Red Hat Enters The Database Market

tekBuddha writes: "It looks like Red Hat is trying to make its way into the database market. This article on Yahoo! says Red Hat is about to announce its own product ingeniously named 'Red Hat Database' next Monday. "

Re:Uncreative title

[Anonymous Coward]

It is good you don't read Anon posts, cause this one will be from an Anon for reasons.

I have some particular inside information into some of this debate but wish to remain anonymous.

Redhat is developing this database mostly because when they recently looked around for an internal database to update their aging and plague ridden Oracle Database, Oracle themselves came again to pitch their product and flubbed it.

To say they flubbed it is probably mild, Oracle came to Redhat and decided to pitch only a Windows compatible version, refusing to even think about making a Linux variant for a Linux OS company to use. Its almost like AOL going to Microsoft and saying, hey we'd like you to scrape IE and use Netscape if you want our business. Oddly enough, many of the internal Marketing department actually loved the Oracle presentation. Luckily cooler heads have prevailed. Admittedly marketing got them back by taking a month to come up with the name for the new database product :) Anonymous Informant

Re:What code are they using

[Anonymous Coward]

And SAPDB - SAP recently opened that one up. It is a full-fledged database, more functional than interbase or any of the other free alternatives (but bigger too).

Now, it was pretty clear that SAP opened their database as a strike against Oracle -- Oracle is moving into the SAP marketplace and hardly anyone uses SAPDB without SAP, so SAP loses no revenue. SAPDB may not be as good as Oracle, but it is no slouch. It would make sense for RedHat to pick it up and run with it, especially if they got a little backing from SAP too.

Re:wtf are they thinking?

[abischof]

Yes, it's a meta [jargonfile.com] announcement, if you will ;).

Alex Bischoff

Re:What code are they using

[Micah]

The tables have actually somewhat turned in the MySQL/Postgres performance battle. [phpbuilder.com]
Check this out...

---

so who wants to bet it looks like this:

[Wakko Warner]

bash-2.05# find /usr/src/MySQL -exec sed s/MySQL/Red\ Hat\ Database/g {} \;

--
Forget Napster. Why not really break the law?

Re:PostgreSQL

[The Man]

And why not? You don't give any reasons why doing so would put anyone at a disadvantage.

sap-db?

[emil]

Is this the most advanced open-source database available now?

Does sap-db give you the ability to roll-forward a transaction log in a recovery situation? Can you do something like this with Postgres right now?

Re:Aargh!

[CaseyB]

I can't believe they listed MSSQL next to DB2 and Oracle!

You mean MySQL. Microsoft SQL Server wasn't mentioned, and it is an ACID database.

In fact IBM or DB/2 weren't mentioned either. What article did you read?

Re:sap-db?

[sql*kitten]

I understand it means a lot to have a comparable strength with an enterprise database...but talk about comparing old and obsoleted products....

Well, the brief was, find out what commercial products this could be used in place of. Oracle 7 is still a very capable product, and I wouldn't be at all surprised if many of the apps now being coded against Oracle 8.1.6 could be made to run on it with trivial changes (again, so long as you aren't using the exotic data types, yadda yadda).

Re:sap-db?

[sql*kitten]

Is this the most advanced open-source database available now?

I believe so, yes. I conducted an exhaustive study of this product as a consultant, and found it was equivalent to Oracle 7.3.2. If you're working with straight, transactional SQL in a relational schema (i.e. no OR features, none of the Oracle extensions for Time Series, Spatial Data et al) it's a fine choice, and decades ahead of MySQL.

Re:RedHat Database == PostgreSQL

[Jason Earl]

Actually, one of the problems with PostgreSQL (and I am a huge PostgreSQL fan) is that it really isn't possible to upgrade it in place like you can upgrade MySQL. The core developers have changed the on disk format of every single release that I am aware of (I have played with PostgreSQL since it was known as Postgres95). This means that an upgrade of PostgreSQL requires a dump to a very large text file and a reload. This process takes quite a bit of time, and is inherently dangerous.

There has been some talk on the PostgreSQL mailing lists about writing a tool that would read the various on-disk PostgreSQL formats and change them on the fly, thus allowing the user to upgrade in place. Instead of dumping your database to a text file, upgrading PostgreSQL, and reloading your database from the text file, you could simply shut down the postmaster, upgrade PostgreSQL, run the automatic fix-it tool, and restart your new postmaster. If RedHat has written such a tool for PostgreSQL that would be very cool (and well worth $2K a year IMHO).

In other words the trick isn't replacing rpm -i postgres-..., but rather the trick is replacing rpm -Uvh postgres-...

Re:RedHat Database == PostgreSQL

[Jason Earl]

Well, yes and no. First of all the different PostgreSQL backends are already quite adept at reading the binary formats, and the changes (from what I understand) have generally not been that large. Binary formats certainly are more fragile than human editable text, but clearly it is possible to manage binary formats fairly safely, otherwise any database would be a poor place to put your information. Second of all, since the postmaster would be off the tool could safely do a

tar zcvf backup.tgz data/

Before running the format munger. The tool would have to guarantee you had enough space before hand, but that isn't a big deal (and if you failed you could simply exit with an error message).

Besides, buman readable is only useful to a point. For example, the backups on my PostgreSQL database are big enough that I run them through split so I don't end up with a single file greater than 2G. If something happens to one of my backups recovering that data is going to be a pain (though it would probably be possible).

When upgrading I generally shut down my PostgreSQL server for an entire day as it takes a long time to upgrade (during which time the database is unavailable). When I upgrade I generally dump out the database, load the database into a test machine running the new version (while making a tarball of my data directory just in case), I then verify that all is well, upgrade my production machine's version of PostgreSQL, and then load the database into my production machine. This takes a long time. Updating the binary files would almost certainly be faster, and with proper precautions it wouldn't be any more dangerous.

