Ok, you have a big data problem, but can we stick with Java now?
It’s funny how it goes. I’ve worked at Rogue Wave for 13 years and you probably remember us as a C and C++ company only. So you’ll probably understand why I never would’ve thought I’d say this: “can we stick with Java now”?
Years ago, I was a bit dumbfounded when I first met people at low frequency trading seminars in NYC or Chicago talking about trading algorithms in Java. Even one of my best friends moved to Dublin to write artificial intelligence algorithms in Java for a market maker.
Java first triggered so much interest because it was easy to use and a lot of libraries and application servers were there to help you. But later, people became experts, building entirely new real-time systems with it! Of course, not everyone got there, but Java empowered developers to do new things that they couldn’t do before. As a matter of fact, the big buzz word was Web 2.0 at the time, and Java was the language meant to deal with the Web Problem.
So what happened to C and C++? Did they become irrelevant? Well, actually the opposite is true. They’re clearly still here, constantly switching places with Java as some of the most-used development languages out there – not everything you do needs an application server to run. Not everything fits the model of a Java Virtual Machine. When you trade, it’s still mostly happening in C++.
C/C++ and Java are side by side, playing an important role in our future development landscape. I’d just stick with Objective-C and Android-flavored Java as good examples.
I’m sure that if I drop my Java, I’ll make a mess.
Why would one think about dropping Java? Well, there’s a new sheriff in town and it’s name is big data. New problems and new rules means a new language, the same way Web 2.0 promoted Java. So, when your boss asks you to look at the “big data initiative”, what’s your first step?
Drop the Java and start learning SaS, Matlab, R, or another programming language. Why? In the same way that Java offered nice add-on’s that let you focus on your business flow without having to write a database or web connector, new data analytics tools and languages enable you to focus on your data problem without writing data connectivity or complex charting. It’s all there!
So is Java dead or will it stay around to handle your data needs in way you haven’t seen yet?
First, the landscape is so fluid in term of big data solutions, that no one can be sure of what the next flavor of big data Java will be – will it be universally accepted or, like the next big data Visual Basic, universally avoided?
Second, it’s highly probable that Java will be used in ways that will keep it relevant to tomorrow’s big data initiatives. The perfect example is the rise of Hadoop, Cassandra, and Java-based technologies where Java remains a platform of choice for transparent virtualization.
|And another perfect example is a project that one of my co-workers, Mark Sweeney, has been working on.||Read the tech tutorial by Mark:
Embedding analytics into a database using JMSL
So how does all of this relate to big data?
My point is really simple, Java remains relevant in the big data world the same way that C/C++ has remained relevant. Jumping on new big data tools and languages is the right way to go, the same way that accepting Java was. Tools like Hadoop and Cassandra along with many interested people are taking the initiative to develop new ways to use Java effectively with big data.
My friend Mark has developed a creative and unique way to use Java in an Oracle database to solve complex data problems. It even outperforms today’s preferred solutions, Oracle Advance Analytics based on R. So why not simply write your analytics in Java within Oracle instead of R? This webinar shows you how: