Anyone know where I can find some good resources to help us choose between 
SQL and Oracle ( Progress Openedge as well ) . Any comments on what you 
would choose ?? We are creating a new Warehouse Management System which wil 
manage our very large inventory.

Anyway comments suggestions welcome

Thanks
Paul 

this is something like comparing oranges and apples but lets try...

Some questions arise:
- what platform are you intend to work Unix, Windows ?

Well, progress is dead meat, forget it.

SQL Server has many features and looks good, but in mission critical systems 
it is still
five years behind of Oracle.

Oracle has been and will be RDBMS for many critical system, choose this.

Others are wellcome to comment ;)

"Paul" <paul@home.com> wrote in message news:5yWYf.4870$4S.829@edtnps82...
> Anyone know where I can find some good resources to help us choose between 
> SQL and Oracle ( Progress Openedge as well ) . Any comments on what you 
> would choose ?? We are creating a new Warehouse Management System which 
> wil manage our very large inventory.
>
> Anyway comments suggestions welcome
>
> Thanks
> Paul
> 

The user base for people using Microsoft SQL Server for their mission 
critical database is massive, probably in terms of installations far out 
numbers that of Oracle (notice I said installations rather than revenue).

Whilst SQL Server has a very strong BI offering that comes as part of the 
product and included in the cost - Oracle doesn't, it relies on extensions 
to its relational database and has no reporting facilities - you need to pay 
extra to get BI with Oracle.

Others are wellcome to comment ;)

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


"Jack" <none@INVALIDmail.com> wrote in message 
news:ug8Zf.124$_g4.0@read3.inet.fi...
> this is something like comparing oranges and apples but lets try...
>
> Some questions arise:
> - what platform are you intend to work Unix, Windows ?
>
> Well, progress is dead meat, forget it.
>
> SQL Server has many features and looks good, but in mission critical 
> systems it is still
> five years behind of Oracle.
>
> Oracle has been and will be RDBMS for many critical system, choose this.
>
> Others are wellcome to comment ;)
>
> "Paul" <paul@home.com> wrote in message news:5yWYf.4870$4S.829@edtnps82...
>> Anyone know where I can find some good resources to help us choose 
>> between SQL and Oracle ( Progress Openedge as well ) . Any comments on 
>> what you would choose ?? We are creating a new Warehouse Management 
>> System which wil manage our very large inventory.
>>
>> Anyway comments suggestions welcome
>>
>> Thanks
>> Paul
>>
>
> 

Paul wrote:
> Anyone know where I can find some good resources to help us choose between 
> SQL and Oracle ( Progress Openedge as well ) . Any comments on what you 
> would choose ?? We are creating a new Warehouse Management System which wil 
> manage our very large inventory.
> 
> Anyway comments suggestions welcome
> 
> Thanks
> Paul 

This type of question generally invites a bloody good flame war but I 
will try to present some of the deltas that flavour Oracle in the least
inflammatory way that I can.

Be very careful when comparing these two products as the verbiage may be
the same but the concepts and technologies can be very different. For
example a database in SQL Sever equals a schema in Oracle and has
absolutely nothing to do with the Oracle concept of a database. Neither
is a SQL Server instance the same concept as an Oracle instance.

Both products have log files but they work in very different ways. In
SQL Server one sizes log files to handle the largest transaction and its
rollback. In Oracle one can perform an infinitely large transaction
using a couple of small log files.

Both products have tables but that is pretty much where the similarity
ends. Oracle provides heap tables (same as a SQL Server table) but also
two types of global temporary tables (the tables aren't temporary ...
the data is), external tables, compressed tables, index organized
tables, nested tables, partitioned, and XML tables. Not to mention
objects such as sorted hash clusters.

Talk about indexes and in SQL Server you find BTree and Bitmap indexes.
In Oracle you will also find bitmap join, compressed, descending, 
function-based, reverse key, and no-segment indexes.

And this type of difference extends throughout the products. For example
SQL Server has no object types that perform the functions of Oracle's
packages, sequences, user-defined operators, rule-sets, and many more.

Look at security and you will find the differences are very substantial
as is the range of operating system options. You'll never run SQL Server
on Linux. And on Windows you will always be the target of every virus,
trojan, worm, and disgruntled employee that knows anything about the 
o/s. Be sure to look at auditing with Sarbanes-Oxley and similar laws
in mind.

How important is 7x24 operation? There is no SQL Server technology
equivalent to Oracle' Real Application Clusters. They are working hard
in Redmond to get it in a future version but that is years away. And
how important would a capability such as resumable transactions be as
Oracle provides with their DBMS_RESUMABLE built-in package?

Be sure too to look at the differences in the transaction models. They
are completely different. In Oracle reads never block writes and writes
never block reads and there are an infinite number of row-level locks.
Lock escalation does not exist.

Finally, in spite of marketing types naming things to help sell them,
the fact is that SQL Server's Enterprise Edition is approximately
equal to Oracle's Standard Edition. Oracle's Enterprise Edition contains
essentially nothing but features SQL Server does not offer. Only the
name "Enterprise" is the same.

Feel free to contact me off-line if you wish as I have no interest in
fueling the inevitable name-calling any further than I already have.
Also feel free to visit my web site "Morgan's Library" at www.psoug.org.
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)

they both work.

do you prefer a chevy pickup, or a dodge pickup?


but, this begs the question, WHY ARE YOU CREATING AN INVENTORY SYSTEM
FROM SCRATCH????????   go buy an application that fits your needs, and
use whatever they tell you to.

Doug wrote:
> they both work.
> 
> do you prefer a chevy pickup, or a dodge pickup?
> 
> 
> but, this begs the question, WHY ARE YOU CREATING AN INVENTORY SYSTEM
> FROM SCRATCH????????   go buy an application that fits your needs, and
> use whatever they tell you to.

Most applications are available supporting multiple database platforms.

Due diligence requires that first you make sure that the system complies
with the laws in your jurisdiction. If you can't achieve Sarbanes-Oxley
in the US or Basel II in Europe you can stop right there. And then there
is the entire issue around internal development to be considered.
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)

On Wed, 05 Apr 2006 21:22:09 GMT, Paul wrote:

>Anyone know where I can find some good resources to help us choose between 
>SQL and Oracle ( Progress Openedge as well ) . Any comments on what you 
>would choose ?? We are creating a new Warehouse Management System which wil 
>manage our very large inventory.
>
>Anyway comments suggestions welcome

Hi Paul,

The two major considerations are: what platform are you (or your staff)
already familiar with, and what features do you need.

Moving to a new platform is costly. You lose lots of productivity
attending courses or studying books. Then, you'll be less productive
while getting acquainted and using your new knowledge for real for the
first time. And you'll inevitably lose time cleaning up after using a
syntax that turns out to be disastrous in the new platform. If you're
familiar with either Oracle or SQL Server, stick with what you know!

Features: The platforms have a lot in common, but both also offer some
unique features. Visit MS' and Oracle's sites: both companies are very
good at pointing out the features their competitors lack. If any feature
that is critical or very important for your application is in one of the
products only, choose that product.

Price should be last on your list. Not because price is unimportant, but
because the other issues are even more important. You'll have to do your
own research, but here are some things that I have heard numerous times:
- Licensing cost for MS SQL Server is cheaper than for a comparable
  Oracle setup. Of course, you do save on OS licensing costs if you
  run Oracle on Linux.
- Oracle is more efficient. You have to spend less on hardware to get
  the same performance.
- Oracle databases are harder to manage. A single DBA can easily manage
  five or more SQL Servers, but you need at least three DBA's to manage
  a single Oracle instance.
(I can't quote any URL for any of those, though - they might but they
might be urban legends just as well. I encourage you to check the facts
and find either confirmation or denial of what I read)

And if price can't help you decide either, you'll just have to flip a
coin, I guess. Or consider where you feel more at home - at a platform
that sees it's advocates roaming the newsgroups of the competition in
order to win some souls, or at a platform that has a strong community pf
users who'll try to find out what's best for you?  <WINK>

You'll have noticed that my comments are fairly broad. I can't give more
details, though, as I have no experience with Oracle. Everything I know
about Oracle is hearsay. This precludes me from posting an in-depth
comparison of Oracle vs MS SQL Server.

-- 
Hugo Kornelis, SQL Server MVP

> Price should be last on your list. Not because price is unimportant, but
> because the other issues are even more important. You'll have to do your
> own research, but here are some things that I have heard numerous times:
> - Licensing cost for MS SQL Server is cheaper than for a comparable
>   Oracle setup. Of course, you do save on OS licensing costs if you
>   run Oracle on Linux.

Linux is cheaper, in fact it is free. However, if you use one of the big 
boys (Red Hat, etc), then you may end up paying a large amount for 
support costs. I've seen Linux cost more than Windows...and I've seen it 
cost less too. That being said, you can always run Oracle on Windows and 
it becomes a wash.

> - Oracle databases are harder to manage. A single DBA can easily manage
>   five or more SQL Servers, but you need at least three DBA's to manage
>   a single Oracle instance.

I would disagree here. At least, I would disagree if you are using 
Oracle 10g. I do caveat that with the notion that I have not experienced 
SQL Server 2005 yet. But I do have plenty of experience with SQL Server 
2000 and much more with Oracle. Oracle 10g is the most self-managing 
database Oracle has ever put out. It is much easier to manage than 
previous Oracle versions.

Cheers,
Brian


-- 
===================================================================

Brian Peasland
oracle_dba@nospam.peasland.net
http://www.peasland.net

Remove the "nospam." from the email address to email me.


"I can give it to you cheap, quick, and good.
Now pick two out of the three" - Unknown

Hugo Kornelis wrote:

> Moving to a new platform is costly. You lose lots of productivity
> attending courses or studying books.

The average time it takes competent (the operative word here being
competent) Windows admins to learn Linux or most flavours of UNIX
sufficient to work with a database product whether Oracle or Sybase
or whatever) is less than a week.

Oracle on any flavor of UNIX is identical to Oracle on Windows once
the initial installation is completed. The only skill required after
that is minor variations of basic DOS navigation skills.

If you were talking about moving an entire enterprise I would agree
with you. But one or two database servers. I'd come in and do it in a
single day and so would many consultants.
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)

On Thu, 06 Apr 2006 15:57:05 -0700, DA Morgan wrote:

>Hugo Kornelis wrote:
>
>> Moving to a new platform is costly. You lose lots of productivity
>> attending courses or studying books.
>
>The average time it takes competent (the operative word here being
>competent) Windows admins to learn Linux or most flavours of UNIX
>sufficient to work with a database product whether Oracle or Sybase
>or whatever) is less than a week.
(snip)

Hi DA,

I refered to the cost of moving to another database platform, ie someone
familiar with SQL Server moving to Oracle, or someone familiar with
Oracle moving to SQL Server.

-- 
Hugo Kornelis, SQL Server MVP

oh spare me.  sarbanes-oxley isn't going to tell you what db to run.

again, i think you are pretty new at corporate level implementations of
an application. installing the OS is TRIVIAL compared to developing the
app.
Developing the app is trivial compared to implmeneting it.
implementing it is trivial comparted to long term ramifications of
whether the app meets the business needs.

find a solution that works for someone else, steal it, buy it, or
borrow it. Writing it from scratch is a LAST resort.

Yes, it is very difficult to compare the two products, often Oracle DBA's 
like yourself concentrate purley on what the RDBMS engine has to offer.

They ignore that SQL Server is a product set that comes (as part of the 
cost), an enterprise ETL tool, an enterprise OLAP tool, Reporting tool, 
Notification Services, Service Broker (messaging tool), Textual Search, Data 
Mining, a powerful set of developer tools, SQL Profiler, Database Engine 
Tuning Advisor again - all come with the product.

With Oracle you pay for everything seperately and it costs a lot - the cost 
of the profiler and database tuning tool comparitive is more than the cost 
for SQL Serve itself!

The entry bar to 'Enterprise' has changed significantly, 5 - 10GB databases 
used to be classified 'enterprise' 10 years ago but they are common place 
now in the small / medium business tiers.

Most required Enterprise availability features are now in SQL Server 2005, 
there are a number of 'specialist' features like RAC's that aren't yet but 
it will come, only if business tells MS they want it though. I've heard so 
many sob stories around RAC, I have the impression in a real situation they 
don't really work.

There are different mindsets to creating a 24x7 environment, within the SQL 
space (and application tier) its all about scaling out; you can do that 
using peer to peer replication for instance, that can be implemented so that 
the servers are on different continents, you do need to think around data 
partitioning but it works.