Not that I am complaining. PostgreSQL's structured text file dumps (like most everything in PostgreSQL) work quite well. But it's not very fast or convenient.

My Guess

[Amphigory]

So I'm thinking about this... It seems to me that RedHat is probably buying one of the smaller commercial database companies and planning to make its product open source. Otherwise, I can't see why we wouldn't have heard about this -- Redhat has not typically done high security for new development projects. I just hope they are not planning to announce without code being imminent.

--

Re:wtf are they thinking?

[ksheff]

Maybe they are working with some of these other companies or projects, adding RH developed software along with it and marketing sales and support to their existing and new customers? How is that really any different from what they're doing with Linux? They take Open and Free Software, add to it, package it, sell it, and offer support contracts. A nice open source turn key DB system would be nice for small to medium businesses. Given that they haven't even announced it, the reporter may have misunderstood what they are doing and seems to be just going on rumors & speculations. Wait until they actually announce the product.

Re:Flawed benchmark

[WWWWolf]

Yesterday's Tux benchmark wasn't worthless (at least not if you serve static content).

::deep sigh::

Once more, in case someone missed: Tux does serve dynamic content. (People seem to be thinking of kttpd... khttpd, the thing that comes with the kernel, serves only static pages.)

Re:wtf are they thinking?

[Tet]

Redhat Database Management System = RDBMS

Yes, but the article implies that they're likely to call it the Redhat Database. That would lead to the equally confusing RDB [oracle.com] acronym...

Re:ODBMS?

[Tet]

The real values that a good rdbms provide are: Relations, relational constraints, SPEED

*Cough* A good RDBMS is many things, but fast isn't one of them. The reason the industry moved to relational databases (and SQL in particular) was flexibility. Non-relational databases like Berkeley DB [sleepycat.com] or BTrieve [pervasive.com] are significantly faster. The reason they're not more widely used is that the flexibility of SQL vastly outweighs the performance loss it incurs. It's a matter of using the right tool for the job. Netscape and Sendmail use Berkely DB because it makes sense to do so for what they need. Similarly, Remedy uses Oracle because it needs the flexibility more than the speed. Yes, a good RDBMS is faster than a poor one, but they're both slower than a good non-relational database.

I wonder.

[Petra]

1: Look at what Red hat did with Tux
2: look at the past few days of /. postings regarding Tux vs appache vs IIS ... specificly the comment about CGI being out of date in realtion to IIS 5.0's ISAPI.
3: Red Hat annouces A database.

Interesting eh?

Here's my $20.00, I'm betting Red Hat has found a way to tie a database system / databse server system into tux, and into the kerenel. Imagine the power of a kernel space http server, combined with a kernel space (or something simalar to tux) RDBMS. tied together with a API to rival ISAPI ... I wonder if the chili soft ASP! people have caught wind of this yet.... could really turn a few heads.

Re:Flawed benchmark

[stripes]

Once more, in case someone missed: Tux does serve dynamic content. (People seem to be thinking of kttpd... khttpd, the thing that comes with the kernel, serves only static pages.)

I didn't say it didn't. I said the benchmark didn't, I quote from the article:

The 60.7MB of static Web content was small enough to easily fit into RAM,

That said, no, I didn't know that Tux can't do dynamic content, but I'm also not foolish enough to make sweeping statements about things I don't know about.

How does Tux do dynamic content? Does it pass it up to a userland web server on the same box? On another box?

Re:Flawed benchmark

[stripes]

A benchmark is inherently flawed if it isn't made by an impartial entity.

No, it is inherently suspect. If it includes enough information to reproduce the results, and they are reproducible odds that it is useful go up. If they lose odds go up more. If it is simple enough that you can see if it merely played to the strengths of that software or if it is a real test of how it might be used...well that helps too.

After all SPEC benchmarks are run by the venders (on programs and data sets that a group of venders all agreed on).

Yesterday's Tux benchmark wasn't worthless (at least not if you serve static content).

Most benchmarks are not done by impartial entities. Many are run by magazines that get advertising money from only some of the products. Some are run by the potential users that have a vested interest in picking the cheapest one.

SuSE Email Server

[IGnatius T Foobar]

The same thing happened with the product called "SuSE Email Server." It turned out to just be Postfix, Cyrus, and some other things pre-integrated and shipped in a nice-looking box bundled with a support agreement. Considering the magnitude of writing a database, I'd expect Red Hat's offering to be something similar. They didn't survive the Linux company shakeout and turn a profit by being stupid.
--

Re:PostgreSQL

[PCM2]

Actually, Greatbridge DOES sell PostgreSQL:
http://www.greatbridge.com/product/software.php [greatbridge.com]
--

Re:I wonder.

[johnnyb]

However, one nice thing about Tux is that it sparked the movement of most of its code to more generic parts of the kernel, thus allowing the creation of the X15? web server which is entirely userland, but even faster than Tux. The way this works is that a lot of the code from Tux was re-written more generically to make well-written 2.4 apps really scream. Thinks like zero-copy networking and the like. Anyway, it is possible that they did a similar thing with the database - used Tux's zero-copy stuff to make PostgreSQL excruciatingly fast.

However, my guess is simply that they are just packaging Postgres in a more user-friendly way, not actually changing it too much (although they might). Then again, they might not even be using Postgres. It looks like we're just going to have to wait and see :)

Re:I wonder.

[johnnyb]

Of course if I were RedHat, before I put any work into an idea, I would leak it to the press and look and see what ideas Slashdot had about what I was thinking about doing, and then decide (1) if it's worthwhile and (2) what the best plan of attack is. Then, on Monday, I would just announce that I'm developing whatever the best idea on Slashdot was.

Premature?

[doomicon]

This seems to be a little premature, as well as
a tough market to enter. With the number of
Enterprise databases available, as well as the
free databases, it's gonna be a little crowded
out there.

