Database market share 2004

[Jonathan Leffler]

Re: Database market share 2004

Leythos wrote:[color=blue]
> ... they run 24x7x365 and do so for years.[/color]

At least seven years at a time if your claim is correct...

:-)

It probably means you're still on Windows 98, too; remarkable!

--
Jonathan Leffler #include <disclaimer.h >
Email: jleffler@earthl ink.net, jleffler@us.ibm .com
Guardian of DBD::Informix v2005.01 -- http://dbi.perl.org/

[Neil Truby]

Re: Database market share 2004

<david@smooth1. co.uk> wrote in message
news:1117837690 .805470.259020@ z14g2000cwz.goo glegroups.com.. .[color=blue]
>
>
> Redhat say upgrade the firmware..I assume it fixes something...SO DO
> IT..
>
> If problem persists...get Dell to quote required OS patches and apply
> them.
>
> If Dell just quote a specific OS issue then quote that back to Redhat.
>
> If it's a known issue then Dell and/or Redhat should be able to quote
> required
> patches.
>
> If it is still a problem then Redhat should be able to put in more
> logging
> to give problem details. Dell should be able to help after all how do
> firmware
> patches get created?[/color]

I don't know if this irony, David. I suspect not. But either way, I think
illustrates well the actual point I was trying to make: that there remains a
widespread perceptiona mongst serious corporations that you're asking for
Support Hell if you buy an OS not supplied by he hardware manufacturer.

[Stu Charlton]

Re: Database market share 2004


DA Morgan wrote:
[color=blue]
> There is another important reason too: Instrumentation . If they are
> slow diagnosing why is a question of making guesses. That may be a
> reasonable approach when supporting a small non-commercial web site.
> It is a non-starter when talking terabytes and a requirement for 7x24.[/color]

A rather depressing thought is that the reason many of these open
source platforms are gaining popularity are a general and widespread
lack of understanding of the essentials of performance engineering.
That eyeballing utilization and ratios are enough.

Part of my job is to firefight important projects that use my
employer's technology (BEA). I find the "task forces" are formed with
the various vendors and don't actually wind up finding the problem --
they just look under their own rock, and give the thumbs up, and point
fingers to the other guy. It's hard to get a disciplined focus on the
end-to-end service and wait times.

Usually when I come in, I find the problem is "good J2EE 101" or "good
Oracle 101"; stuff that Tom Kyte has pointed out for years: Java
developers not using bind variables, developers building extremely
chatty systems (18 small round trips instead of 1 larger one), lack of
statement caches to avoid parsing, not collecting statistics regularly
and properly, etc. And management, desperate to show some kind of
action to their executives, just decide to add hardware. I can't blame
them, though they usually are aghast when I'm brought in and suggest
that this may even make the situation worse.

I look at the MySQL's and JBoss' of the world, and I see a focus on a
particular audience of developers whose itches are being scratched: the
shiny, stylish, and inexpensive environment that requires "just enough"
in-depth knowledge to keep the developer around to run the thing. But
when problems creep up outside their domain, all hell breaks loose.

Cheers
Stu

[DA Morgan]

Re: Database market share 2004

Neil Truby wrote:
[color=blue]
> I don't know if this irony, David. I suspect not. But either way, I think
> illustrates well the actual point I was trying to make: that there remains a
> widespread perceptiona mongst serious corporations that you're asking for
> Support Hell if you buy an OS not supplied by he hardware manufacturer.[/color]

Likely based on real-world experience.
--
Daniel A. Morgan

http://www.psoug.org

damorgan@x.wash ington.edu
(replace x with u to respond)

[DA Morgan]

