Oracle's enemies of choice: IBM and SAP; Examining the prospects

Oracle's enemies of choice: IBM and SAP; Examining the prospects

Summary: It's clear following Oracle's fiscal third quarter earnings call that CEO Larry Ellison has two enemies of choice: SAP and IBM. However, Oracle's prospects are much more believable vs. SAP. It remains to be seen whether Sun can tackle the high-end server market and wrest control from IBM.


It's clear following Oracle's fiscal third quarter earnings call that CEO Larry Ellison has two enemies of choice: SAP and IBM. However, Oracle's prospects are much more believable vs. SAP. It remains to be seen whether Sun can tackle the high-end server market and wrest control from IBM.

Analysts were generally upbeat about Oracle's quarter. The integration of Sun is moving along, Exadata is showing progress, and -- on the applications front -- Oracle's Fusion effort is on deck. Meanwhile, Oracle is notching customer wins as SAP works out its vision amid management changes.

Also see: Oracle: Sun integration 'better than expected'

Now onto the enemies. Simply put, Ellison wants to be number one in apps and high-end servers. Oracle isn't there yet on either front. Here are some excerpts from Oracle's conference call transcript and the reaction to the comments.

First, there's IBM. Ellison, who loves Exadata, sees the machine as an IBM killer. Ellison said:

It is not uncommon for an Exadata benchmark, when going up against a big IBM pSeries machine, to beat them by a factor of 10. Because they just don't have a clustered architecture, they're stuck with kind of an old-fashioned SMP architecture for transaction processing. So we're taking advantage of new technologies, not just InfiniBand by the way, but Flash -- very large scale memory -- integrated memories. You will see all of that has allowed as to deliver stunning performance using -- if you will, commodity parts. And not only do we run much faster than the IBM pSeries, but also our architecture is clustered -- there are multiple servers, multiple piece of storage. There is no single point of failure, so much more reliable, while delivering this fantastic performance. And finally it is much lower in cost, because the components themselves if you will are commodities. The components that we -- the disk drives and memory and processors which we assemble with our software. So again we think we can challenge IBM on the high end of the server market. Better performance, lower cost and much more reliability.

Later in the call he responded to an analyst question about hardware plans:

Well, stage one of the version one of Exadata was really aimed at Teradata, and Netezza -- and really the specialists data companies that focused on building custom hardware designed to tackle large-scale data warehouses. And that was Exadata version one. Exadata version two, handles not only data warehouses but also handles transaction processing. We added a lot of flash memory into it. And we are going to have new Exadata models, and I don't want to pre-announce anything, but we're going to get better and better at very large scale transaction processing. Our intent is that the Exadata line challenged the biggest IBM pSeries machines, and beats them badly in performance, reliability, and cost. And we think -- and we think we can do that where -- in transaction processing where we're twice as fast , and in data warehousing, we're 10 times as fast. Those are our goals, to be twice as fast as IBM's biggest, best box, and again, at a dramatically lower price.

Credibility check: When Ellison focuses on something---even when you think his statements are initially nuts---you should pay attention. Analysts, however, are in the wait-and-see mode on Sun. Right now, Oracle has to fix Sun, focus on higher-margin servers, eradicate old business practices, and get used to the hardware business. In addition, Ellison alluded to new product launches---specifically an Exadata server line. That's a lot of work just to get to the starting line. Simply put, it's unclear what happens with Oracle's hardware business once all the low-hanging integration fruit has been plucked. JMP Securities analyst Patrick Walravens quipped: "If Larry Can Fix Sun, Imagine What He’ll Do for the Warriors." For those of you not following along, Ellison is trying to buy the Golden State Warriors. The rehab work for both the Warriors and Sun are similar. Also see: Oracle-Sun’s great society vision: Will customers bite? Oracle-Sun: Strategy set; Will buyers go for the integrated stack?

And now let's look at SAP. Ellison and president Charles Phillips talked a lot about SAP on its conference call. That's nothing new, but Oracle's comments had a lot of detail this go round. As a veteran of Oracle conference calls, the SAP bluster runs high usually. Ellison cranked up the bluster-o-meter, but also made some interesting points. Overall, it's hard to argue that Oracle isn't in a better spot than SAP today. SAP is the big dog in apps, but it's changing management and is building the narrative around its product line. Simply put, it's not that hard for Oracle to win some deals with its focus on industries, functionality, a good story and some good old fashioned fear, uncertainty and doubt (FUD).