Also, who would be the target customer? Those
who have the cash to spend on Oracle, are not
going to take a step down. And those with mySQL
and Post, I can't see them paying for an unproven
product... So that leave's MS and SQL2K, but I
can't imagine companies already entrenched in the
MS platform changing.

To establish a profitable customer base in such a
tough market is going to take ALOT of money, and
time. It just doesn't seem that RH is focusing
on the correct market.

Anyone, have another angle on this?

Online backup

[XNormal]

According to the above mentioned Yahoo article [yahoo.com] RedHat's offering will also include online backup and recovery. For small and medium sized systems this could be a really big advantage. Hopefully it will support encryption...


-

Good luck!

[alsta]

I am not very savvy with what the differences between various OpenSource databases are, but I have a couple of thoughts.

Oracle is the database of choice when entering the big league. A few people are going to say that SQL Server is just as good or better, but what one usually sees is Oracle on Solaris. Oracle and Sun are in bed with eachother. If it weren't for one or the other, they would probably have had a much harder time. Sure, VERITAS is in there and gets sloppy seconds too. But the strategic corporate harmony is with Oracle and Sun. Oracle people like sun, because Oracle likes Sun. Sun people like Oracle, because Sun touts Oracle as being the best.

So now that Oracle has let Red Hat in the door to "watch", why would they jump out of this and slam the door after themselves. Why should Oracle maintain their RDBMS for Red Hat, now that Red Hat can't even stay away from offering an RDBMS? Sun and Oracle seem to have a secret code with eachother:

Sun: I swear that no Database product shall wear my name...

Oracle: I swear that no Operating System/Hardware produced by me shall compete with Sun...

Bottom line is, that Sun and Oracle already have a good reputation out there among pointy-haired bosses. Red Hat had lots of fun during the .com days and even though they are profitable, I am sure that the reputation up along the corporate ladder isn't the same.

How many Red Hat Database licenses can Red Hat sell? How long does the product need to live before it is viable with everybody in the top layer so that sales start taking off? We've got NuSphere MySQL and at least a few companies supporting PostgreSQL if that's something that is needed. Why buy Red Hat's?

Don't get me wrong, I think it's cool that Red Hat has balls... But will they succeed with this one?

wtf are they thinking?

[banky]

I don't understand this at all.

OK, so, you can't use Oracle, because its proprietary. Fine, use MySQL. No, wait, that's largely banked by VA (IIRC). Also rule out Postgres, and Interbase for similar reasons.

So *develop your own*? That just makes no sense. "NIH" (Not Invented Here) kills companys, it blinds their thinking from the best tool for the job.

Why can't Red Hat just partner with NuSphere/Greatbridge/etc? Is it NIH, or something about business I just don't understand?

Last point: This has (apparently) appeared out of nowhere. Some press, but that's it. No beta program. No white papers. No conspicuous hiring of RDBMS gurus - hell, check the Borders in Durham, I doubt they've even bought a textbook on databases. What gives here? Are they going to launch a product without first having running code?

It'll likely be open-source anyway...

[tuffy]

...since everything else Red Hat does is. That way, the people on all the other Unices can test it out and submit patches to ensure portability across platforms.

It'd be nice if Red Hat would come up with a better name, tho.

Re:The Databse-market?

[woggo]

actually, the US State Department has done a pretty good job of collecting data on where the largest BSE (bovine spongiform encephalopathy, or mad cow disease) infestations are, thereby letting me know where not to eat beef.

Therefore, http://travel.state.gov [state.gov] is my preferred "data-BSE", although CNN is fine for less enterprise-class data requirements. For "data-BSE" metadata, I'd suggest you consult the USDA [usda.gov].

What kind of database?

[Mr. Flibble]

From the article: "The pricing has really come down" on database products, he said. "It kind of closes the door on the open-source guys."

Funny, I thought Red Hat, being a Linux Company (and now posting a profit!) understood what the GPL is and what Open Source is. I can't see them "Closing the door" as that quote states...

The real question is what kind of database? My guess is that it will be some form of SQL, but a little more advanced than MySQL in that it will be a relational database (like Pervasive SQL).

What would be a smart idea, is to have the Red Hat Database come out as a CROSS PLATFORM standard. Yes, I said cross platform. The database that we use here at work (Pervasive SQL) runs on Novell, Windows, Linux, and Solaris. When we need to provide a database solution for our customers, we can put the database on whatever type of server they prefer (more and more of our customers are choosing a Linux server however, due to cost reasons.)

Having the database run on more than one platform gives our customers a choice, our customers like that. I sadly doubt that Red Hat Database will work on anything other than Linux, but then I cannot blame Red Hat for doing that. It only makes sense for it to run on Red Hat. (AFAIK, Access only runs on Windows.)

I expect to see this database come out open (although maybe not GPL?), so it stands a chance of getting ported. It would be great to see Red Hat Database out competing Access or some other Microsoft Database on their home turf (NT).

Probably won't happen, but its nice to dream anyways.

Re:ODBMS?

[Panaflex]

libxml rocks. I've been writing quite a few systems using it, and it is superb.

One thing... use libxml2!!!

Pan

Re:Winning move or fatal blunder?

[Panaflex]

SAP has open sourced the DB. I'm betting all my redhat stock (ehehe) that their using SAP DB.

Look at the timeing... SAP just released their DB a few weeks ago under the GPL.

Pan

Re:Uncreative title

[ethereal]

Then you say that RHAT has a marketing department that works against the engineers out of spite? Sounds like a sick company to me, in the black or not.

Really? Sounds like a typical technical company to me :)

Caution: contents may be quarrelsome and meticulous!

Re:wtf are they thinking?

[Skeezix]

There is no indication yet that Red Hat developed a new database from scratch. Their product may very well an optimized, packaged version of an existing open source database with configuration and interoperability tools.

