The crazy people that said that Oracle bought Sun just to shut it down and use the patents to sue Google are looking pretty non-crazy now. I guess once the lawsuit was lost (and boy was it expensive) the shutdown was inevitable.
7 replies (most recent on top)
" probably thought they could run the business better than my little pony was doing."
No, they definitely thought that.
Right after the layoffs, there was a major attitude of "time to clean house! Let's show these Sun guys how it's done!"
I was at an Oracle HQ briefing sometime within the first year when one of their top level execs had to tell us, "yeah, we didn't get HW supply chain at all, so that's why we can't deliver your systems yet. But we totally get it now!"
Oracle wants to make money. There are hardware units that make revenue and profits and cutting them doesn't make sense.
It seems Oracle wanted to get on the low-cost RISC CPU game that IBM kind-of started, but it failed when Sonoma didn't make the expected sales. Again, it impacts SPARC in a big way, but as long as there is a market for large SPARC machines, such as OLTP, Oracle will continue to offer something, particularly since the current SPARC roadmap looks right and low-risk. Without changing the underlying layout, just by going from 28 nm to 16 nm at TSMC will achieve 40% higher frequency (that roughly corresponds to 1.4x increased throughput) and reduced die size per core will allow 50% more cores (for 1.5x thread strength).
And while database acceleration in hardware, which used to be touted as the killer feature, works, it simply cannot match the flexibility of an FPGA chip, and it seems the entire industry is moving in that direction, so controlling your own architecture no longer offers the advantage it used to.
No, they still sound crazy. Really - you think the master plan is Oracle divests business units on a 7 year cycle? Java and the customer install base were the key incentives to buy Sun, the hardware was just sauce on the side and probably thought they could run the business better than my little pony was doing. But I doubt anyone thought you could make big margins like Oracle is used to with the hardware. Should have cut back long before now on declining business. Their focus is cloud and Exadata, not hardware and storage.
Hardware business at Oracle will go away eventually. Slowly and steadily they will keep cutting. Matter of time.
b---s---.
First off, they didn't lose the lawsuit yet, the court of appeals might rule in Oracle's favor.
Second, Oracle didn't need to keep the hardware business to sue Google. They basically just needed to have ownership of Java and active development for that lawsuit.
And come on. Oracle held on to Sun hardware for 6.5 years, 26 quarters since Sun LEC. All those consultants claiming that it would be the quarter that Oracle will drop hardware were bound to be right finally. It was obvious at some point, Oracle would review its hardware business and decide to lay off people. And even now, not everything was dropped. Some engineering teams were hit harder than the others, but generally, it was not total carnage.
Suing for success financial and otherwise makes as much sense as cutting your way to profitability. I'd say Sun/Solaris/Sparc had a much longer life under Oracle than it would have if IBM purchased them.
Amen!