Re: Database market share 2004

Stu Charlton wrote:[color=blue]
> DA Morgan wrote:
>
>[color=green]
>>There is another important reason too: Instrumentation . If they are
>>slow diagnosing why is a question of making guesses. That may be a
>>reasonable approach when supporting a small non-commercial web site.
>>It is a non-starter when talking terabytes and a requirement for 7x24.[/color]
>
>
> A rather depressing thought is that the reason many of these open
> source platforms are gaining popularity are a general and widespread
> lack of understanding of the essentials of performance engineering.
> That eyeballing utilization and ratios are enough.
>
> Part of my job is to firefight important projects that use my
> employer's technology (BEA). I find the "task forces" are formed with
> the various vendors and don't actually wind up finding the problem --
> they just look under their own rock, and give the thumbs up, and point
> fingers to the other guy. It's hard to get a disciplined focus on the
> end-to-end service and wait times.
>
> Usually when I come in, I find the problem is "good J2EE 101" or "good
> Oracle 101"; stuff that Tom Kyte has pointed out for years: Java
> developers not using bind variables, developers building extremely
> chatty systems (18 small round trips instead of 1 larger one), lack of
> statement caches to avoid parsing, not collecting statistics regularly
> and properly, etc. And management, desperate to show some kind of
> action to their executives, just decide to add hardware. I can't blame
> them, though they usually are aghast when I'm brought in and suggest
> that this may even make the situation worse.
>
> I look at the MySQL's and JBoss' of the world, and I see a focus on a
> particular audience of developers whose itches are being scratched: the
> shiny, stylish, and inexpensive environment that requires "just enough"
> in-depth knowledge to keep the developer around to run the thing. But
> when problems creep up outside their domain, all hell breaks loose.
>
> Cheers
> Stu[/color]

Precisely duplicates my experiences.
--
Daniel A. Morgan

http://www.psoug.org

damorgan@x.wash ington.edu
(replace x with u to respond)

[knorth]

Re: Database market share 2004

Stu Charlton wrote:[color=blue][color=green]
> > The number of downloads is probably a better measure of the influence
> > of open source products. Trying to use capitalization or revenue as a
> > yardstick is problematical.[/color]
>
> Agreed, but arguably "number of downloads" is moreso. I remember back
> during the dot-com boom, that was how ISVs measured their success.
> Didn't get them very far...[/color]

The first example that comes to mind is the Java Developer Kit. Because
it was a free download, you couldn't look to Sun financials for "JDK
sales" as an accurate measure of Java adoption. In the ramp-up period
for Java, Sun regularly publicized the number of JDK downloads as a
measure of Java's growing popularity.
[color=blue][color=green]
> > The economic effect of open source isn't limited to direct revenue from
> > licenses and support contracts. Companies that manufacture computers
> > know that open source contributes to the revenue stream from selling
> > hardware (e.g., 40 million web servers running Apache).[/color]
>
> Great, but since the only major integrated hardware & software
> companies are Sun & IBM, that's a rather limited set of companies.[/color]

Your integrated hardware & software company list is too narrow. For
example, HP's quarterly revenue from software was $277 million. That's
$205 million more than Borland's most recent quarterly revenue.

The more fundamental question is why you'd want to use an "integrated
hardware & software company" prism when looking at the effect of open
source software on hardware revenues.

Not every organization buying hardware to run Linux and Apache web
server will limit its vendor choices to only IBM and Sun because they
are integrated hardware and software companies.
[color=blue]
> Whither the ISV? What happens to Oracle, Symantec/Veritas, BEA, etc?[/color]

Open source clearly has a different impact on software companies than
on hardware vendors. If you're selling Xeon server boxes and RAID
farms, your revenue isn't likely to decline if there's a net gain of 6
million in the number of users of the Linux/Apache/JBoss combination.

On the other hand, if you're selling application server software, the
Linux/Apache/JBoss combo is a threat. Look at BEA's numbers for 2003
and 2004:

http://www.sqlsummit.com/Trends/AppServerMarket.htm

[Data Goob]

Re: Database market share 2004

Neil Truby wrote:[color=blue]
> "Jurgen Haan" <jurgen@fake.do m> wrote in message
> news:429f0643$0 $17153$e4fe514c @news.xs4all.nl ...
>[color=green]
>>Stu Charlton wrote:
>>[color=darkred]
>>>So the question is whether OSS eventually sucks the direct revenue out
>>>of the market. If there are enough vested interests to fund that sort
>>>of effort, sure, it could happen. Apache is a great example of that
>>>effect. I think the next area probably will be operating systems.
>>>Databases will take longer, if ever.
>>>[/color]
>>
>>I think databases will happen sooner than you think.
>>Hardware is getting cheaper while getting faster,
>>storage is getting cheaper while getting faster
>>OSS Databases are getting faster and more stable while still being free.
>>The one point missing in OSS databases is support.
>>It's so damn hard to get any support on OSS databases.
>>
>>That's why many companies still go for the mainstream databases,
>>for the false sense of safety (I pay for it so it must be good) and for
>>the support (which I still think is one of the most important issues).[/color]
>
>
> There is a very widespread perception that open-source operating systems -
> RedHat for example - do not provide adequate support either. We have
> encountered many risk-adverse users who would rather stick with Sun, HP or
> IBM because of the absence of demarcation disputes between the OS and
> hardware support vendors.
>
>[/color]
RedHat is not Linux. RedHat is RedHat-Linux, a fork in the road. But I
will say RedHat has provide me with excellent phone support, with a rather
short wait-time. RH lacks a lot of admin tools, and requires a bit more
hands-on than say SuSE, so you might find yourself using the support more
at first till you get comfortable with their limited, but usable tools.

RH does have some benefits in terms of how you can manage networking, but
it isn't necessarily better than SuSE, because you have to know what you're
doing. Not acceptable in new-to-Linux environments or for those risk-adverse
environments. Fedora is a great way to get into RedHat Linux because it
doesn't really cost anything. Of course you can get SuSE Pro for around
$99 USD so the question really becomes, how much will we want to support
it for ourselves. To the extreme, you can go Linux-From-Scratch and
really DIY ( www.linuxfromscratch.org ) but you will really learn Linux.

Use SuSE and get the support with it. SuSE has excellent admin tools
and allows your people to be successful with Linux and make it work
quickly without that queasy feeling that it isn't going to work. SuSE has
been excellent support on every single client I know that uses it. They
respond quickly and typically are spot on in problem solving. IBM has been
pushing "SuSE certified" systems for some time now, so you get a lot of
support not only from SuSE, but also from IBM.

To be sure you can expect problems with any release of Linux, but you should
have someone on staff that can help you as well as a support contract. The
nice thing about SuSE is that they really only have two products, one for
workstations that actually can be a server ( Pro ), and the other is a server
product that requires a support contract or you can't really update it unless
you really want to support it yourself ( SLES ). SLES has a lot of the
features you'd expect for a corporate server environment, deeper SCSI
support, etc. I think the latest release has morphed into a better
product with Novell networking built-in, which is even better, if you
are looking for that kind of product.

[Data Goob]

Re: Database market share 2004

david@smooth1.c o.uk wrote:[color=blue]
>
> Redhat say upgrade the firmware..I assume it fixes something...SO DO
> IT..
>
> If problem persists...get Dell to quote required OS patches and apply
> them.
>
> If Dell just quote a specific OS issue then quote that back to Redhat.
>
> If it's a known issue then Dell and/or Redhat should be able to quote
> required
> patches.
>
> If it is still a problem then Redhat should be able to put in more
> logging
> to give problem details. Dell should be able to help after all how do
> firmware
> patches get created?
>[/color]

We did not have this problem when using SuSE Linux--or Gentoo for that
matter on our Dell servers.

Dell has a mailing list you can subscribe to that will provide you
with input from a lot of heavy-hitters who use Linux on Dell, in
addition to using paid support. Much of the problems with Linux on
Dell have to do with RedHat, not necessarily Dell. It's like any
hardware platform, sometimes it's the hardware, or software or both.

Head over to Dell's Linux support website, and subscribe to their
list for a while and you can see what people are saying--and you
can ask them what they would recommend before actually using Dell
and Linux. Dell has been pretty good at improving their Linux
support, but they have been mostly RedHat because of their resistance
to support more than one Linux version--which points to a lack of
understanding or really digging into Linux because the bulk of their
sales is on Windows. When more shops use something else besides
RedHat they will get the message and start to support other releases.

[Christopher Browne]

Re: Database market share 2004

DA Morgan <damorgan@psoug .org> writes:[color=blue]
> Neil Truby wrote:[color=green]
>> "DA Morgan" <damorgan@psoug .org> wrote in message
>> news:1117485555 .324290@yasure. ..
>>[color=darkred]
>>>Larry wrote:
>>>
>>>>Daniel,
>>>>
>>>> What makes this year different from any other? It's always
>>>> backwards-looking, but that's the only way to capture actual
>>>> data.
>>>>
>>>
>>>Gartner also makes predictions. Not accurate ones but predictions.[/color][/color][/color]
[color=blue][color=green]
>> Don't know if it's the same in the US, but over here in Blighty the
>> supposedly independent analysts Gartner have attracted a lot of
>> criticism for advocating off-shoring in their editorials whilst at
>> the same time having an active consultancy arm that, er, helps
>> companies off-shore![/color]
>
> In my opinion, and it is only "my" opinion ... Gartner Group has the
> ethics of a great white shark in a feeding frenzy. Not sure about
> "here in Blighty" but they are most certainly a blight: If not a boil.[/color]

They have a history of having opinions in keeping with the sorts of
monies flowing towards them.