Hint, Hint, Wish, Wish!

[HiThere]

Maybe they'll include a dialog creator (ala Glade), a scripting language (perhaps Python), and a report writer. Then we'd have some real competition for MS Access.

Caution: Now approaching the (technological) singularity.

And the don't waste money either...

[sterno]

So they take an existing open source database, repackage it, and retool their support model a bit and suddenly they have a brand new revenue stream with very little effort put in. And people wonder how open source companies make money :).

I'm betting my money this is a repackaged version of PostgreSQL. Does anybody have details?

---

All your database are belong to us

[Gaewyn L Knight]

sorry... couldn't resist....

But seriously, if they can provide a product that comes close in performance I see a lot more companies going over to it. I know a lot of companies that would love to use open source but they want someone to blame when it goes wrong.

Hopefully Redhat can still turn a profit when they are the ones catching the blame.

chalk another one up for postgresql

[kneel]

My company recently moved from an all Sun/Solaris/Sybase shop to Dell/Linux/Postgres. I spent alot of time with RedHat engineers and Sales people and my name somehow got around the company, and ended up in the lap of someone who was working in their Product Development department. He gave me a call and interviewed me on my experiences with postgresql, asking me what kind of applications we were using, why we chose postgres, and where we were getting our support from. He hinted that they might be producing a postgresql product.

If RedHat is in fact using postgres, kudos to them for picking a RDBMS with row-level locking, failover, and several other enterprise level database product features. And kudos to the PGSQL team for creating a wonderful product. Our websites run much better on our new platform than our old.

Article short on details

[evilpenguin]

The article is way to non-specific to judge what RedHat is doing. If they took one of the solid Open Source databases (I would say PostgreSQL because it presently has features needed for transactional "enterprise class" multi-user databases that MySQL lacks, at least in the last version I used -- I know the developers were talking about adding these features -- such as multi-op transactions, record versioning (for non-blocking transactions) and so on) and, in essence, built a "distribution" that included all the tools folks would want around it (ODBC/JDBC stuff, admin tools, GUI tools a la Oracle Designer), and so long as they continue their habit of making the tools they roll themselves GPL'ed, I say, "What a great idea!" BTW, I'm not saying what the world needs is a Designer-like tool for PostgreSQL, but there are "softies" in the DBA world, just like there are "MSofties" in the programming and sysadmin world. Why not make a buck by giving them what they want? Heck, that's what RedHat's success is based on! They didn't write Linux (okay, so Alan Cox has been working for them for a long time, so one might say they did a lot of good; they also did the abomination that is RPM [flame on!]), but they have built a business out of selling it.

A really good packaging of PostgreSQL with lots of nifty doodads like those the "big boys" sell with their DB engines would be a good product to have.

To me, the whole Linux thing is about freedom. I can always roll my own if I want to or have to, but 95% of the time, I'd rather re-use someone else rolled-up DB environment. So long as they stick with their GPL tradition, I think this is a great idea.

Mind you, this is a vast pile of speculation based on a mere atom of fact. Not that I've ever let that stop me before...

Re:Article short on details

[evilpenguin]

Continuing my tradition of replying to my own posts, I will add that if they make their "database distro" or whatever it is REQUIRE RedHat Linux, then they are a) retarded, b) evil, and 3) going to fail.

Red Hat "Linux"

[cpeterso]

What would the impact be, compared to the fact that Oracle would soon be running on Linux and also Informix (acquired by IBM) and DB2 would soon be running on Linux clusters ?

Oracle, Informix, and DB2 might running on Linux, but perhaps they won't be compatible with Red Hat 8? What prevents Red Hat from introducing incompatibilities into their operating system? The gcc 2.96 controversy is just the beginning of the proprietarization of Red Hat "Linux".

Re:RedHat Database == PostgreSQL

[cpeterso]

Don't you think that munging the binary database file (in one of many possible format versions) is much more dangerous than just exporting/importing a structured text file?

Re:How will it stack up, though.

[RyanGWU82]

Why not do this?

SELECT DISTINCT c.name
FROM customers c, collections cl
WHERE c.id = cl.id
AND cl.debt > 0

This should work in any SQL server even if it doesn't support subselects. Have you found the subselect query to be faster/better? Or were you just trying to use an example subselect, not necessarily designing it to be "correct"?

Re:Would it make an impact ?

[DenialS]

IBM DB2 on Linux [ibm.com] already runs on Intel clusters, and has for about 6 months (see DB2 UDB Enterprise Extended Edition [ibm.com]). In fact, it also runs on Linux for the S/390 platform and Linux for the Itanium platform. I don't think that Oracle claims to be playing in either of those areas yet...

I'm looking forward to reading more details about Red Hat's database product. This article was just a teaser.

Re:What code are they using

[Misha]

If they are trying to steal market from Oracle (small businesses and such), they will need their database to be transactional. Thus it's not MySQL in its current incarnation. And from what I remember in DB course, not accounting for transactions in the early stages of design, makes a hell of a time rewriting the product later.

Postgres, while transactional, lacks the performance. I have not tested it myself, but I seem to remember a bunch of benchmark articles a while back which seemed to point to this. (CORRECT ME, DON'T FLAME ME, IF I AM WRONG). It kinda makes sense for MySQL to be faster if it doesn't support LOCK.

Works for me.

[cybrthng]

Windows has worked for me for the past 5 years no problem. I wonder what planet your from.. did you buy your pc from radio shack or the dude selling from the back of a truck?

Linux is just another choice, and both kick ass and crash or have weird quirks. I have yet to be able to use KOffice or even mature star office for more then an hour without barfing, i think i can leave office 95 running on my laptop for months without a single problem.

For all intents and purposes, microsoft's products are a whole in one. Without microsoft the PC would have never evolved. It would have been left to IBM or unix guys and then everything would have been screwed up.

