It should be obvious to everyone that technology is evolving faster than ever, and we found ourselves working overtime in 2014 trying to keep up. New languages, frameworks, and platforms for building enterprise mobile apps -- invariably powered by Node.js -- demanded our attention. The arrival of Java 8 prompted refreshes of the major Java IDEs. The PaaS wars led us to the inevitable dilemma: Cloud Foundry or OpenShift OpenShift took the crown this year, but keep an eye on IBM Bluemix in 2015.
Little big iron
Even data center hardware is evolving at a breakneck pace as the vendors work to shove more compute power, more IO, and more storage into smaller spaces.
One of our hardware winners, HP Moonshot, packs a whopping 45 server cartridges -- ARM, Atom, Opteron, or Xeon -- into a 4.3U chassis. With energy-sipping cartridges designed for specific workloads such as virtual desktops and distributed data processing, Moonshot is a triumph of high density and high efficiency.
Another winner, the Dell PowerEdge R730xd, brings together so much flash, disk, and elegance, it makes an ideal platform for software-defined storage solutions such as VMware's remarkable Virtual SAN (also a winner). The R730xd shows that innovation still lives in 2U, two-socket servers.
It's fascinating what some vendors can do with commodity parts. The magic, of course, is in the software. In the Tintri VMstore, for example, the Tintri OS uses inline deduplication, compression, and other tricks to turn one part MLC flash and 10 parts SATA disk into something akin to an all-flash array for virtual machines. Lightning-fast flash delivers the IO, while disk provides the capacity.
Managing storage with Tintri might seem magical because Tintri operates exclusively in terms of virtual machines and virtual disks, rather than intermediate concepts like LUNs and volumes. After all, isn't it high time we assumed that all machines are virtual machines
You might think so ... until something better comes along -- say, like Linux containers, which offer the isolation of a virtual machine in a much lighter footprint. Think tiny, OS-less VMs with an instant-on button.
It's a rare product that earns a Technology of the Year Award before reaching version 2. But have we ever seen anything like Docker A new tool for building, running, and sharing Linux containers, Docker addresses so many pain points in application development and application lifecycle management, it would be difficult to overestimate its impact. Docker makes it easy to ensure that applications are built the same way every time and run exactly the same in development, testing, and production.
Docker helps streamline and simplify devops. Developers own what's inside the container, and operators own what's on the outside.
Docker is even changing how applications are designed (as loosely coupled containers running microservices) and how the operating system is conceived (as a minimalist runtime for containers). It's an ambitious project with big issues remaining to be solved (networking, security, orchestration), but it has broad industry support and a rapidly growing user base. It's exactly the sort of technology that the Technology of the Year Awards were made for.
After so much experimentation in 2013, it was an oddly quiet year in mobile devices, with Apple's iPhone 6 and iPhone 6 Plus the clear highlights. Apple impressed us even more with Handoff, a feature in the latest versions of OS X and iOS that allows you to begin a note or an email or another task on one device, and pick up right where you left off on another. Dubbed "liquid computing" by my InfoWorld colleague Galen Gruman, Handoff is the kind of innovation that really matters to users and an easy pick for a Technology of the Year Award.
Something old, something new
Technology of the Year veteran Hadoop has owned the big data stage for years, but the time has finally arrived for the elephant to share the spotlight. Recent versions of Hive, a companion Apache project, finally deliver on the promise of real-time, SQL-like queries of Hadoop data, moving us closer to the day when the open source big data stack can replace the traditional enterprise data warehouse.
The push toward real-time and event stream processing is also shifting attention to projects such as Apache Spark and Apache Storm. Spark offers an ingenious way to do distributed processing in memory, and it can run on Hadoop or a stand-alone cluster. Storm is a distributed stream processing system that integrates with any message queue and any database, and work is underway to integrate it with Hadoop.
So just as Hadoop has been freed from the bonds of MapReduce and batch processing, these fresh new options for distributed, real-time processing have emerged to take their place alongside it. They also take a place alongside 30 other 2015 Technology of the Year Award winners.
To learn about all 32 of our winning products, see our slide show.