I remember starting to _really_ disbelieve them when their "Itanium
everywhere" opinions started coming out back in about 2000.

I daresay I haven't yet seen a single Itanium system deployed. I
gather that they exist, but in such tiny quantities as to be not of
commercial importance.
--
"cbbrowne","@", "ca.afilias.inf o"
<http://dev6.int.libert yrms.com/>
Christopher Browne
(416) 673-4124 (land)

[Christopher Browne]

Re: Database market share 2004

Paul <paulsnewsgroup s@hotmail.com> writes:[color=blue]
> That has to be market share in some shape or form, but was
> completely ignored in the article which only mentioned PostgreSQL,
> and even then only provided a link, with no mention of its market
> share![/color]

It is also an unmeaningful question in that the notion of "market
share" is difficult to formulate for a piece of software which, since
it does not have any organization that owns the right to licensing
fees, doesn't realistically factor in to have a well-defined notion of
either "market" or "share" of that market.

If it isn't "sold," then using the term "market" is misleading, at
best...
--
"cbbrowne","@", "ca.afilias.inf o"
<http://dev6.int.libert yrms.com/>
Christopher Browne
(416) 673-4124 (land)

[Chris Browne]

Re: Database market share 2004

Paul <paulsnewsgroup s@hotmail.com> writes:[color=blue]
> DA Morgan <damorgan@psoug .org> wrote:[color=green]
>>There is another important reason too: Instrumentation . If they are
>>slow diagnosing why is a question of making guesses. That may be a
>>reasonable approach when supporting a small non-commercial web site.
>>It is a non-starter when talking terabytes and a requirement for 7x24.
>>And then there's that little problem with government requirements
>>around audits. They just aren't ready for prime-time.[/color]
>
> I have worked for a company (big player in its field) and had an
> interview with another (a giant in its field) where they had systems
> that ran on all of the major db's - SQL Server, Oracle, DB2 and
> Sybase.
>
> These db's contain(ed) neither triggers, SPs, PL/SQL, Transact SQL
> or any other proprietary features of any of these db's. The only
> coding that varied between the db's was how to get an
> autoinc/generator/sequence value and that was in some sort of middle
> layer. They did make use of DRI in the db creation scripts.
>
> I'm not saying that that's a good thing (in fact I'm appalled), but
> the fact of the matter is that many companies pay for expensive db's
> when in fact they are little more than bitbuckets. I think that it's
> only a matter of time before the IT industry is going to wake up to
> the reality that (at least as far as *_I_* have seen) very little of
> the capacity of an Oracle or DB2 is actually being used and make the
> switch to cheaper or Open Source db's.[/color]

You may want Oracle's instrumentation for terabyte-sized DBs requiring
plenty-o-nines' uptime, but there are a whole lot of instances of
"department al" applications out there which are quite amenable to the
'leap' to free software databases.

For applications that are mostly used during office hours by the
limited population of staff within a department, you don't get a lot
of value out of the difference in licensing costs between Firebird or
PostgreSQL and Oracle.

And there are plenty of those sorts of applications around.
--
(format nil "~S@~S" "cbbrowne" "acm.org")

http://www.ntlug.org/~cbbrowne/sap.html

Rules of the Evil Overlord #78. "I will not tell my Legions of Terror
"And he must be taken alive!" The command will be: ``And try to take
him alive if it is reasonably practical.''"
<http://www.eviloverlor d.com/>

[Data Goob]

Re: Database market share 2004

You folks might want to get the online version of this
magazine:

http://www.eosj.com

They are talking about many of the myths-n-shit you guys are
propagating. This issue is one you won't want to miss.




