7. Machine learning. This is pretty much the new name for artificial intelligence. On the one hand, it's important not to overpromise the near-term potential of machine learning. On the other, it's essential in making sense of big data, and open source projects like Mahout and Spark/MLlib are easing the path. As James Kobielus noted earlier this year, machine learning is so pervasive, we can often assume its presence in big data applications. IBM is mainstreaming that idea by opening Watson APIs, while startups like the Andreessen-backed Adatao are applying today's abundant compute horsepower to revive neural net algorithms.
8. The return of devops. This mashup of "development" and "operations" is really about increasing the efficiency of operations in order to make agile development real. The devops trend first emerged five years ago, but vendors have revived it, grandfathering in application lifecycle management, automated testing tools, database virtualization, release automation, configuration management, application performance monitoring, platform as a service, and related technologies.
In some circles, devops is considered a way to give developers ongoing responsibility for applications in production, but that doesn't scale. It's probably best to think of devops as shorthand for the most modern and efficient way to configure dev and test environments, which must scale up to meet today's almost universal business need for more and better applications.
9. The end of network switches. No, we won't see network switches disappear in 2015. But virtual network devices, software-defined networking, and the abundant horsepower of servers are leading to a major rethink of the data center network. The long-term prospect of the network being reduced to "the wires between the servers" is becoming more real.
Cumulus Linux brings the network control plane to industry standard hardware, and within reach of today's server orchestration tools, while preserving wire-speed network operations. A recent OpenFlow project released by InfoBlox called LINCXthis year shows the potential power of a completely software-programmable network. Meanwhile, NFV (network function virtualization) -- leveraging server virtualization and data center orchestration to deliver load balancing, firewalling, WAN acceleration, and other network functions as a service -- is all the rage amongservice providers and cloud platforms like OpenStack.
The open source imperative
A common thread runs through most of these nine trends: Open source is leading the way in technology development. It's become the vehicle of choice for startups to gain traction, as customers -- mainly developers within companies -- take new technologies for a spin, provide feedback, and eventually put them into production. Meanwhile, other developers see what's hot and start building an ecosystem around a core project, as has occurred with Docker, Hadoop, OpenStack, and others.
The simple model of open source development -- collaborative, self-organized, and distributed -- even appears to be having an impact on enterprise app dev. That trend will take years to unfold, although some companies are now experimenting with it.
Sign up for CIO Asia eNewsletters.