I think oracle specialists like yourself and some others a frequent the 
oracle forums get judgement clouded and think that the way oracle have 
implemented solutions to solve business problems like availability is the 
only way to do things - that just isn't true and shows a complete arrogance 
within the industry - there are several ways to skin a cat and all have 
merits and downsides, RAC for instance - the skill level and man power 
required is high.

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

"DA Morgan" <damorgan@psoug.org> wrote in message 
news:1144339895.223894@yasure.drizzle.com...
> Paul wrote:
>> Anyone know where I can find some good resources to help us choose 
>> between SQL and Oracle ( Progress Openedge as well ) . Any comments on 
>> what you would choose ?? We are creating a new Warehouse Management 
>> System which wil manage our very large inventory.
>>
>> Anyway comments suggestions welcome
>>
>> Thanks
>> Paul
>
> This type of question generally invites a bloody good flame war but I will 
> try to present some of the deltas that flavour Oracle in the least
> inflammatory way that I can.
>
> Be very careful when comparing these two products as the verbiage may be
> the same but the concepts and technologies can be very different. For
> example a database in SQL Sever equals a schema in Oracle and has
> absolutely nothing to do with the Oracle concept of a database. Neither
> is a SQL Server instance the same concept as an Oracle instance.
>
> Both products have log files but they work in very different ways. In
> SQL Server one sizes log files to handle the largest transaction and its
> rollback. In Oracle one can perform an infinitely large transaction
> using a couple of small log files.
>
> Both products have tables but that is pretty much where the similarity
> ends. Oracle provides heap tables (same as a SQL Server table) but also
> two types of global temporary tables (the tables aren't temporary ...
> the data is), external tables, compressed tables, index organized
> tables, nested tables, partitioned, and XML tables. Not to mention
> objects such as sorted hash clusters.
>
> Talk about indexes and in SQL Server you find BTree and Bitmap indexes.
> In Oracle you will also find bitmap join, compressed, descending, 
> function-based, reverse key, and no-segment indexes.
>
> And this type of difference extends throughout the products. For example
> SQL Server has no object types that perform the functions of Oracle's
> packages, sequences, user-defined operators, rule-sets, and many more.
>
> Look at security and you will find the differences are very substantial
> as is the range of operating system options. You'll never run SQL Server
> on Linux. And on Windows you will always be the target of every virus,
> trojan, worm, and disgruntled employee that knows anything about the o/s. 
> Be sure to look at auditing with Sarbanes-Oxley and similar laws
> in mind.
>
> How important is 7x24 operation? There is no SQL Server technology
> equivalent to Oracle' Real Application Clusters. They are working hard
> in Redmond to get it in a future version but that is years away. And
> how important would a capability such as resumable transactions be as
> Oracle provides with their DBMS_RESUMABLE built-in package?
>
> Be sure too to look at the differences in the transaction models. They
> are completely different. In Oracle reads never block writes and writes
> never block reads and there are an infinite number of row-level locks.
> Lock escalation does not exist.
>
> Finally, in spite of marketing types naming things to help sell them,
> the fact is that SQL Server's Enterprise Edition is approximately
> equal to Oracle's Standard Edition. Oracle's Enterprise Edition contains
> essentially nothing but features SQL Server does not offer. Only the
> name "Enterprise" is the same.
>
> Feel free to contact me off-line if you wish as I have no interest in
> fueling the inevitable name-calling any further than I already have.
> Also feel free to visit my web site "Morgan's Library" at www.psoug.org.
> -- 
> Daniel A. Morgan
> http://www.psoug.org
> damorgan@x.washington.edu
> (replace x with u to respond) 

What about the time differential in a person actually doing something - it 
takes a lot longer to fart about running queries in a DOS window then it 
does using a nice feature rich UI.

Your comment here really does show you are class room bound.

There is more to this that just installing the box, there are DR strategies, 
connectivity issues arrising from clients being on Active Directory and your 
box being on something else, your software needs to work with a different 
set of db access drivers.

The last thing a company needs is to start implementing different vendor 
databases, unless the plan is to migrate - skill set differences are high on 
the database products themselves, staffing costs finding a level of 
expertise in both products is high, you'd probably have to have multiple 
staff because you will stretch to find real experts in both competancies.

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

"DA Morgan" <damorgan@psoug.org> wrote in message 
news:1144364223.213063@yasure.drizzle.com...
> Hugo Kornelis wrote:
>
>> Moving to a new platform is costly. You lose lots of productivity
>> attending courses or studying books.
>
> The average time it takes competent (the operative word here being
> competent) Windows admins to learn Linux or most flavours of UNIX
> sufficient to work with a database product whether Oracle or Sybase
> or whatever) is less than a week.
>
> Oracle on any flavor of UNIX is identical to Oracle on Windows once
> the initial installation is completed. The only skill required after
> that is minor variations of basic DOS navigation skills.
>
> If you were talking about moving an entire enterprise I would agree
> with you. But one or two database servers. I'd come in and do it in a
> single day and so would many consultants.
> -- 
> Daniel A. Morgan
> http://www.psoug.org
> damorgan@x.washington.edu
> (replace x with u to respond) 

Hugo Kornelis wrote:
> On Thu, 06 Apr 2006 15:57:05 -0700, DA Morgan wrote:
> 
>> Hugo Kornelis wrote:
>>
>>> Moving to a new platform is costly. You lose lots of productivity
>>> attending courses or studying books.
>> The average time it takes competent (the operative word here being
>> competent) Windows admins to learn Linux or most flavours of UNIX
>> sufficient to work with a database product whether Oracle or Sybase
>> or whatever) is less than a week.
> (snip)
> 
> Hi DA,
> 
> I refered to the cost of moving to another database platform, ie someone
> familiar with SQL Server moving to Oracle, or someone familiar with
> Oracle moving to SQL Server.

Sorry I misunderstood. Changing platforms can be expensive. But it
seemed to me, from the OP's question, that they were not considering
a change but rather an initial platform.
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)

Doug wrote:
> oh spare me.  sarbanes-oxley isn't going to tell you what db to run.

No. But the auditor that signs off on your financial statement is going
to tell you that if a DBA or System Administrator can make unaudited
changes they will refuse to sign. I saw that happen with a public
company here in Seattle last year. You've never seen such a scramble
in your life. The replatformed in 3 weeks.