As for innovation bs, thats for the birds. Just make your products and sell your shit. I'm not buying innovation, i'm putting my money into something that gets the job done or gives me gratification.

Use Oracle 9i for Linux.

[cybrthng]

Oracle 9i is Glibc 2.2 based, works on all the newer distros (althought it could appear GCC 3.0 and the new libraries are sure to break that very soon).

Oracle 9i supports too many features to list, and certainly can't be beat by anything out there yet.

XML, ODBC, JDBC, J2SEE, PL/SQL, SQL99, Flat File support (Yes, you can link flat files to the database and query them via sql now), Clustering, Apache + ModPLSQL + XML + Caching + Portal and tons of other features.

Not to mention performance increases, the huge and highly successull structures for maintaining and reporting system status. The new updated rollback and logging system as well as more security and labeling features.

It runs on RedHat and many other unix varients as well as other Unix operating system

RedHat is shooting themselves in the foot if it is only a linux product. Even IBM knows DB2 is needed on NT, and other platforms. Sure linux runs on other hardware, but hardly a business solution for enterprise environents.

Especially the fact no conversion documents, white papers or applications using the database exist. Won't cause anyone to change. You can already order 9i books and run it on Linux, Solaris and HPUX with NT version coming out soon.

Re:ODBMS?

[Webmonger]

While the text format of XML is what most people think of when they talk about XML, this isn't what people mean when they talk about XML databases.

XML is also a way of structuring data; trees, parents, children, attributes, etc. When people talk about XML databases, they mean a way of storing the XML data, without storing it in the XML text format.

Presumably, this sort of approach would be much faster than using the text format (and far fewer files!) but would also be much easier when to use with XML data than relational databases are. It would also allow the database to prevent races.

Re:Portability

[Darren.Moffat]

Solaris 8 shipped with rpm2cpio ;-)

Re:What kind of database?

[GauteL]

It is not the custom of the OSS-community to bind software to one platform, and AFAIK Red Hat is still pretty much in touch with the community, despite what some trolls seem to suggest at times.

The Databse-market?

[GauteL]

That would be a fantastic move... as far as I know there are no other players in the databse-market, so Red Hat should be able reak in heaps of profit!

I would also look into the databae-market, which is also unexploited.

:-)

Re:Who is their target audience?

[helarno]

The article states that Red Hat is specifically going after small to medium sized business, who don't see Oracle as a reasonable expense.

Oracle is expensive and charged on a per CPU (actually per Mhz) basis. So if you're building something that is expected to scale, you're looking at a lot of money.

For example, if you're building a cluster of web servers on stock x86 boxes, you'll find that the price of a single Oracle license on a single box will probably cost you in excess of five of those plain white boxes. For a small company, that is unacceptable.

Aargh!

[Grendel Drago]

I can't believe they listed MSSQL next to DB2 and Oracle! Sheesh!

Yes, MSSQL runs on expensive hardware. Oracle and DB2 run on *really* expensive hardware. And pass the ACID test. And so on.

Guh...

-grendel drago

Portability

[SecretAsianMan]

Anybody think that they'll actually try to make it protable across all Unices?

--
SecretAsianMan (54.5% Slashdot pure)

PostgreSQL

[radulovich]

According to this article [fortune.com] at fortune, Frank Batten, lead investor/owner of Greatbrigde, is also the largest individual shareholder in RedHat.

Also note that Greatbridge does NOT SELL PostgreSQL - it sells services. I would bet money that it is PostgreSQL that RedHat produces, and contracts out tech support to Greatbridge.

Synergies exist, there is no direct competition, and the owner is an old friend.

I think the writing is on the wall.
-Mark

No. It's not NIH.

[robl]

NIH syndrome is what happened when IBM didn't want to use something like C or AT&T Unix and instead chose to use their own in house proprietary tools / OS's instead.

This, my friend, is called competition. Yes, they're going up against the Oracle goliath, but it's competition none the less.

Fine, use MySQL. No, wait, that's largely banked by VA (IIRC)

Okay, then by that logic, Mandrake should never have based their own distribution off the Red Hat version, opting to instead, start completely from scratch. Luckily for Mandrake, they didn't follow your logic.

And maybe RedHat isn't gunning for press coverage. Tney aren't spending marketing money near to the extent that IBM, Oracle, or M$ are. Continuing in the speculation mode, maybe they just want to produce a "Economy Class" database to compete with the "Rolls Royce" Oracle.

Re:What code are they using

[Zigg]

If they are starting from someone else's code base, I'm betting it's PostgreSQL. It seems to me they've always had a soft spot for it.

Re:What code are they using

[Ronin Developer]

I've used Interbase since 1996 and like it. If you follow the uninstall procedures (i.e run the uninstall program) as specified by Borland, then it uninstalls just fine and will not corrupt other database drivers. It's when people try to manually delete files that that they get into trouble. If you were using the BDE, manually uninstalling drivers is a bad thing in general and is not recommended.

Additionally, InterBase is not orphaned and is very much supported by Borland. I suggest you go to their website and see what they are offering. Last I checked, IB was running on a myriad of platforms and is included on the Kylix CD. I assume its also on the Delphi 6 CD...but as I don't have it...it's just an assumption.

If you are looking for a free alternative, I suggest you consider the FireBird project. It's essentially InterBase but based on the Open Sourced code. It runs under Linux and Windows.

Best tool to use for accessing InterBase or FireBird is Jason Wharton's InterBase Objects.

Finally, if you are running Windows, use a utility like CleanSweep to record your installation process. This will make a clean uninstall a piece of cake. One thing about Windows I really don't like is that there are no reference counts to files, libraries, or drivers. But, then again, I don't know of too many install programs or OS's that actually track this information. Oh well.

Re:wtf are they thinking?

