Oracle Linux Explored

M-Saunders writes "Two days ago Slashdot reported on Oracle's move into the enterprise Linux market, and how it may challenge Red Hat. Red Hat's stock has already dropped, and there's a great deal of talk about the implications of this act. Linux Format got hold of the 'Unbreakable' distro to find out what's going on under the hood. Is it a breakthrough for Linux in the corporate market, or just another RHEL respin? See the article for all the info and screenshots — including an 'interesting' choice of GRUB colours."

Comments

Re:Installing Oracle on linux

[kripkenstein]

I would be more than satisfied if they come with an easy solution for installing Oracle flawlessly on most linux flavors!

That would be nice, but how about if instead of a full-fledged distro, they put out a barebones Linux+Oracle, all set up and configured, that is then run in a virtual machine. Sort of an "Oracle Appliance". Saves the hassles of supporting various distros, and even saves the hassle of supporting an entire single distro (since people will install other things than Oracle on their "Unbreakable Linux"es).

I haven't used Oracle products in several years. Anyone know why they aren't doing this (or are they, and I am just ignorant)?

This will help others adopt Linux

[ciurana]

Greetings.

There is a wrong perception that large companies don't adopt Linux because they prefer commercial offerings. This is only half right. It's not that they like commercial software per se, or that they don't know or understand the benefits of open-source software. The real issue for the lack of adoption is the perceived legal exposures of running software and becoming liable for it (SCO, anyone?). These large companies would be happy to bring Linux in-house as long as a larger company offers some kind of indemnification clause in their contracts.

Many large companies offer Linux distributions and absorb the indemnification. It's no wonder then that superior distributions like Ubuntu aren't on the enterprise shopping list: there is little or no viable indemnification offered. Red Hat is a big fish among open-source vendors but not large enough to convince many large enterprises to take the plunge. That's why IBM has made a good play in this arena: their Linux offerings are rather crappy, but they offer the magic word: INDEMNIFICATION. This has opened many doors for them that remained shut to other vendors.

An Oracle offering brings the same "large company support" that will let the pussies in legal departments and the dumbass middle managers sleep well at night. Oracle is already known to work well with Linux; couple that that with Red Hat functionality and Oracle support (especially if other Oracle products are involved) and that makes a very attractive proposition for all the parties involved. If Oracle plays this right they can start by offering Red Hat dressed in Oracle garb as they came out of the gate, and then provide a migration path toward Ubuntu or another Linux distribution with better tools.

Oracle didn't get that big by being idiots. They are smart and they are aggressive. I think that this is overall a good thing. It creates more competition for IBM, who perhaps now will actually push for real Linux offerings that work, for Novell with SuSe, for Sun and Solaris, and it opens the door for upstarts like Canonical who are well-positioned to make Ubuntu a household name. Last, it will open doors to Linux that would otherwise remain shut. Oracle Linux marks the maturity phase of the first round of consolidation and is the harbinger of the next distribution wars. The next five years will be very interesting.

Cheers,

Eugene Ciurana

Our biggest competitors are our customers

[Anonymous Coward]

'Our biggest competitors are our customers' - paraphrase of a shareholder statement for a high tech company.

Companies don't NEED Red Hat support. All the documentation they need is freely available. They could provide all their own Linux support. The reason companies buy support from Red Hat is because it is cheaper and more reliable than doing it in-house.

Companies do need to buy support from Oracle because it is closed source. On the other hand, the open source databases are getting better. Oracle has two challenges: it has to provide better support than Red Hat (Oracle has a lousy rep.) and it has to fight off the steadily improving open source databases. In the long term, things don't look that good for Oracle. In the short term, we will see if they can use their superior size to crush Red Hat.

Re:Its the support costs that are interesting

[anandsr]

The Price of 1999$ is for Premier unlimited support for 1 year. The Lifetime warranty is for 5 years and beyond. Basically if you buy support for 5 years, then they will support you indefinitely except a doing certification with a third party. I believe that it is fail safe, as in Oracle will probably not be the leader in DB after 5 years anyway. And I wouldn't see anybody committing for 5 years to any software even if it is Oracle.