The issue is not which product ... but rather the auditing and security
provided by the product. SQL Server, or perhaps more correctly Windows,
just isn't there yet.

There is a reason most large corporate line-of-business apps run on some
flavour of UNIX. It isn't just a desire to waste money.
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)

We can go either way ( Unix/Windows ) doesn't really matter to us - Why 
would you say Progress is dead meat ? What has you experience been that you 
can justify that ? Just curious is all . I tend to agree with you but the 
person I report to will want to know why.


"Jack" <none@INVALIDmail.com> wrote in message 
news:ug8Zf.124$_g4.0@read3.inet.fi...
> this is something like comparing oranges and apples but lets try...
>
> Some questions arise:
> - what platform are you intend to work Unix, Windows ?
>
> Well, progress is dead meat, forget it.
>
> SQL Server has many features and looks good, but in mission critical 
> systems it is still
> five years behind of Oracle.
>
> Oracle has been and will be RDBMS for many critical system, choose this.
>
> Others are wellcome to comment ;)
>
> "Paul" <paul@home.com> wrote in message news:5yWYf.4870$4S.829@edtnps82...
>> Anyone know where I can find some good resources to help us choose 
>> between SQL and Oracle ( Progress Openedge as well ) . Any comments on 
>> what you would choose ?? We are creating a new Warehouse Management 
>> System which wil manage our very large inventory.
>>
>> Anyway comments suggestions welcome
>>
>> Thanks
>> Paul
>>
>
> 

Paul wrote:
> We can go either way ( Unix/Windows ) doesn't really matter to us - Why 
> would you say Progress is dead meat ? What has you experience been that you 
> can justify that ? Just curious is all . I tend to agree with you but the 
> person I report to will want to know why.

I'd agree Progress is no longer viable.

The demands for meeting governmental regulations (auditing and security)
plus the ability of organizations to find competent internal and
external resources mitigates against use of any RDBMS other than the top
4 or 5. If you can't get the staffing you need and the phone/web support
you need no organization is going to be happy in the long run.

Take a serious look at the toolset for recovering from media failure or
corruption. Progress isn't in the running.
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)

Hugo Kornelis wrote:
> - Oracle is more efficient. You have to spend less on hardware to get
>   the same performance.

I'd put a question mark here - at least for the general way you put it. 
  It always depends on what you do with the tool.  SQL Server's network 
protocol seems to be more efficient as it allows faster round trip. 
This can make a difference in applications that touch the DB frequently 
accessing only small data sets.

Kind regards

	robert

On Mon, 10 Apr 2006 17:19:42 +0200, Robert Klemme wrote:

>Hugo Kornelis wrote:
>> - Oracle is more efficient. You have to spend less on hardware to get
>>   the same performance.
>
>I'd put a question mark here - at least for the general way you put it. 

Hi Robert,

I'd put a question mark at all the points I mentioned in that list. Like
I said, I only echoed what appear to be widespread ideas. I explicitly
encouraged Paul to do some check ing - widespread idead CAN be true, but
they can be urban legend just as well.

-- 
Hugo Kornelis, SQL Server MVP

Robert Klemme wrote:
> Hugo Kornelis wrote:
>> - Oracle is more efficient. You have to spend less on hardware to get
>>   the same performance.
> 
> I'd put a question mark here - at least for the general way you put it. 
>  It always depends on what you do with the tool.  SQL Server's network 
> protocol seems to be more efficient as it allows faster round trip. This 
> can make a difference in applications that touch the DB frequently 
> accessing only small data sets.
> 
> Kind regards
> 
>     robert

Actually I've tested this myth in my lab and, so far have never been
able to duplicate it, given a decently tuned system. Do you have any
benchmarks that support it?
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)

Arrrrr. the lab and tester that has a massive and tunneled bias towards 
oracle and is so anti windows that he's bricked up all the windows in his 
appartment.

I'm not suprised, based on past evidence that you cannot repro anything that 
SQL Server on the windows platform is better at!

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

"DA Morgan" <damorgan@psoug.org> wrote in message 
news:1144786257.786355@yasure.drizzle.com...
> Robert Klemme wrote:
>> Hugo Kornelis wrote:
>>> - Oracle is more efficient. You have to spend less on hardware to get
>>>   the same performance.
>>
>> I'd put a question mark here - at least for the general way you put it. 
>> It always depends on what you do with the tool.  SQL Server's network 
>> protocol seems to be more efficient as it allows faster round trip. This 
>> can make a difference in applications that touch the DB frequently 
>> accessing only small data sets.
>>
>> Kind regards
>>
>>     robert
>
> Actually I've tested this myth in my lab and, so far have never been
> able to duplicate it, given a decently tuned system. Do you have any
> benchmarks that support it?
> -- 
> Daniel A. Morgan
> http://www.psoug.org
> damorgan@x.washington.edu
> (replace x with u to respond) 

DA Morgan wrote:
> Robert Klemme wrote:
>> Hugo Kornelis wrote:
>>> - Oracle is more efficient. You have to spend less on hardware to get
>>>   the same performance.
>>
>> I'd put a question mark here - at least for the general way you put 
>> it.  It always depends on what you do with the tool.  SQL Server's 
>> network protocol seems to be more efficient as it allows faster round 
>> trip. This can make a difference in applications that touch the DB 
>> frequently accessing only small data sets.
>>
>> Kind regards
>>
>>     robert
> 
> Actually I've tested this myth in my lab and, so far have never been
> able to duplicate it, given a decently tuned system. Do you have any
> benchmarks that support it?

I don't have figures at hand but it should be fairly easy to set up a 
test.  All you need is JDBC drivers for both (I recommend jTDS for SQL 
Server), an installation of each product and then test timings for SQL 
that does nothing ("SELECT 1 FROM DUAL" for Oracle or "SELECT 1" for SQL 
Server) - you can even use a PreparedStatement to save the parsing.  You 
don't even need to specially tune the db.

Kind regards

	robert

Tony Rogerson wrote:
> Arrrrr. the lab and tester that has a massive and tunneled bias towards 
> oracle and is so anti windows that he's bricked up all the windows in his 
> appartment.
> 
> I'm not suprised, based on past evidence that you cannot repro anything that 
> SQL Server on the windows platform is better at!

I'm not surprised. Much of the resistance to SQL Server has nothing to
do with SQL Server but rather Windows and much of it quite rationally.

