Tonight was a significant evening for Sun & Oracle. The opening keynote of Oracle OpenWorld 2009 was provided by Scott McNealy with an appearance by Larry Ellison. There is a lot to unpack here, so I’m going to break it down into sections.
This is the first event I’ve attended as “press”. As such I got all the press perks, namely access to announcement details in the press room and early access to the keynote for prime seats, in my case, right behind all the Sun Microsystems reserved seating. Talking with others there to cover the event was interesting in and of itself. Clearly everyone was looking for new information regarding the merger. None was expected but it was hoped for and we all listened attentively.
Here you can see (from left to right) Dr. James Gosling (creator of Java), Scott McNealy (Sun Chairman), and John Fowler (Sun EVP Systems). These were the speakers from the Sun side of the house, only Larry came up to represent Oracle.
Scott obviously thought the acquisition was a good thing and spent a lot of time about the history of Sun as an innovation driver (building his legacy). James came up to say that he thought it was a good thing for Java and the only hiccups along the way were with regard to Oracle learning now to interact with a community as large as that around Java. John showed off some of the new goodness from Sun and pointed to Oracle continued interest in Solaris, SPARC, and X86.
Larry wanted to hammer home the commitment they are making to Sun. He talked about the recent ads that Oracle’s been putting out and how they are fighting against IBM who’s trying to capitalize on the confusion. He re-iterated that Oracle will increase the money going into SPARC, Solaris, Java, and added to the list MySQL. He’s very clear that nothing is getting chopped, he needs to whole company. With regard to MySQL he pointed out to Sleepycat (BerkelyDB) and InnoDB as things that Oracle owns and has invested in and been able to make some money with and intends to do the same with MySQL. He maintains that MySQL in no way competes with Oracle.
The more Larry talks the more comfortable it seems everyone is getting with this deal. Early estimates were that almost 50% of the company would be let go and there would be major changes in the companies product lineup. More and more those estimates are dropping below 30% and suggest that nothing will be cut, but rather pruned neatly into a more structured form. Best line of the nite was from larry, “We’re in it to win.”
Larry drove the point about synergies between Oracle and Sun home in 2 ways. The first was talking about the previously released Sun/Oracle ExaData v2 product (pictured above). The second was to show that with Sun’s technology today, pre-acquisition, is the best platform available for Oracle even against IBM’s monster POWER 595 system which consumes 76 standard racks. Sun’s solution that beat it consumes only 9 racks, is fault tolerant, based on SPARC (Niagara), got 25% more throughput, gets 16 times better response times, and obviously uses a hell of a lot less power to boot.
I had a conversation with the PAE guys there and got a lot of great details on the configuration and how they made it work. Here are some highlights…
So the Sun system that beat out the 595 was based on T5440 (UltraSPARC T2) systems connected to the new F5100 Flash Array. In order to make all this work in a fault tolerant way COMSTAR was used and throughout the process required absolutely no modification! Apparently the biggest “problem” they ran into some some minor tweeking in the mpt and sd drivers because they weren’t designed to hand the extreme number of IOPS coming from the flash arrays. More shockingly, when they got the TPC-C number that beat IBM the CPU’s were 50% idle! And, if you can believe it, during the whole time Sun was working on this benchmark of all the flash modules involved, only a single one failed! Just one!
The F5100 was illuded to a couple months ago by Andy Bechtolsheim; a 1U storage array filled to the brim with Sun SO-DIMM form factor Flash Modules. It can be ordered with as little as 20 modules for 480GB raw or as large as 80 modules for almost 2TB raw capacity. Sequential Write performance on the 80 module unit is rated at 9.7 GB/sec. It physically connects via SAS.
The F20 PCIe Flash Card is just a smaller version. Up to 98GB of Flash rated at 501 MB/s Seq Write. All the goodness of high performance flash storage but you just drop it into a PCIe slot and go. A fantastic solution for databases in need of fast logging capabilities, just plug it into a PCIe slot and define it as your new log device.
Guess who wasn’t present. Jonathan is nowhere to be found. In fact, I haven’t seen him since JavaOne. Since this deal has occured Jonathan has been pushed to the back seat while Scott has insisted on driving. The question is why?
I’m very curious how history will record things with all the details filled in. Did Jonathan sell us down the river? Or, perhaps, Scott’s been driving things far longer than we realize and Jonathan has been something of a pawn in the latter days of the company. Its clear that it was Jonathan’s management of the company that delivered us to the point acquisition was required, but we can’t forget that he did do a number of good things, even if they didn’t actually benefit the company in return.
I’m not going to make a judgment call just yet… but I’m starting to almost feel like Jonathan got screwed here more than we realize. Never the less, he’ll have his millions of dollars to console him while the rest of us are left holding a fist full of memories and broken dreams.
Watch the Keynote!
Make some time, OnDemand Replays available… you’ll enjoy it.