This article is more than 1 year old

Microsoft lays into Oracle (and Sun)

Future belongs to SQL Server/Itanium

Interview IT-Analysis spoke to Gordon Mangione, the vice-president of Microsoft's SQL Server Team, about SQL Server and where he sees it in the market.

How's the year been for SQL Server?

SQL Server has had a truly tremendous year. We've nailed down 35 per cent year-on-year growth. We've broken the $1 billion sales mark. And we aren't leaving it there. We've got Oracle firmly in our sights. We are catching them with ease right now, last year we outdid them on NT whilst they remained flat. Our database business is one of the best places you could wish to be right now.

I think Oracle would probably beg to differ on a few of those points.

If you look at Oracle, it's a one product company. Oracle is in serious trouble. It's revenues are down, its applications aren't any good and its database revenues fell 5 per cent. And we are stealing market share from them very easily. Oracle is getting hit from all angles. IBM is hammering from it from one side and we are hammering it from the other. They're trying, there's no doubt about that, but Oracle is a company in trouble.

And how do you believe that Microsoft is hurting Oracle?

We are hurting them with our database and its OLAP solutions. Oracle is trying to keep up with its relational stuff but it really can't compete. If you look at what customers are demanding today, they want real business intelligence generated on the fly, in real time, so that they can respond there and then. If you're going to tackle this right you need to be able to develop many dimensions in your cubes and with a relational database that's very difficult to do. It is possible but as you build them you bloat your database and you slow the queries to snails pace. That's no good and that's just one of the areas where Oracle is messing up.

How do you believe your BI solutions are superior?

Our business intelligence runs across all platforms, you can stick it wherever, Oracle, IBM, and we're finding that more and more people want to do this because our OLAP proposition is so strong. This is a very viable business for us. We can sell our OLAP solutions right on top of Oracle and the customers love it because it so fast. It beats Oracle hands down.

But you operate different markets. Microsoft plays at the low end whilst Oracle does business at the high end. How can you seriously say you compete?

Just look at our growth rates. Look at the market. The database market is huge right now and growing at impressive rates. We are well positioned to take that market from Oracle. The work that we have been doing with Unisys and Compaq running very large, enterprise-class systems shows that isn't true. We have benchmarks on Unisys systems with SQL Server running SAP with more than 20,000 concurrent users. SQL Server is used to run businesses.

The fact remains that SQL Server is not considered an enterprise class system and it won't be until Windows is considered an enterprise class system - and that's a good few years away yet. Isn't it about time you ported your products?

I do appreciate that we have a perception battle to fight in the high-end and we will continue to do that. But as far as porting our database that simply will not happen. We will not port to other operating systems. We are absolutely committed to Windows. If you start porting your software you have to get into all sorts of deals and all sorts of concessions and ultimately you end up with a preferred platform anyway. I don't believe that porting is good for the product. We work with Unisys and Compaq very closely to make sure that our Windows systems run very well in the data centre and we are completely committed to that.

I don't understand how you can feel confident in winning a database battle if you can't operate in the Unix environment.

If you look at the investment that we are putting into this and look at the investment that Intel is putting into this you can see how we can feel that way. True, we may be behind with our efforts for the Itanium chip, and Mckinley is going to take a while, but there is no way that Sun can keep up with Intel in the chip market and that's where the lines will cross. Sun will find itself behind the curve and Oracle will find itself losing ground. If Sun can't keep up with the game then Oracle has lost the fight. These two are the same company. Oracle and Sun are one. As soon as Sun loses the high end so does Oracle and that's when we will take it.

What makes you think Sun can't keep up in the chip race?

It can't maintain that level of investment. Sun has got inventory problems. Sure it had a very good couple of years with the dot coms but since all of that collapsed Sun has found a lot of second hand kit in the market and its inventory is stuffed. Now Sun is struggling.

Okay, you might think you can take Oracle, but that still leaves IBM out there. How are you planning to take IBM out of the race?

IBM is a very serious, tough, competitor. We have a better product but IBM has a huge solutions sales that some people seem to like. Some of the buyers out there like to be able to buy hardware, software, services - the whole lot - in one go. But I think people fail to realise that Microsoft, with its OEM's, can provide that very same solution. That's how we plan to tackle IBM. There is the perception that IBM has its one stop shop, but the chance to have multiple OEM partners, with their dedicated and honed skills, is a very strong proposition and it provides us with another route into IBM's territory.

And what about the open source movement. Red Hat database must be expected to apply pressure to Microsoft at the low-end. How are you going to cope with that?

To be honest with you, I don't think that you will find many people running production systems on Red Hat. But please don't think that we don't take it seriously because we most certainly do. Bill and Steve are very concerned about Linux and so are the rest of us. We won't be taking our eyes of Red Hat's database. We will be watching very carefully to make sure that it doesn't start to dent us at the bottom. We will be very focussed on this effort believe me. Having said that, and I really don't want to diminish the efforts of Red Hat, but I think that the dedicated SQL Server team of more than 800 professionals dedicated to the job is an investment that can't be matched. And that will continue to give us our edge.

I imagine that the inclusion of your OLAP engine into SQL Server must have hurt plenty of OLAP vendors. Did that take many of them out of the race?

I think if anything the OLAP engine has given people a common platform to develop to. We can go out to other OLAP vendors and give them a toolkit that helps them build for this solution. We've had a lot of vendors do that since we built this stuff into SQL Server. I don't think that it has damaged the market. This is a very big marketplace and there is plenty of money for people to get. Some OLAP vendors have aligned with us and others have chosen to focus on their own customers and developing serious solutions for them. Our OLAP solution has done some good things.

Do you see OLAP as key to SQL Server nowadays?

I think it is, yes. The market is increasingly looking for market data that is immediate and that can be turned into reports on the fly so that it can be responded to immediately. This is especially true in the area of web analytics and click-stream analysis. And OLAP really puts us ahead of the game in this area. This gives us a huge opportunity.

How does SQL Server fit into .Net?

SQL Server is the foundation of .Net. We took a decision some time ago that SQL Server would be key to all of application developments and that is precisely what we are dong with .Net. It has given us a new strategy to rally around and everybody is working so hard to achieve this. Every application from now will be built with SQL Server in mind. Naturally, other databases will be compatible but all of our systems will be optimised for SQL Server.

You have to realise that we have a very unique offering at Microsoft. We can do things with our software that no one else can do. We can say to people use our solutions and you will find that, as a whole, they are much, much more than the sum of their parts. That's what .Net is all about and that's what are heading for.

Copyright © 2001 IT-Analysis.com. All rights reserved.

More about

TIP US OFF

Send us news


Other stories you might like