Here's Ellison's scene setter:

In applications -- SAP is the leader. But their technology -- that they use for their applications -- is a proprietary technology, a German programming language called ABAP, where later this year -- and that is a 25-year-old technology. But it is still the center of their architecture and strategy for applications going forward is ABAP. The center of our strategy going forward is Java, and a modern service-oriented architecture. And during this calendar year, we will deliver our Fusion applications. We have been working on them for a while, but we have rewritten -- and written in Java, all of our accounting software, all of our supply chain software, all of our HR software, our sales automation, our service automation -- it's all been rewritten in Java with a modern service oriented architecture. And we're going to go compete with SAP's 25 year old technology. The interesting thing is we're competing quite well against SAP now.

We think once we deliver Fusion, we're going to be well-positioned to challenge for the number one slot. One of the important things about our Fusion applications is they are designed not simply to run on premise, which of course they do. But they're all on-demand or if you prefer, cloud-ready. So we'll be delivering those applications, both by selling the software directly -- kind of the old way of doing it, which is still the most popular way, by the way. We'll be selling the Fusion applications integrated with our hardware, our servers, and our storage and our networks. We'll be selling it on the cloud, all modern service-orientated 21st century stuff, competing against SAP. We think SAP is vulnerable. And we can take them on in a variety of industries. The other thing that we're doing, that SAP is not doing, is we're emphasizing industry functionality. So it is not just technology where we are competing with SAP, we're also competing with them in functionality. So we think we have much more functionality for intelligent -- for a telco, a phone company, a large-scale retail operation, in insurance and in banking. We have industry-specific applications for a variety of industries. Health care, I could go on. So our strategy is to have much better industry focus than SAP, in terms of functionality, and a much more modern technology underlying all of that functionality. Again it is a company that we think is vulnerable, and we think we have an excellent chance of becoming number one in applications.

Phillips, after touting customer wins, added:

I'm certainly seeing a shift there, where I would say two years ago SAP was considered a safe choice even if the technology was a little old and complex to implement and all that. But that seems to be changed, as customers appear more nervous about their technology and strategic directions, and are more open to discussing things with Oracle, even the SAP customers, some of the largest customers. And secondly, they don't have anything new to talk about.

Of the customer wins, the one that caught my eye was an Oracle win with the Australian government, which is building out a national broadband network. See ZDNet Australia for the history on that project.

Later in the call, Ellison said that Oracle's ERP sales were up 26 percent in its fiscal third quarter. Ellison noted success in specific verticals---health care, pharma and telecom---and executing while SAP was switching coaches. While SAP was revamping management, Ellison was going on sales calls.

Credibility check: Analysts had no problem believing that Oracle was a big threat to SAP on apps. Sure, Oracle has a lot of ground to cover, but SAP is in transition on many fronts.

Walravens said in a research note:

We believe that Oracle is significantly better positioned than SAP. We see two main areas of comparison. The first is the attempt by the world’s largest on premise software companies to adapt to the computing as a service trend. Oracle seems to be taking the approach of selling shovels in the gold rush – rather than risking adopting the on demand model. Oracle has a full infrastructure stack – middleware, operating system, database, servers – which it can sell to operators of public and private clouds. Oracle also has a new generation of applications coming to market, which it argues can be made available as software as a service solutions. SAP seems to be stuck in the middle in a classic innovator’s dilemma. In the small and medium-sized market, SAP plans to accelerate the deployment of its Business by Design solution in 2010. One problem with on demand accounting solutions for SME is that the SMEs tend to have a lot of accounting questions – and they burn up the phone lines. This is a problem NetSuite struggled with in the past.

Heather Bellini, an analyst with ISI Group, produced this organic growth chart for license revenue that tells the tale.

The big question here is whether Oracle has a significant long-term technology and functionality edge over SAP as it claims, or is taking advantage of a company in turmoil. In other words, when will SAP get off the mat?

Topics: Enterprise Software, Data Centers, Data Management, Hardware, IBM, Oracle, SAP, Servers, Software

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • SAP will be a part of Oracle - eventually.

    Oracle has made the right decisions while SAP has been stagnant. In my opinion, SAP will in some form, become part of Oracle.

    It's the EU that would hold things back.

    Oracle rules!
    • Not likely with IBM around

      It would be more likely that IBM would buy out SAP before Oracle get them.
  • Larry, you've forgot the biggest enemy

    evil M$!
    Linux Geek
    • Wow what a surprising and yet insightful response - I'm shocked

      Didn't Ellison try that and failed or walked away limping? Then went a new direction buy everything ERP based and then target SAP and IBM. It will be interesting to see what happens. So far his record is not that great. They must have him on new meds.
  • One of Oracel's biggest customers and users is

    SAP. I have always found it funny that Oracle hates SAP. Well like MS they write them off the next thing you know they are still there and ever present. We are an SAP shop. They have tried to switch us over and they are seriosly lacking in the integration area. Maybe one day, but right now, you need an army of programmers to one wants to pay for that and with SAP that is their strength, they integrate their systems better than just about anyone we have brought in house. POCs are great with consultants and they ALWYAS fall flat on their face - sad really.

