but, its recorded
> now so I will let readers of the thread judge (and read up) for themselves,
> the truth is out there as one good tv series puts it, and its certainly
> doesn't come from you!
>
> --
> Tony Rogerson
> SQL Server MVP
> http://sqlserverfaq.com - free video tutorials

>
>
> "DA Morgan" <damo...@psoug.org> wrote in message

> news:11406316...@jetspin.drizzle.com...
> > Tony Rogerson wrote:
> >>>I do get the sound track however. It's pretty hilarious. Don't you hate
> >>>it when demo's go wrong ?
> >>
> >>
> >> I'm not responsible for the Microsoft site - take it up with them if you
> >> can't play it.
> >>
> >> Yes, sometimes it is funny; however, that demo did not go wrong !
> >>
> >> The example SQL is here:
> >> http://www.sqlserverfaq.com/controls/kbase/store/KB_31_CONCURRENCY.zip
> >
> > Tony please take a good and considered look at the following:
> >
> > http://en.wikipedia.org/wiki/Internet_troll

jg
--
@home.com is bogus.

http://news.com.com/HP+Labs+marks+40th+with+high-tech+coffee+table/2100-1008_3-6041758.html

paul c

unread,

Feb 22, 2006, 8:55:40 PM2/22/06

to

Joel Garry wrote:
> Tony Rogerson wrote:
>
>>Guys, I simple corrected Joel or whomever it was on their original post and
>>its gone from there, for each myth posted about SQL Server 2005 I've
>>responded and kept responding, its not me driving this - its you guys!
>
>
> You mean where I said "So that's an average of 2TB/week. Do you really
> defend what Tony said:
> Windows/pair of SATA drives can do that? "
>
> Or where I said "Well, here's where I disagree. The design necessary
> to get SQL Server
> to handle TB's of data along with random transactional queries puts the
>
> onus on programmers to do it right. All you need is one newbie and you
>
> are screwed. Unless you use the new feature that makes it work like
> Oracle. So let's see: New unproven feature or risk of manual error.
> New unproven feature or risk of manual error. New unproven feature
> that probably has bugs (like with ALL vendor databases new features),

> or near-100% chance of manual error. ...

World seems full of people, not mentioning numbers, who throw quantities
around without specifying them. A week's got 168 hours in it, that's a
lot of disks in a year unless you own a disk company, or spend half your
time deleting instead of querying.

Apologies for not paying attention to all the details, even if being a
half-assed assembler programmer, I think I can see the big picture in
this particular case. *

I guess the phrase 'recorded history' has changed since Einstein's day.
Walmart won't last if they persist with their drill-down dogma, but
if they want my advice in the meantime, I'll take their money but freely
tell them they are wasting their lives.

One point in favour of all this history is that the big-time gigantic
systems have always been proved in the field, not in the lab.

> I'll pass.

I'm with you.

pc

ps: looking forward to watching the great innovators and artists of
modern hockey. Too bad there isn't a Fabian P to pronounce on what
really goes on in that game.