Re:Installing Oracle on linux

[Anonymous Coward]

"I really dont see any threat to RedHat"

I am not so sure about that. I use CentOS and it works fine. However, if I try to convince my boss we don't need to spend thousands of dollars for Red Hat support (which outside of getting patches I have never used) he says "What the hell is a CentOS?" Now, if I say we can get "Oracle Linux" for a fraction of the cost, he might say, "Oracle, wow that is a big name that gives me a nice warm fuzzy feeling!" (OK he probably wouldn't say that exactly)

Point is, PHB likes a big name. If he can get a big name product cheap, you better bet he will think about it.

Re:Oracle certified on most linuxes....

[ajs318]

They don't certify it on $RANDOM_DISTRO because it's a proprietary, closed-source product, and there's simply no way to ensure that it will work with every possible configuration when there are so many variables over which they don't have control; paths, library versions &c.

Back in the day, even proprietary software used to be semi-open source. You actually got the source to compile (almost no two computers were similar enough to be binary-compatible, which was why C was invented in the first place) and tweak if necessary; you just weren't allowed to distribute copies.

Oracle may be losing relevance

[mcrbids]

I'm CTO of a small but growing just-barely-post-startup. (EG: We're profitable, and growing fast)

For me, Oracle is a non-starter. It's big, expensive, and reportedly has a high management overhead. So why would I bother?

So far, I've seen massive growth easily and handily supported by PostgreSQL. It's been rock-solid, very stable, secure, and installation consisted of typing two commands:


yum install postgresql-server;
service postgresql start;


We're experimenting with Slony PG clustering, with the intention of rolling that out over Christmas break. (when nobody's looking) Currently, we're snapshotting and mirroring databases hourly, but we want real-time failover...

Re:Oracle may be losing relevance

[kpharmer]

> For me, Oracle is a non-starter. It's big, expensive, and reportedly has a high management overhead. So why would I bother?

you often wouldn't for a start-up - assuming modest data volumes postgresql is a great choice

But let's say that you've got a 200 gbyte table, and your query is doing table scans because the selection criteria identifies more than 5% of the data in the table. Ok, on db2 or oracle with partitioning, parallelism and very good query optimization that might take you, say, 2-5 seconds on $40k in hardware. Not too bad. How long for postgresql? 20x as long for the table scan and 4x as long for the serial activity. So, 160-400 seconds. Meanwhile you're pounding the snot out of your server.

Of course, you could buy a million dollar machine to crunch the data more quickly with your free database. But it would probably be cheaper, easier and faster to just buy a $20-40k database.

And sure, you could keep the data in 200 separate tables and use a union-all view to concatenate them together. Postgresql will do this part. Assuming you want to keep your data in 200 tables, and assuming that you can take the performance when you do have to scan through them all - and it tries to union them all together.

So, yeah - postgresql is a great database, and I'd probably want to use that too if I was in your shoes. But in my shoes I've got a lot of data, and need to scan tons of it quickly - to ensure my users get subsecond response time. Saving money on postgresql here ultimately loses money in customer revenue.

Re:Do these people even know what "Enterprise" mea

[DrunkenPenguin]

I don't think Oracle linux isn't enterprise class, I just think linux suffers a stigma of gross amateurism.

I'm Unix administrator in Fortune 200 company so I guess I should know what Enterprise class is. Did you happen to see the list of other partners who joined the Unbreakable Linux program? Have a look. Hmm.. let's see. HP, IBM, EMC, BMC..etc.

Linux with EMC Symmetrix high end fibre channel storage support, Linux with HP Service Guard mission critical high availability cluster management software, Linux with BMC enterprise class system monitoring tools... Hmm, well, I'd say that's pretty Enterprise class - in fact, that's as much of Enterprise class as you can possibly get today. All these solutions I mentioned have been implemented in Linux today - they are right here, right now. We're not talking about future.

Re:CentOS?

[Anonymous Coward]

Red Hat says its a fork, Oracle says it isn't. Who to believe. See my remarks on my blog http://www.theciocompanion.com/