Chris Browne wrote:[color=blue]
> Paul <paulsnewsgroup s@hotmail.com> writes:[color=green]
> > DA Morgan <damorgan@psoug .org> wrote:[color=darkred]
> >>There is another important reason too: Instrumentation . If they are
> >>slow diagnosing why is a question of making guesses. That may be a
> >>reasonable approach when supporting a small non-commercial web site.
> >>It is a non-starter when talking terabytes and a requirement for 7x24.
> >>And then there's that little problem with government requirements
> >>around audits. They just aren't ready for prime-time.[/color]
> >
> > I have worked for a company (big player in its field) and had an
> > interview with another (a giant in its field) where they had systems
> > that ran on all of the major db's - SQL Server, Oracle, DB2 and
> > Sybase.
> >
> > These db's contain(ed) neither triggers, SPs, PL/SQL, Transact SQL
> > or any other proprietary features of any of these db's. The only
> > coding that varied between the db's was how to get an
> > autoinc/generator/sequence value and that was in some sort of middle
> > layer. They did make use of DRI in the db creation scripts.
> >
> > I'm not saying that that's a good thing (in fact I'm appalled), but
> > the fact of the matter is that many companies pay for expensive db's
> > when in fact they are little more than bitbuckets. I think that it's
> > only a matter of time before the IT industry is going to wake up to
> > the reality that (at least as far as *_I_* have seen) very little of
> > the capacity of an Oracle or DB2 is actually being used and make the
> > switch to cheaper or Open Source db's.[/color]
>
> You may want Oracle's instrumentation for terabyte-sized DBs requiring
> plenty-o-nines' uptime, but there are a whole lot of instances of
> "department al" applications out there which are quite amenable to the
> 'leap' to free software databases.
>
> For applications that are mostly used during office hours by the
> limited population of staff within a department, you don't get a lot
> of value out of the difference in licensing costs between Firebird or
> PostgreSQL and Oracle.
>
> And there are plenty of those sorts of applications around.
> --
> (format nil "~S@~S" "cbbrowne" "acm.org")
> http://www.ntlug.org/~cbbrowne/sap.html
> Rules of the Evil Overlord #78. "I will not tell my Legions of Terror
> "And he must be taken alive!" The command will be: ``And try to take
> him alive if it is reasonably practical.''"
> <http://www.eviloverlor d.com/>[/color]

[Buck Nuggets]

Re: Database market share 2004

Jurgen Haan wrote:[color=blue]
> DA Morgan wrote:[color=green]
> > Jurgen Haan wrote:[/color]
>[color=green]
> > There is another important reason too: Instrumentation . If they are
> > slow diagnosing why is a question of making guesses. That may be a
> > reasonable approach when supporting a small non-commercial web site.
> > It is a non-starter when talking terabytes and a requirement for 7x24.
> > And then there's that little problem with government requirements
> > around audits. They just aren't ready for prime-time.[/color]
>
> 24x7 is actually no problem with OSS db's like Postgres.
> At the company where I work we have a DB2 DB and a Postgres DB running.
> Neither of them have to be taken down during maintenance.[/color]

Next week I get to swap out two storage arrays on a 500-gbyte db2 udb
database server: the new arrays are larger and faster. Anyhow, one
nice feature about db2 is that I can add the new arrays & remove the
old ones in a relatively simple operation in which the database handles
rebalancing all data automatically, and everything is online the entire
time. Didn't think postgresql was up to that yet.
[color=blue]
> What's a big problem with postgres (and actually one of the main reasons
> why we don't use it for our sensitive information) and that's that
> postgres is extremely unrelyable in high TPS situations.
> Scalability with OSS databases just plain sucks (if any).[/color]

Although I really like postgesql, the biggest reasons not to use it in
production for us are:

1. If it can't do everything then we need db2 as well. Standardizing
on db2 is simpler than maintaining a mix of skills, procedures,
products, etc.

2. postgresql doesn't support query parallelism or partitioning. This
means that db2 table scans run about 40x the speed of postgresql table
scans (assuming a 4-way server and typical partitioning benefits).