[Nailer]

My theory is, they're banking on the confusion of PHBs who read about "the latest RDBMS technology" and notice that the Redhat RDBMS is a whole bunch cheaper than Oracle.

Makes sense. These are the same PHBs who get confused into thinking any SQL server is MS `SQL Server'.

GO RH!

[ndfa]

Forbes has this to say about Red Hat [forbes.com]..
NEW YORK - Red Hat Software has squeezed out a first-quarter profit distributing a product that, for all intents and purposes, it doesn't own.

So its a lot like Microsoft which has made profits distributing a product that, for all intents and purposes, doesn't work.

Re:Works for me.

[ndfa]

It was a freaking joke... maybe not in the best of tastes...but jeez....also i am using windows 2k right now and cant complain... running within VmWare it has not crashed for a very long time.. I also use win2K at home to play games etc...

and thanks so much in the history lesson about PC's. and here i was thinking that they just won the OS race early leaving OS2 behind.. you almost make M$ role in the evolutons of PC's sound noble... hey i dont mind their OS, i think they do their limited tasks well. But lets not forget that this is also the firm that is responsible for making a joke out of standards. Do you remember JAVA.. hell just look at HTML and what
MS did with that.... they are a company that enjoy's its position in the industry and uses it to try and move ahead everywhere else!

Re:Would it make an impact ?

[scoove]

How does it work if a decision in a big company for a _new_ DB system (usually for a special purpose project) has to be made?

I'll tell you how I've seen it done in telecom. You sit around in one meeting after another, gathering a fantasy list of user requirements, only to discover that the sponsor blew their budget and expected the IT department to cover the costs.

Then, when you get to the sticker shock, you get questions like:

"What's the $250K for? Licenses? For what? Database software? Can't you just write something? After all, what's all that development time for if you're just going to go out and buy stuff from Oracle? Why are you paying all those programmers and database people if you're just going to buy it anyways?"

By that time, open source is looking to be the only option that'll come close to the user's budget. True, you've got the developers, and probably the time (wait in line), but without the funds for the big budget commercial database, open source sure beats writing from scratch.

Plus, I've yet to meet a commercial database that came pre-constructed to do exactly what I need to do for telecom billing, so it's assumed that you'll be conducting a good amount of development on the system regardless of its origin.

But please folks, stop dreaming that RH will eat away the piece of the big guys in the upperscale solution market.

No. Microsoft already has done that. In too many vertical markets, Oracle/IBM are not adding any value. At least Microsoft doesn't charge you nearly as badly based on use. (Vendors: Take a hint. Charge me on use and you'll guarantee I'll be looking for a way to get rid of you as I grow!)

Per the previous example, I've tired of Oracle quotes that demand:

- extra per processor
- extra if you're going to use the data on the Internet somehow (even if it is interfacing a Microsoft SQL server first, which is talking to the web via IIS and Oracle never sees a public IP)
- a huge chunk of change for maintenance (aka bug fixes)

Then again, it's not easy to make Larry Ellison wealthier than Gates...

You're also right that PostgresQL and other open source varients can't touch Oracle's very high end capabilities. Their stuff is "trivial" [amazon.com] (thanks Clayton Christensen for the model), and is the only thing the very, very high end customers (5%) of Oracle can use. The other 95% do just fine with the open source alternatives.

*scoove*

Re:wtf are they thinking?

[rkent]

wtf are they thinking?

Redhat Database Management System = RDBMS

Relational Database Management System = RDBMS

My theory is, they're banking on the confusion of PHBs who read about "the latest RDBMS technology" and notice that the Redhat RDBMS is a whole bunch cheaper than Oracle.

D'oh!

---

Re:I wonder.

[selectspec]

Not too sure about that. Tux has been around for awhile. Webservers like Tux don't really do all that much in the kernel. Most of the complicated functionality such as auth, dynamic content, etc live in user space with Tux. All that is in the kernel is the file serving part and it works faster because the memory copying is reduced (since the memory is all in kspace) and the files served by webservers tend to be large.

Putting a database into the kernel is an entirely different proposition. Unlike with a webserver, the bottleneck with a database is not memory copying or even I/O. The bottleneck is with bandwith in the memory bus do to the large size of tables and with the high latency of the I/O bus to the drives. Plus, you wouldn't be able to process SQL queries and do index lookups any faster in the kernel than you would in user space. Databases already can talk to raw partitions and bypass the filesystems within the kernel. Generally the data is stored and retrieved from databases in small chunks, so the memory penalty of having the database in user space is really minor.

The only bennifit I can see for putting the database in the kernel is to assist in clustering of databases. Given the intensive peer-to-peer communications of a cluster and the need for low latency I/O, putting as much as you can closer to the hardware is always a plus.

Re:Would it make an impact ?

[AugstWest]

oracle already runs on linux. i've run 8.1.5, 8.1.6 and 8.1.7 on redhat 6.2 without any problems. i have a development 8.1.6 server that has been up and getting hammered for almost a year now without a reboot or a restart or oracle.

Familiar Strategy.

[mikej]

How often have you heard this story:

Headline: $company Announces Move Into $field.

$company announced today that they will be moving into $field software, with product details to be announced in roughly six months. At the news, developer interest in $field products from companies already in the market have dried up, as $company's version will likely become a major, if not the primary, industry standard.

Then, in the intervening months, all competition in the market goes belly-up, $company buys whatever former competitor they think is most useful, and becomes the Only Game In Town.

Set $company to "Microsoft" and everyone knows this story. It hasn't worked if set to "RedHat" until now.

Re:wtf are they thinking?

[Pennywise]

So I guess this is an announcement of the announcement? :)

Who says they're making something _new_?

[BierGuzzl]

The article has no details about what type of rollout this is expected to be. Given the amount of work that goes into creating a new database, there would have been some noise about it by now. Likely, what we're looking at is yet another Mysql/postgres in a new box, with a nifty configuration/control panel and a manual.

