Computerworld Australia talks to Angus MacDonald, Chief Technology Officer at Oracle's A/NZ Systems Line of Business about the local implications of the Oracle and Sun merger.
How far down the road is Oracle with its Sun merger?
A lot of people are looking at this as if it occurred in April last year which would suggest we have had over a year to sort out what goes where. But the acquisition wasn’t completed until January this year, so the acquisition is a lot younger than what most people think…
What has been the customer feedback around the merger?
What IT managers and CIOs really want [is]… some comfort around is that nearly everything they were used to in the Sun Systems product line is still there and is still being invested in.
The only real rationalisation … has been a complete refresh of our x86 product line which we did very quietly… around the Nehalem and Westmere processors. It’s what customers really wanted and it removes confusion. We have also consolidated some of our storage strategy around our 6000 SAN and Fibre Channel connected arrays around unified storage which Sun used to call Open Storage, and around the focus on flash and tape.
There really hasn’t been a lot of significant investment from (Sun) customers in Australia which [customers] would say has been damaged, other than potentially the moving away from Hitachi storage, which was a decision taken largely because our own storage portfolio is broadening in that area.
Are customer fears around the merger, which are leading some to migrate from Sun, unfounded, then?
I think the fears are unfounded, but it would be fair to say we have a fair task in front of us to go and explain this openly to all of the customers. As you can imagine, for the period after the acquisition was originally announced, and with the European Union asking questions, a lot of customers became a little nervous around the product line.
What are your key messages to CIOs and IT managers around the merger?
For CIOs and IT managers, the key message is what Oracle has done with the acquisition of Sun has not jeopardised the investments, direction architectures, strategies these people were building on. The CIO can look and be comfortable that past investment is being protected, enhanced and developed. On top of that is that as a result of the totality of the stack, Oracle is now in a position to do what neither Sun nor Oracle alone could do. If you look at the Exadata 2 product we have taken part of the database code itself and taken it down into the storage product itself. We can only do that because we own both the IPs.
How will refreshing Sun’s x86 line address Sun’s declining server market share?
The flagship product in our x86 line is an eight socket Nehalem server, so if you consider cores, sockets and hyperthreads that manifests itself as a 128 thread machine as far as operating systems and virtualistion are concerned. That is an unusual addition. What we are finding is that now there is some certainty, customers are starting to recommit. We are also talking to customers about what used to be the old UltraSPARC T2 [microprocessor]. We are release a T3 based system a little later this year. Oracle is also investing a lot more in Solaris than Sun has been doing and the same thing applies to SPARC.
What’s happening at the higher end?
The N series continues. We have been talking about how that proceeds and what we’ll be doing next year. People who have invested in that have treated it as a mainframe class and are looking for many years of investment return on that. They have started refactoring it into their architectures and we expect significant success in that area. At the top end arena, there are things like the tape libraries. The SL8500, the flagship tape library, we have seen substantial sales of those since the change in control occurred. People investing in the M9000, 8000, 3000, 4000 and 5000 will continue to have the ability to do that. There is a roadmap available to customers under non-disclosure which details that.
What can you tell us about the future of Solaris and SPARC?
I don’t observe as many customers leaving those platforms as I keep hearing. I think the most positive message we have been able to give those customers, which has staunched any flow, is the increased investment. The release of Solaris 11 by the end of this calendar year… the focus on scalability, being able to cope with 10s of terabytes of memory, thousands of threads and substantial improvement in I/O scalability customers seeing that are able to recommit and those who haven’t moved away are feeling a lot more comfortable about continuing with it.
On the SPARC front it is similar. Bring forward the T3 processor has given a lot of customers confidence that it is serious, and it has given customers the confidence to come back if they had shied away from it.
How will a commercial Solaris affect its community and developers?
We have stated quite openly that Solaris will be a superset of Open Solaris. That is not suggested that Open Solaris is going away. We will continue to focus and invest in open source generally. I think people who may be concerned with Oracle’s approach to open source need only look at our pedigree in Linux which is probably the longest proof point we have. I personally don’t see a challenge in it.
What is the future of Java?
Java is something Oracle has a very strong vested interest in and continues to develop. Most of Oracle own applications are written in Java.
Follow Tim Lohman on Twitter: @TLohman
Follow Computerworld Australia on Twitter: @ComputerworldAU