    Despite the talking heads and Ellison which history has shown about as accurate as tea leaves, SAP is ingrained in the ERP culture in so many companies world wide. A wholesale switch for a company in this economy would be risky - it will be interesting if anyone is going to drink the Ellison Kool-aid.
  • RE: Oracle's enemies of choice: IBM and SAP; Examining the prospects

    Any less ambitious goals would not be worth Larry?s attention. Way to go Larry!
    Ivan Vinogradov, Systems Architect
  • DB2 PureScale vs. Oracle Exadata

    Larry should keep up with industry news. IBM's DB2 PureScale is a cluster solution for massive scale-out with shared data files.

    Exadata2 is a nice package. They've eliminated IO bottlecks to disk by going to Infiniband and Falsh Memory.

    There's nothing stopping anybody running any database server from buying these components off the shelf. So if you're IO bound these are easy solutions to add in without buying Exadata.

    RACs scaling problems has always been cache coherency. Every node needs to talk to every node. As nodes numbers increase, there is a geometric increase in inter-node communications.

    DB2 PureScale copies DB2 mainframe Data Sharing with a centralized point (mirrored of course) in a star like communication. This allows scaling to a much higher degree. Communications between the individual instances and the center is done via Infiniband remote memory copy commands. This is a very low latency protocol.

    Of course, Oracle has lots of mind share and market share to drive their sales.
  • SAP slaveshops

    SAP costs monumental amounts to keep afloat if you get it. The whole regime is 1970s and designed to suck money out of the customer. As for their implementation language.... Someone said that you never can know that any item of data has actually GONE from a SAP installation. SAP deserve to go and I feel that their star is inexorably being eclipsed.
  • RE: Oracle's enemies of choice: IBM and SAP; Examining the prospects

    Actually i don't think acquiring SAP would make
    any thing profitable for Oracle, including the
    bad reputation SAP trade mark it self, because
    in most SAP implementations - Including the
    company i work in - the case is always a
    troubled, complex, tensioned, unfinished and
    consequently an endless paid support for a
    never ending SAP project.

    That's why if Oracle acquired SAP they may not
    even loose for SAP but they may also loose the
    Oracle company itself.
  • RE: Oracle's enemies of choice: IBM and SAP; Examining the prospects

    The only chance Ellison has of accomplishing his goal of going to war with SAP and IBM is to join forces with MS. Otherwise, he is barking at the moon again...
    Woned B. Fooldagan
  • RE: Oracle's enemies of choice: IBM and SAP; Examining the prospects

    Absolutely. MS being excellent on front end & oracle with
    back end.. it has been a good combination since client-
    server days. Combination can certainly be a challenge to
  • The real problem with Oracle...

    is not the Oracle + Sun solutions that could make quite a (to come) interesting integrated portfolio, but the tremendous arrogance of most of the company executives, Larry Ellison first, but Charles Philips, Safra Catz and Vishal Bhagwati not being that far behind! Bla bla bla and a bunch of FUD!

    BTW, speaking about benchmarks, I just found some interesting information comparing a newbie (G-WAN) to most of the current purported market leaders! Take a look at
  • Oracle is very reliant on SAP Customers for revenue....

    SAP Customers and systems are a major cash cow for Oracle's licensing / maintenance revenue. The stacked design of SAP is such that it is very easy (though time consuming if you have teraabytes of data) to move from one DBMS to another - all the logic and business rules are in the SAP proprietary layer or the Customer layer (a better description would take up lots more space).

    So, if Mr Ellison gets too aggressive with SAP (whether its in the ERP market in general or over licensing Java or the Java trademark for the SAP Netweaver stack or some other triviality), then SAP do have the nuclear option - saying say something like
    "We're giving you free support to export your existing Oracle DBMS to any other SAP supported DBMS, so long as you do it within 6 months. Any one remaining on Oracle after that will be COMPLETELY unsupported".

    IBM are in a similar situation, except
    a) the impact of SAP Customers on IBMs bottom line is a much smaller proportion, and
    b) IBM's Services division are very willing to make / provide around hardware / IBM Software / 'Consulting Fees' to get work (including SAP implementation and support).

    In fact, the more I think about it... Apart from Larry Ellison, who else cares enough about Oracle to keep THEM off the merger / acquisition menu ?
  • RE: Oracle's enemies of choice: IBM and SAP; Examining the prospects

    Oracle rocks

    If you want to find more information on ERP software reviews, go to You will get alot of tools and reviews to find the best software application for your business