Re:Article short on details

[teg]

MySQL does support transactions for some of the table types it supports, but it lacks other essentials like foreign keys and subselects. More details can be found in their docs [mysql.com]

Re:Makes Sense

[teg]

MySQL is included with Red Hat Linux 7 and 7.1.

Re:A pattern forms...

[festers]

Maybe not every small/medium business can afford Oracle, etc.? Ever stop to consider that possibility from your oh-so-high-and-mighty position? That seems like a pretty reasonable market to go after: Companies that can't justify the expense of Oracle but need something more than a RDBMS hack.


--------

Oracle vs RedHat

[pravel]

Previous Oracle release, 8i, was certified on RedHat 6.2, and installed cleanly out of the box. The latest 9i release, released last week, is certified on SuSe, and does not install on unmodified RedHat Linux.

It seems that Oracle is moving away from RedHat.

Re:ODBMS?

[StandardDeviant]

Well, they do ship PostgreSQL, which is at least an ORDBMS. (And it's really cool!</advocacy>). Plus even the latest versions of Pg typically are a zero sweat install (rpms from the Pg site).

--
News for geeks in Austin: www.geekaustin.org [geekaustin.org]

Re:ODBMS?

[nojomofo]

If you think of a Relational Database as "tables and rows", then perhaps you should use XML instead. The real values that a good rdbms provide are: Relations, relational constraints, SPEED (which you seem to have neglected entirely in your thoughts) and the like. Having data doesn't do you much good if you can't get it out fast, and get it out in a useful fashion (joins).

If you really neead a "reason to use a database at all", then you clearly aren't doing anything interesting with any volume of data.

Re:Would it make an impact ?

[jallen02]

Oracle did not instantaneously become a enterprise scale RDBMS overnight either

I mean if they continue to develop the product and add features who knows.

No not a treat yet but after a few years of sucess in the midsized world they could have a mature enough operating system and product to really make a break through. Think big big picture, not just, the next two months ;)

Jeremy

Re:What code are they using

[Tassach]

As a Sybase DBA, this would make me VERY happy. I've administered various Sybase, Oracle, and MS-SQL installations over the years. Of them all, I've found that Sybase requires the least care and feeding to stay happy. The only drawback I've seen is that T-SQL lacks some of the neato (propriatary) features of PL-SQL, but that's an acceptable trade-off for me. The Sybase optimizer has some interesting quirks that can be frustrating at times, but they can usually be worked around without too much effort.

One of the BEST features about Sybase on Linux is that it's trivial to move an existing MS-SQL database off of NT -- this can often be done without changing any code. This makes it a very attractive migration path for those looking to decrease their reliance on M$.

How will it stack up, though.

[don_carnage]

There are quite a few very well established databases out there already: MySQL [mysql.com], Oracle [oracle.com], Postgres [crimelabs.net] to name a few. Just so long as I can do this, I'll be happy:

• select c.name
• 
  from customer c
  where c.id in (select cl.id
  from collections cl
  where cl.debt > 0)
  

--

Re:How will it stack up, though.

[don_carnage]

Sorry...just an example subselect. I get very irritated when I can't do that sort of thing in MySQL...especially with updates:

• update customers c
• 
  set c.salary = (select newsalary from othertable o where o.id = c.id)

--

What code are they using

[ibirman]

The article does not mention where they are getting the code. It is probably too big a job to start from scratch - so their only real choice is probably Postgres. Are there any other open source databases they can start with besides MySQL?

Just what we need...

[duffbeer703]

Yet another database engine...

RedHat will probaly be the only one to use it, I bet they originally wanted a more powerful DB for rpm.

Would it make an impact ?

[cOdEgUru]

What would the impact be, compared to the fact that Oracle would soon be running on Linux and also Informix (acquired by IBM) and DB2 would soon be running on Linux clusters ?

Do you think Redhat could have some clout in convincing the market that it can build a solid database compared to the solutions already out there. Agreed, that Linux is a solid server OS, but that itself wouldnt do. RedHat needs to have a solid solution which doesnt depend on Linux, but be able to stand on its own, in terms of TPS and Performance issues.

The ACID test

[CritterNYC]

Actually, I think most databases perform ok while their admins are on acid.

R & D Spending increase

[1alpha7]

Well, the article said they had increased their R&D spending. I just didn't think that meant something as mundane as a database. Oh, well, maybe they're hiding the exotic stufff behind this cover . . .

1Alpha7

Re:What code are they using

[11223]

Don't forget InterBase, from Borland...

Re:Would it make an impact ?

[iseletsk]

I don't think they are going after enterprise market. But the whole idea that they provide an enterprise support, and they are a big company, will be a major plus for any database they would promote. Latest version of Oracle costs about $40k per CPU. DB2 & SQL Server are not much cheaper. If you want a disent "comercial grade" database, for a reasonable price, with an option to get support "latter on", there is almost no way to go.
There are some good opensource databases. Yet they all lack clustering (even that MySQL has some rudimentary form of it). If RedHad would come up with some solution that has clustering (even for an extra $), and if RedHad would throw its name behind the product - there is a posiblity of success. No, they will not get to the enterprise market, but they might get bunch of smaller accounts that were going to MS SQL before that.

Re:ODBMS?

[graveyhead]

Don't use DOM, use SAX instead. It's much more painful, but also at least 10x faster, without the memory requirements of DOM. Also, which parser are you using? I have found that Apache Xerces is a good API but slow, and that libxml is tougher to use, but faster.

Well, your fingers weave quick minarets; Speak in secret alphabets;

Re:ODBMS?

[graveyhead]

OK I see your point on speed. I generally work with smaller scale applications that don't require thousands of connections per second, and don't have terrabytes of data.