One factor we routinely see with Oracle is that one can take a single
piece of hardware. First load Oracle on Windows XP SP2 on it and run a
load. Then format the hard disk and perform the exact same test using
RedHat Linux. The difference in scalability and performance is hard to
miss.
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)

Robert Klemme wrote:
> DA Morgan wrote:
>> Robert Klemme wrote:
>>> Hugo Kornelis wrote:
>>>> - Oracle is more efficient. You have to spend less on hardware to get
>>>>   the same performance.
>>>
>>> I'd put a question mark here - at least for the general way you put 
>>> it.  It always depends on what you do with the tool.  SQL Server's 
>>> network protocol seems to be more efficient as it allows faster round 
>>> trip. This can make a difference in applications that touch the DB 
>>> frequently accessing only small data sets.
>>>
>>> Kind regards
>>>
>>>     robert
>>
>> Actually I've tested this myth in my lab and, so far have never been
>> able to duplicate it, given a decently tuned system. Do you have any
>> benchmarks that support it?
> 
> I don't have figures at hand but it should be fairly easy to set up a 
> test.  All you need is JDBC drivers for both (I recommend jTDS for SQL 
> Server), an installation of each product and then test timings for SQL 
> that does nothing ("SELECT 1 FROM DUAL" for Oracle or "SELECT 1" for SQL 
> Server) - you can even use a PreparedStatement to save the parsing.  You 
> don't even need to specially tune the db.
> 
> Kind regards
> 
>     robert

That's what I've done. I've never seen it. But running SELECT 1 may be
more of a test of how clueless the person doing the test is than of the
hardware and operating system. As the two events are not at all similar
in what they do.
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)

> One factor we routinely see with Oracle is that one can take a single
> piece of hardware. First load Oracle on Windows XP SP2 on it and run a
> load. Then format the hard disk and perform the exact same test using
> RedHat Linux. The difference in scalability and performance is hard to
> miss.

So you are comparing an OS mean't for the desktop (XP) against an OS mean't 
for a server environment.

Like I say, your bias of anti-MS tunnels your judgement.

If I get some free time I'll try a comparison between linux and windows 2003 
r2 server edition which is a more comparable test.

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

"DA Morgan" <damorgan@psoug.org> wrote in message 
news:1144897930.909259@yasure.drizzle.com...
> Tony Rogerson wrote:
>> Arrrrr. the lab and tester that has a massive and tunneled bias towards 
>> oracle and is so anti windows that he's bricked up all the windows in his 
>> appartment.
>>
>> I'm not suprised, based on past evidence that you cannot repro anything 
>> that SQL Server on the windows platform is better at!
>
> I'm not surprised. Much of the resistance to SQL Server has nothing to
> do with SQL Server but rather Windows and much of it quite rationally.
>
> One factor we routinely see with Oracle is that one can take a single
> piece of hardware. First load Oracle on Windows XP SP2 on it and run a
> load. Then format the hard disk and perform the exact same test using
> RedHat Linux. The difference in scalability and performance is hard to
> miss.
> -- 
> Daniel A. Morgan
> http://www.psoug.org
> damorgan@x.washington.edu
> (replace x with u to respond) 

DA Morgan wrote:
> That's what I've done. I've never seen it. But running SELECT 1 may be
> more of a test of how clueless the person doing the test is than of the
> hardware and operating system. As the two events are not at all similar
> in what they do.
Rating high on the cluelessness on SELECT 1 myself, would you care to 
elaborate?
My working hypothesis is that SELECT 1 returns a result set of one row 
and one column to the client.
So does SELECT 1 FROM DUAL, AFAIK.
If you are alluding to the fact that Oracle may actually do a table 
lookup on DUAL then that could hardly be the competitions fault and you 
would be free to supply a better approach for how to return a result set 
without driving IO.

Cheers
Serge
-- 
Serge Rielau
DB2 Solutions Development
IBM Toronto Lab

Might be aluding to this....

SELECT 1 FROM ACTIVITY_ENTRY WHERE (1=0) FOR UPDATE

Which Oracle gives a table / index scan rather than realising it doesn't 
need to do anything.

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

"Serge Rielau" <srielau@ca.ibm.com> wrote in message 
news:4a6vb6FrjsldU1@individual.net...
> DA Morgan wrote:
>> That's what I've done. I've never seen it. But running SELECT 1 may be
>> more of a test of how clueless the person doing the test is than of the
>> hardware and operating system. As the two events are not at all similar
>> in what they do.
> Rating high on the cluelessness on SELECT 1 myself, would you care to 
> elaborate?
> My working hypothesis is that SELECT 1 returns a result set of one row and 
> one column to the client.
> So does SELECT 1 FROM DUAL, AFAIK.
> If you are alluding to the fact that Oracle may actually do a table lookup 
> on DUAL then that could hardly be the competitions fault and you would be 
> free to supply a better approach for how to return a result set without 
> driving IO.
>
> Cheers
> Serge
> -- 
> Serge Rielau
> DB2 Solutions Development
> IBM Toronto Lab 

Tony Rogerson wrote:
>> One factor we routinely see with Oracle is that one can take a single
>> piece of hardware. First load Oracle on Windows XP SP2 on it and run a
>> load. Then format the hard disk and perform the exact same test using
>> RedHat Linux. The difference in scalability and performance is hard to
>> miss.
> 
> So you are comparing an OS mean't for the desktop (XP) against an OS mean't 
> for a server environment.

No. When and where did I say "Desktop"? You said it I didn't.

But the same test has been run against all 32bit Windows Server
implementations with the same result. Hope that clarifies it.

> Like I say, your bias of anti-MS tunnels your judgement.

I use a lot of MS products. Just not SQL Server for line-of-business
apps. That hardly qualifies as an anti-MS bias. You just don't seem to
like the message that there are somethings where Windows and/or SQL
Server are not the best tool for the job. That is not my bias. That is
the result of benchmarking.

> If I get some free time I'll try a comparison between linux and windows 2003 
> r2 server edition which is a more comparable test.

Given your knowledge of Oracle and ability to set it up properly I think
one can pretty much assume the result is predetermined.
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)

Serge Rielau wrote:
> DA Morgan wrote:
>> That's what I've done. I've never seen it. But running SELECT 1 may be
>> more of a test of how clueless the person doing the test is than of the
>> hardware and operating system. As the two events are not at all similar
>> in what they do.
> Rating high on the cluelessness on SELECT 1 myself, would you care to 
> elaborate?