3. db2 is pretty inexpensive in its small-server editions (workgroup,
express, etc). So the savings of going with postgresql is more than
wiped out by the additional hardware requirements.
[color=blue]
> But still I think OSS databases are to be reconed with.
>
> It's the same as the early 90s, Linux, what a cute little project, but
> it surely will never be of any importance. Now, just take a look a the
> linux population among Internet web servers.[/color]

Yep, postgresql is on track to be a cool DBMS.

buck

[Michael Krzepkowski]

Re: Database market share 2004

Data Goob wrote:[color=blue]
> You folks might want to get the online version of this
> magazine:
>
> http://www.eosj.com
>
> They are talking about many of the myths-n-shit you guys are
> propagating. This issue is one you won't want to miss.[/color]
It uses a bizzare page reader and does not work very well with Firefox.

But they sure mention IE!!!!


[color=blue]
>
>
>
>
> Chris Browne wrote:
>[color=green]
>>Paul <paulsnewsgroup s@hotmail.com> writes:
>>[color=darkred]
>>>DA Morgan <damorgan@psoug .org> wrote:
>>>
>>>>There is another important reason too: Instrumentation . If they are
>>>>slow diagnosing why is a question of making guesses. That may be a
>>>>reasonabl e approach when supporting a small non-commercial web site.
>>>>It is a non-starter when talking terabytes and a requirement for 7x24.
>>>>And then there's that little problem with government requirements
>>>>around audits. They just aren't ready for prime-time.
>>>
>>>I have worked for a company (big player in its field) and had an
>>>interview with another (a giant in its field) where they had systems
>>>that ran on all of the major db's - SQL Server, Oracle, DB2 and
>>>Sybase.
>>>
>>>These db's contain(ed) neither triggers, SPs, PL/SQL, Transact SQL
>>>or any other proprietary features of any of these db's. The only
>>>coding that varied between the db's was how to get an
>>>autoinc/generator/sequence value and that was in some sort of middle
>>>layer. They did make use of DRI in the db creation scripts.
>>>
>>>I'm not saying that that's a good thing (in fact I'm appalled), but
>>>the fact of the matter is that many companies pay for expensive db's
>>>when in fact they are little more than bitbuckets. I think that it's
>>>only a matter of time before the IT industry is going to wake up to
>>>the reality that (at least as far as *_I_* have seen) very little of
>>>the capacity of an Oracle or DB2 is actually being used and make the
>>>switch to cheaper or Open Source db's.[/color]
>>
>>You may want Oracle's instrumentation for terabyte-sized DBs requiring
>>plenty-o-nines' uptime, but there are a whole lot of instances of
>>"departmental " applications out there which are quite amenable to the
>>'leap' to free software databases.
>>
>>For applications that are mostly used during office hours by the
>>limited population of staff within a department, you don't get a lot
>>of value out of the difference in licensing costs between Firebird or
>>PostgreSQL and Oracle.
>>
>>And there are plenty of those sorts of applications around.
>>--
>>(format nil "~S@~S" "cbbrowne" "acm.org")
>>http://www.ntlug.org/~cbbrowne/sap.html
>>Rules of the Evil Overlord #78. "I will not tell my Legions of Terror
>>"And he must be taken alive!" The command will be: ``And try to take
>>him alive if it is reasonably practical.''"
>><http://www.eviloverlor d.com/>[/color]
>
>[/color]

[DA Morgan]

Re: Database market share 2004

Data Goob wrote:
[color=blue][color=green]
>>For applications that are mostly used during office hours by the
>>limited population of staff within a department, you don't get a lot
>>of value out of the difference in licensing costs between Firebird or
>>PostgreSQL and Oracle.[/color][/color]

What planet do you reside on?

For those of us in the US.

If those apps involve any medical information you MUST comply with HIPAA.

If those apps involve any financial information including projections,
assets, liabilities, inventories, etc. you MUST comply with Sarbanes
Oxley.

If you have more than 0 employees you MUST comply with the regulations
published last Wednesday by the Federal Trade Commission.

Ignore any of the above you your attorney will be looking for a
retainer. Oh and that will be for criminal defense ... not civil
litigation if you attract the wrong kind of attention.

Do you think you can comply with those laws with Firebird?
I can answer that for you.
How abot PostgreSQL?
Same answer.
You'd better be thinking Oracle, DB2, or Informix or you had better not
be lecturing others about their bad habits and criticizing others that
break the law.

So yes ... if you are storing your mother's cookie recipies I would
think you'd be pretty safe in the open source databases to which you
refer.
--
Daniel A. Morgan

http://www.psoug.org

damorgan@x.wash ington.edu
(replace x with u to respond)