As for "not doing anything interesting", I have to disagree. Have you looked at xslt? Using it in combination with some programming logic, you can do some seriously interesting things. For example, a job site I am currently working on uses many separate xml files with different purposes. Some logic comes along and stiches the files together, and then processes the whole lot using xslt. The result is similar to RDB JOIN, with the notable difference that I have ultra fine control over every input and output element. Systems like this allow me to easily generate client-specific HTML output, and special case output such as simplified xml for syndication.

Of course when speed isn't a critical issue, a database or xml will work for a system like this. In the end, it comes down to a matter of personal preference.

Well, your fingers weave quick minarets; Speak in secret alphabets;

Re:ODBMS? - as in crappy performance

[graveyhead]

I guess you don't have much experience working with ODBMS's (or CORBA) in real-world situations.

Yes, you are correct about this. Why? Because ODBMS's are currently all non-free. I only work with free software. My point was partially that it would be nice to see such a thing GPLd because then I (and a million others) could learn the concepts, and in the purest vein of free software development, improve it. The problems you mention could be overcome by free software developers.

Everyone in this little thread I've started is complaining about speed. The speed of xml sucks, the speed of RDBMS sucks. Well? Not every application needs to handle a zillion connections a second, you know. I think speed is overrated and design elegance and system maintainability is seriously underrated.

Well, your fingers weave quick minarets; Speak in secret alphabets;

ODBMS?

[graveyhead]

The article is a little light on the details, but it would sure be nice to see an object database bundled with the next RH release. The flat "tables and rows" paradigm (did I really just say that?) is ok for some things, but in most cases, I often ask myself "why not use XML instead?" An ODB would give me a better reason to use a database at all. It would be sweet to store serialized C or Java objects this way, or access them through CORBA.

Well, your fingers weave quick minarets; Speak in secret alphabets;

Re:Who is their target audience?

[hearingaid]

The database market is enormous. While it's true that the big dollars are all in Oracle, it's shocking to note that many mid-level businesses are using Access.

If RedHat has sense, what they should be going after are the types of people who like databases like FileMaker. Access is an awful database program; it's just extremely limiting. On the other hand, mySQL does require some hair value to set up correctly. And the problem is, since M$ confusingly dropped FoxPro (a vastly superior program to Access), there's nothing out there for the semi-computer-literate guys & girls who need decent databases.

Also, if they get a good little server-based database with a decent GUI, not a great one (think FileMaker) then they've got a huge lead on the SOHO market, many of who are interested in Linux but are frightened of SQL. FileMaker did really well with these people in the Mac market; I don't see why RedHat can't do well too in the x86 market.

Anyway, that's what I think they should do. If you're listening, boys, there you go. :)

Flawed benchmark

[OblongPlatypus]

If you read some of the comments to that article, in particular the responses from the article's author, you'll notice he seems extremly partial to Postgres, much more so than he did in the article.

A benchmark is inherently flawed if it isn't made by an impartial entity.

An other point entirely, which was also made in the comments, is the known flaw in glibc which causes severe MySQL performance problems on SMP systems. Running the same benchmark on a single processor machine would probably have shown quite different results, as would a patched glibc.

math

[simpl3x]

the profit was actually in the period they were developing the database. so they were investing in r 'n d at the same time, like many businesses which intend on remaining in business.

Re:Who is their target audience?

[tb3]

Oracle is changing their pricing model; the "power-unit" pricing was too annoying. They're now going for a flat per-cpu charge.

There's an article on news.com about it here [cnet.com]

Whether or not this is going to mean lower prices remains to be seen.

"What are we going to do tonight, Bill?"

Re:Who is their target audience?

[tb3]

Adabas, included with StarOffice, is a perfectly good desktop (single-user) database for Linux, if that's all you're looking for. IMHO, Access is a decent Windows desktop database, while Foxpro is an aging monlith, built on 20-year-old flat-file technology(i.e. dBase II for CP/M).

Anyway, the current market for Linux is the server market, not the desktop market, so they really don't have much to gain by building a desktop database.

"What are we going to do tonight, Bill?"

RedHat Database == PostgreSQL

[Wolfgar]

Please refer to this Yahoo article [yahoo.com] which explicitly states that all RedHat is doing is essentially packaging up PostgreSQL and selling "up2date" type functionality for it.

Of course, why anyone would pay for something as simple as "rpm -i postgres-..." is beyond me.

Hmm... sounds suspiciously like...

[powerlinekid]

I don't want to sound crazy... but Red Hat making their database sounds kind of Micro$oft-istic *ack, gasp*. Hear me out here... lets see... we have Oracle... we have mySQL, we have DB2 (which runs great on Redhat), basically we have a vast number of databases. Yet RedHat decides to make their own. I question their intentions... hopefully they haven't been looking towards Redmond for a business model. The day Red Hat tells me I can only use the Red Hat Database connected to Red Hat Tux 2.0 is the day I tell Red Hat that they can keep their damn ISO.

Re:sap-db?

[germania]

yes.
PostgreSql 7.1 will re-do the logs files in a recovery situation.
(it's called WAL in postgres, Write Ahead Log)

/sergio

SAP involvement? Hmmmmm?

[digitalamish]

I posted before when SAP opened up their db code [slashdot.org] that something like this may be coming. I wouldn't be surprised to see an announcment shortly that SAP runs on the new database [slashdot.org]. They've even benchmarked [slashdot.org] using Intel boxes. Could this be a precursor to SAP and Redhat launching an Oracle and DB2 free ERP solution?

---
No electrons were harmed in the transmitting of this message. EOL

Good for Red Hat

[reposter]

It's nice to see an open source company finally finding viable niches in which they can actually make money, justifying the hype. Although the whiney open-source `advocates' who don't understand the concept of value-added might complain about the sticker price, when you're looking at salaries in the 80K/yr. range for experts on the subject, paying 2K for a year of support ends up making a lot of sense.
Three cheers for open source companies with viable business models.