Depending on which version of Oracle dual may or may not be a table and
calls using dual may behave in very different ways.

The OP assumed something not in evidence.
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)

Tony Rogerson wrote:
> Might be aluding to this....
> 
> SELECT 1 FROM ACTIVITY_ENTRY WHERE (1=0) FOR UPDATE
> 
> Which Oracle gives a table / index scan rather than realising it doesn't 
> need to do anything.

In different versions dual behaves very differently. A comparison of 
SELECT 1 from Oracle (version dependant) and SELECT 1 in SQL Server are
not necessarily comparable.
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)

"DA Morgan" <damorgan@psoug.org> wrote in message
news:1144953446.98074@yasure.drizzle.com...
> Tony Rogerson wrote:
> >> One factor we routinely see with Oracle is that one can take a single
> >> piece of hardware. First load Oracle on Windows XP SP2 on it and run a
> >> load. Then format the hard disk and perform the exact same test using
> >> RedHat Linux. The difference in scalability and performance is hard to
> >> miss.
> >
> > So you are comparing an OS mean't for the desktop (XP) against an OS
mean't
> > for a server environment.
>
> No. When and where did I say "Desktop"? You said it I didn't.

Actually you did by the very fact you said XP.  That's an OS intended for
desktop use and as such is tuned very differently than Server 2003.

>
> But the same test has been run against all 32bit Windows Server
> implementations with the same result. Hope that clarifies it.

What about 64 bit versions out of curiosity.

> -- 
> Daniel A. Morgan
> http://www.psoug.org
> damorgan@x.washington.edu
> (replace x with u to respond)

> No. When and where did I say "Desktop"? You said it I didn't.

XP is a desktop OS, Windows Server editions are optimised for IO and 
throughput and as such give better throughput than XP for applications like 
SQL Server. There are also other considerations like, one important one is 
disk cache - is it turned on / off, this will have a  massive impact on 
write operations.

> But the same test has been run against all 32bit Windows Server
> implementations with the same result. Hope that clarifies it.

On checking TPC there is a comparitive benchmark where SQL Server beats 
Oracle hands down on the same hardware (HP Integrity Superdome), SQL 
Server -> 1.2million; Oracle -> 1million tpmC 
(http://tpc.org/tpcc/results/tpcc_perf_results.asp) - no bias there!

What about the 64bit - you still aren't comparing like with like. The only 
thing it clarifies is you have little experience or knowledge of the windows 
platform, your skill level on windows is that of a 'user'.

> I use a lot of MS products. Just not SQL Server for line-of-business
> apps. That hardly qualifies as an anti-MS bias. You just don't seem to
> like the message that there are somethings where Windows and/or SQL
> Server are not the best tool for the job. That is not my bias. That is
> the result of benchmarking.

Yes, you probably do - but as a user and not an architect or system engineer 
setting and specifying kit. Your bias and anti-windows comes through on my 
past experience with you and your postings and online attitude.

Your benchmarks are floored, even from what you've posted here that you used 
XP that is quite evident.

> Given your knowledge of Oracle and ability to set it up properly I think
> one can pretty much assume the result is predetermined.

Yes, unfortunetly you need to tweak and configure oracle to make it work 
well, that is not the case with SQL Server; so long as you get the hardware 
and disk configuration and placement of database files correct then you 
don't have to start tweaking the engine.

And yes, my findings will be biased and for one reason - because I have no 
knowledge as a SE for linux, just as you in the windows environment.

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

"DA Morgan" <damorgan@psoug.org> wrote in message 
news:1144953446.98074@yasure.drizzle.com...
> Tony Rogerson wrote:
>>> One factor we routinely see with Oracle is that one can take a single
>>> piece of hardware. First load Oracle on Windows XP SP2 on it and run a
>>> load. Then format the hard disk and perform the exact same test using
>>> RedHat Linux. The difference in scalability and performance is hard to
>>> miss.
>>
>> So you are comparing an OS mean't for the desktop (XP) against an OS 
>> mean't for a server environment.
>
> No. When and where did I say "Desktop"? You said it I didn't.
>
> But the same test has been run against all 32bit Windows Server
> implementations with the same result. Hope that clarifies it.
>
>> Like I say, your bias of anti-MS tunnels your judgement.
>
> I use a lot of MS products. Just not SQL Server for line-of-business
> apps. That hardly qualifies as an anti-MS bias. You just don't seem to
> like the message that there are somethings where Windows and/or SQL
> Server are not the best tool for the job. That is not my bias. That is
> the result of benchmarking.
>
>> If I get some free time I'll try a comparison between linux and windows 
>> 2003 r2 server edition which is a more comparable test.
>
> Given your knowledge of Oracle and ability to set it up properly I think
> one can pretty much assume the result is predetermined.
> -- 
> Daniel A. Morgan
> http://www.psoug.org
> damorgan@x.washington.edu
> (replace x with u to respond) 

Greg D. Moore (Strider) wrote:

>> No. When and where did I say "Desktop"? You said it I didn't.
> 
> Actually you did by the very fact you said XP.  That's an OS intended for
> desktop use and as such is tuned very differently than Server 2003.

http://www.microsoft.com/technet/prodtechnol/winxppro/default.mspx
You tell that to Microsoft. They don't listen to me.  ;-)

>> But the same test has been run against all 32bit Windows Server
>> implementations with the same result. Hope that clarifies it.
> 
> What about 64 bit versions out of curiosity.

Don't have hardware available from which I can perform an
equals-to-equals evaluation. When I do ... I will.
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)

Tony Rogerson wrote:
>> No. When and where did I say "Desktop"? You said it I didn't.
> 
> XP is a desktop OS, Windows Server editions are optimised for IO and 
> throughput and as such give better throughput than XP for applications like 
> SQL Server. There are also other considerations like, one important one is 
> disk cache - is it turned on / off, this will have a  massive impact on 
> write operations.

I have a copy of 2003. I'll rerun the comparison this summer during the 
break.
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)

> I have a copy of 2003. I'll rerun the comparison this summer during the 
> break.

There is no point, I'd rather trust the TPC for independence instead of what 
you will produce ie. a biased benchmark based on your anti-microsoft stance 
and lack of technical ability with that platform.

Like I say, check out the benchmark on tpc.org.

