WavebreakMediaMicro - Fotolia

News Stay informed about the latest enterprise technology news and product updates.

Cloud computing technology trends in 2015

As the IT world looks ahead to 2015, which new technologies will take off? Our cloud experts predict the top cloud technology in the New Year.

As the IT world bids adieu to 2014, enterprises and IT pros are left wondering which direction the industry will go next. Last year, our experts predicted big things for big data, a stronger emphasis on security and a movement toward platform services, among other insights. And for the most part, their predictions were right on the money.

The need to house big data certainly didn't decrease and developers continued to try to find new ways to meet demands. Cloud security was a hot-button issue in 2014, as hackers claimed major retail breaches at Target and Home Depot, the iCloud hack and, most recently, the Sony Pictures' attack.

This year, we asked our experts the following question: Which new technologies will take off in 2015 or challenge the status quo? Here's what to expect.

Paul Korzeniowski

I don't expect there to be a lot of new technology, rather a shift in emphasis. Movement to vertical applications will become more common. Companies are looking to move daily business services to the cloud. That change is only possible if they can tailor cloud services to their own operations. Consequently, interest in platform as a service (PaaS) will grow. But finding cloud application developers will become an issue for companies hiring in 2015. Disaster recovery will be another area of keen interest. Traditionally, such systems have been a problem for IT. But cloud enables them to address many previous problems -- testing DR systems, combating the high cost of installing a backup system and accurately mimicking potential disasters.

David S. Linthicum

The most significant developments will be The use of containers around application distribution, cost effectiveness and portability. There will be a renewed focus on automation tools that deal with containers, including those that provide automated app or component portability based on pre-defined criteria.

The value will be the ability to reduce complexity by using container abstractions. Containers remove dependencies on the underlying infrastructure services, which reduces the complexity of dealing with those platforms.

Tom Nolle

Most of the changes in 2015 will be driven by simple business metrics. Cloud margins continue to fall and enterprises have yet to develop cloud-specific applications. Cloud vendors such as Microsoft, HP, IBM and Oracle can help enterprises evolve and provide hosted services to facilitate cloud adoption. That will make them feature players. Telcos and others will have to work to drive down costs, as smaller players will be squeezed out.

Jim O'Reilly

Cloud server technologies to watch in 2015 include fatter servers that use Haswell CPUs and allow larger instances or more virtual machines (VMs) per server; in-memory database instances with as much as 2 TB DRAM and remote direct access memory (RDMA) over Ethernet.

Networking technology will focus on software-defined networks (SDN), which will be the norm with automated and policy driven network management. SDN's rapid evolution will continue through the end of 2015 with new features and performance records. RDMA will become a sandbox item for many, with some large cloud service providers actively exploring deployment.

There also are big changes coming to accommodate big data. Interest in the Internet of Things will build throughout 2015, so positioning clouds and apps for it is difficult. We are moving up from Web-server-sized instances to emulating the best private cluster gear. Everything from orchestration to database tools will need to evolve.

Dan Sullivan

AWS continues to roll out new services at a steady pace. Lambda, in which code is triggered to run in response to an event, is an important step in the abstraction of computing models away from server-centric design. With Lambda, the cloud is becoming a generic compute engine; developers don't need to manage instances -- they just run code.

We will also see more interest in the Go programming language. Large-scale, high-efficiency applications will benefit from Go's performance advantages. Spark will largely replace MapReduce as the go-to model for big data analytics. MapReduce will still be widely used, but developers will choose Spark over MapReduce whenever possible.

Next Steps

What to expect from the cloud market in 2015

Three cloud mistakes to avoid in 2015

Who suffered the most cloud outages in 2014?

Planning for the cost of cloud computing

Dig Deeper on Cloud application development