It was a good first full day at Oracle OpenWorld. It started with the keynote led by Oracle’s CEO. Of course he was very upbeat about Oracle’s products. But, I found his comments about the economy and the way it affects IT spending more interesting than Oracle’s cloud products. My translation is that companies have a lot of older systems that they can’t easily retire or upgrade but they want to quickly add all this new functionality. I see that in my company so it rings true. I don’t really believe that the cloud is the long-term answer but it makes me wonder what the real answer is. I always come back to prioritization. I think prioritizing spending is more important than moving things to the cloud. You can’t afford to do everything so you have to make tough choices about what to spend your IT dollars on. That’s my opinion at least.
Next I went to a session on Coherence. I kind of went out of a sense of obligation since our company owns the product. But, it was a surprisingly good session. It had a lot in it about Java 8 and the new features in it for parallel processing. It made me want to dust off my Java skills and generally think about parallel processing in the Oracle database and how it relates to that in Hadoop, Scala, etc.
I went to two sessions on analytics, again out of a sense that I needed to learn about analytics and not due to any enthusiasm about it. The first session was really interesting, but the 3:30 session almost literally put me to sleep. The first session reminded me of some of the things in Craig Shallahamer’s forecasting book such as making a model of a system and doing validation of the model. Analytics seems to follow a similar process. But, by the late afternoon a non-technical session on analytics in banking nearly knocked me out.
Wedged between my two analytics sessions I went to a very cool In-Memory Option boot camp. I have not had the time or taken the time to look at the In-Memory Option and I got a nice fast hour-long exposure to it. I don’t know if the other people in the class were lost because there were a lot of explain plans flying by but it is the type of stuff I’m interested in so it was nice that it was so technical. The In-Memory Option reminded me a lot of Exadata smart scans and hybrid columnar compression.
Strangely enough multiple speakers pronounced columnar differently than I have done so I guess I will have to change. They emphasize the second syllable but I usually emphasize the first.
I also snuck in to the OakTable World presentation by Tanel Poder. It had to do with querying Hadoop clusters from Oracle databases using odbc/jdbc. Pretty cool. I also got to scope out the venue for my talk tomorrow in the process.
That’s enough for today. I got a lot of good information. I’ve slotted tomorrow for OakTable world so it will be interesting to see what all I can learn there.