On checking TPC there is a comparitive benchmark where SQL Server beats 
Oracle hands down on the same hardware (HP Integrity Superdome), SQL 
Server -> 1.2million; Oracle -> 1million tpmC 
(http://tpc.org/tpcc/results/tpcc_perf_results.asp) - no bias there!

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

"DA Morgan" <damorgan@psoug.org> wrote in message 
news:1145209613.444622@yasure.drizzle.com...
> Tony Rogerson wrote:
>>> No. When and where did I say "Desktop"? You said it I didn't.
>>
>> XP is a desktop OS, Windows Server editions are optimised for IO and 
>> throughput and as such give better throughput than XP for applications 
>> like SQL Server. There are also other considerations like, one important 
>> one is disk cache - is it turned on / off, this will have a  massive 
>> impact on write operations.
>
> I have a copy of 2003. I'll rerun the comparison this summer during the 
> break.
> -- 
> Daniel A. Morgan
> http://www.psoug.org
> damorgan@x.washington.edu
> (replace x with u to respond) 

> http://www.microsoft.com/technet/prodtechnol/winxppro/default.mspx
> You tell that to Microsoft. They don't listen to me.  ;-)

Linking to the XP website proves only one thing - you are clutching at 
straws, I might as well point to oracle.com.

If there was any technical merit in your argument you would point at an 
article that explained the differences between XP Pro and the windows server 
family.

The TPC benchmark seems to have fallen on deaf ears - I would prefer a 
discussion on that - just how you are going to rebuke that should be very 
interesting if not entertaining.

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

"DA Morgan" <damorgan@psoug.org> wrote in message 
news:1145209558.153613@yasure.drizzle.com...
> Greg D. Moore (Strider) wrote:
>
>>> No. When and where did I say "Desktop"? You said it I didn't.
>>
>> Actually you did by the very fact you said XP.  That's an OS intended for
>> desktop use and as such is tuned very differently than Server 2003.
>
> http://www.microsoft.com/technet/prodtechnol/winxppro/default.mspx
> You tell that to Microsoft. They don't listen to me.  ;-)
>
>>> But the same test has been run against all 32bit Windows Server
>>> implementations with the same result. Hope that clarifies it.
>>
>> What about 64 bit versions out of curiosity.
>
> Don't have hardware available from which I can perform an
> equals-to-equals evaluation. When I do ... I will.
> -- 
> Daniel A. Morgan
> http://www.psoug.org
> damorgan@x.washington.edu
> (replace x with u to respond) 

Tony Rogerson wrote:
>> I have a copy of 2003. I'll rerun the comparison this summer during the 
>> break.
> 
> There is no point, I'd rather trust the TPC for independence instead of what 
> you will produce ie. a biased benchmark based on your anti-microsoft stance 
> and lack of technical ability with that platform.

Trust TPC? ROFL!

Their benchmarks are meaningless except to marketing departments and
CTO's looking for cover for their decisions.
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)

Tony Rogerson wrote:

> If there was any technical merit in your argument you would point at an 
> article that explained the differences between XP Pro and the windows server 
> family.

Given that you live in a Windows only world your defense of the one
and only operating system SQL Server runs on seems a bit disingenuous.
Oracle runs on 19 separate operating systems and I routinely work with
Windows, several flavors of Linux, Solaris, HP/UX, and AIX. It matters
to an Oracle architect which o/s is used ... you, on the other hand,
have a choice of Windows or Windows. Congratulations on being able to
make that choice. Which vendor ... oh lets use Microsoft. What should
we benchmark it against? Microsoft again. And the winner is ... drum
roll please ... Microsoft Windows.

> The TPC benchmark seems to have fallen on deaf ears

The ears aren't deaf? Every once in awhile we need a good laugh.

Disclaimer: I am sending this email from a Windows machine and have ton
of Microsoft products here at the U. Heck I used to teach in Mary Gates
Hall and now teach in the Paul Allen Computing Center. But give the
hyperbole a break every once in awhile. There are other competent companies.
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)

> Given that you live in a Windows only world your defense of the one
> and only operating system SQL Server runs on seems a bit disingenuous.

That 'rant' has nothing to do with the fact you chose the desktop version of 
windows aka Windows XP to do your benchmark (as you have stated), my point 
all along has been your benchmark is floored and obviously your knowledge 
and understanding of the windows platform - I would hope that if you have 
any clients outside of the classroom you don't suggest they run their 
production databases on Windows XP, unless the database is required for a 
single user / application, for instance a on the road salesman who stores 
info locally and replicates back into the central database - which would 
run, not on XP but Server editions!

> I routinely work with
> Windows, several flavors of Linux, Solaris, HP/UX, and AIX. It matters
> to an Oracle architect which o/s is used

Perhaps thats what your problem is - knowledge of many, expert in none.

> have a choice of Windows or Windows. Congratulations on being able to
> make that choice. Which vendor ... oh lets use Microsoft. What should
> we benchmark it against? Microsoft again. And the winner is ... drum
> roll please ... Microsoft Windows.

So long as it does what it says on the tin and is able to give my client a 
business solution then it works for me (and my client).

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

"DA Morgan" <damorgan@psoug.org> wrote in message 
news:1145225400.866564@yasure.drizzle.com...
> Tony Rogerson wrote:
>
>> If there was any technical merit in your argument you would point at an 
>> article that explained the differences between XP Pro and the windows 
>> server family.
>
> Given that you live in a Windows only world your defense of the one
> and only operating system SQL Server runs on seems a bit disingenuous.
> Oracle runs on 19 separate operating systems and I routinely work with
> Windows, several flavors of Linux, Solaris, HP/UX, and AIX. It matters
> to an Oracle architect which o/s is used ... you, on the other hand,
> have a choice of Windows or Windows. Congratulations on being able to
> make that choice. Which vendor ... oh lets use Microsoft. What should
> we benchmark it against? Microsoft again. And the winner is ... drum
> roll please ... Microsoft Windows.
>
>> The TPC benchmark seems to have fallen on deaf ears
>
> The ears aren't deaf? Every once in awhile we need a good laugh.
>
> Disclaimer: I am sending this email from a Windows machine and have ton
> of Microsoft products here at the U. Heck I used to teach in Mary Gates
> Hall and now teach in the Paul Allen Computing Center. But give the
> hyperbole a break every once in awhile. There are other competent 
> companies.
> -- 
> Daniel A. Morgan
> http://www.psoug.org
> damorgan@x.washington.edu
> (replace x with u to respond) 

Tony Rogerson (tonyrogerson@sqlserverfaq.com) writes:
> There is no point, I'd rather trust the TPC for independence instead of
> what you will produce ie. a biased benchmark based on your
> anti-microsoft stance and lack of technical ability with that platform. 
> 
> Like I say, check out the benchmark on tpc.org.
> 
> On checking TPC there is a comparitive benchmark where SQL Server beats 
> Oracle hands down on the same hardware (HP Integrity Superdome), SQL 
> Server -> 1.2million; Oracle -> 1million tpmC 
> (http://tpc.org/tpcc/results/tpcc_perf_results.asp) - no bias there!

I think I agree with DA here. TPC is good if all you want to do is 
size-comparisons of male organs, but if you want to test for your actual
application, you should run your own benchmark. And keep in mind that
that benchmark applies to that application, and necessarily not any other
application.

As I understood it, DA's benchmark is for the same Oracle application 
on Windows and Linux, which should be a fairly trivial benchmark to run.
(But it may say more about Oracle's implementation on the two operating
systems, that it tells about the operating systems themselves.)

Running a benchmark for the same application running SQL Server and 
Oracle is a far more devilish game.



-- 
Erland Sommarskog, SQL Server MVP, esquel@sommarskog.se

Books Online for SQL Server 2005 at
http://www.microsoft.com/technet/prodtechnol/sql/2005/downloads/books.mspx
Books Online for SQL Server 2000 at
http://www.microsoft.com/sql/prodinfo/previousversions/books.mspx

DA Morgan (damorgan@psoug.org) writes:
> Greg D. Moore (Strider) wrote:
>>> No. When and where did I say "Desktop"? You said it I didn't.
>> 
>> Actually you did by the very fact you said XP.  That's an OS intended for
>> desktop use and as such is tuned very differently than Server 2003.
> 
> http://www.microsoft.com/technet/prodtechnol/winxppro/default.mspx
> You tell that to Microsoft. They don't listen to me.  ;-)

I may be dense, but on the top this page, it says:

  TechNet Home > Products & Technologies > Desktop Operating Systems

What point were you trying to make?



-- 
Erland Sommarskog, SQL Server MVP, esquel@sommarskog.se

Books Online for SQL Server 2005 at
http://www.microsoft.com/technet/prodtechnol/sql/2005/downloads/books.mspx
Books Online for SQL Server 2000 at
http://www.microsoft.com/sql/prodinfo/previousversions/books.mspx

I think that was my point about the TPC benchmark is that the benchmark 
for -C is using the same application specification and in this case on the 
same kit which to me rings a half decent comparison.

But I do agree that people MUST benchmark their own application then use TPC 
which is, I agree, a simple marketing excercise, but it does show that 
products don't fall apart at such transaction volumes.

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

"Erland Sommarskog" <esquel@sommarskog.se> wrote in message 
news:Xns97A86E3709F54Yazorman@127.0.0.1...
> Tony Rogerson (tonyrogerson@sqlserverfaq.com) writes:
>> There is no point, I'd rather trust the TPC for independence instead of
>> what you will produce ie. a biased benchmark based on your
>> anti-microsoft stance and lack of technical ability with that platform.
>>
>> Like I say, check out the benchmark on tpc.org.
>>
>> On checking TPC there is a comparitive benchmark where SQL Server beats
>> Oracle hands down on the same hardware (HP Integrity Superdome), SQL
>> Server -> 1.2million; Oracle -> 1million tpmC
>> (http://tpc.org/tpcc/results/tpcc_perf_results.asp) - no bias there!
>
> I think I agree with DA here. TPC is good if all you want to do is
> size-comparisons of male organs, but if you want to test for your actual
> application, you should run your own benchmark. And keep in mind that
> that benchmark applies to that application, and necessarily not any other
> application.
>
> As I understood it, DA's benchmark is for the same Oracle application
> on Windows and Linux, which should be a fairly trivial benchmark to run.
> (But it may say more about Oracle's implementation on the two operating
> systems, that it tells about the operating systems themselves.)
>
> Running a benchmark for the same application running SQL Server and
> Oracle is a far more devilish game.
>
>
>
> -- 
> Erland Sommarskog, SQL Server MVP, esquel@sommarskog.se
>
> Books Online for SQL Server 2005 at
> http://www.microsoft.com/technet/prodtechnol/sql/2005/downloads/books.mspx
> Books Online for SQL Server 2000 at
> http://www.microsoft.com/sql/prodinfo/previousversions/books.mspx 

Erland Sommarskog wrote:
> Tony Rogerson (tonyrogerson@sqlserverfaq.com) writes:
>> There is no point, I'd rather trust the TPC for independence instead of
>> what you will produce ie. a biased benchmark based on your
>> anti-microsoft stance and lack of technical ability with that platform. 
>>
>> Like I say, check out the benchmark on tpc.org.
>>
>> On checking TPC there is a comparitive benchmark where SQL Server beats 
>> Oracle hands down on the same hardware (HP Integrity Superdome), SQL 
>> Server -> 1.2million; Oracle -> 1million tpmC 
>> (http://tpc.org/tpcc/results/tpcc_perf_results.asp) - no bias there!
> 
> I think I agree with DA here. TPC is good if all you want to do is 
> size-comparisons of male organs, but if you want to test for your actual
> application, you should run your own benchmark. And keep in mind that
> that benchmark applies to that application, and necessarily not any other
> application.
> 
> As I understood it, DA's benchmark is for the same Oracle application 
> on Windows and Linux, which should be a fairly trivial benchmark to run.
> (But it may say more about Oracle's implementation on the two operating
> systems, that it tells about the operating systems themselves.)
> 
> Running a benchmark for the same application running SQL Server and 
> Oracle is a far more devilish game.

Precisely. Because then you are dealing with SQL*Net vs another
connection mechanism. You are dealing with array processing versus
cursors, and lots of other real-world compromises we all make when
we decide whether to implement best-of-breed functionality or try
to stay generic. I think Tony, in his zeal to defend turf, missed
the point I was trying to make. Next time I will try to be clearer.
-- 
Daniel A. Morgan
http://www.psoug.org
damorgan@x.washington.edu
(replace x with u to respond)