2013 middleware predictions by Red Hat's Craig Muzilla. This post originally appeared on Dec. 18, 2012.

Last year, I offered my thoughts on trends and developments that the market would see in 2012. At that time, I felt that we were looking at a continued emphasis on cloud, while mobile and backend integration technologies would rise in prominence within enterprise IT. I think the industry made progress in all three areas: analysts indicate huge adoption of mobile technologies as a substitute for what would have typically been a laptop or desktop computer; at Red Hat we saw an uptick in interest around some of our backend integration tools like rules-based processing and business process management (BPM); and the industry as a whole saw a solidification around cloud visions and roadmaps from vendors.

One area that turned out to be much bigger than I had anticipated was the explosion and interest in big data. Moving forward, I think applications will be much more data-driven, using information generated by the millions of mobile devices I mentioned earlier.

To-date, the primary use of big data has been related to the analysis of large volumes of information that could not be analyzed using traditional data warehouses. I think this will continue; however, I anticipate a rise in applications wanting to participate in either big data or NoSQL solutions. I expect technologies such as Hive, which allows users to do traditional SQL queries in a big data setting, are going to become very important and popular in the coming year.

Transactional or production-related applications will need to interface with these big data solutions beyond simple analysis to take on some action or activity. Where I predict the market will likely see the most activity is around technologies like middleware that have been on the sidelines but may now begin to develop interfaces for things like MapReduce queries, Hive, or even traditional SQL queries. In this way I expect to see the more traditional application servers and tools like ESBs participate in the big data movement.

Another significant development that I anticipate for next year is the explosion of private cloud. There has been a lot of discussion lately about cloud in general, but public cloud has largely been the focus. Infrastructure-as-a-Service (IaaS) and Platform-as-a-Service (PaaS) technologies have both made a lot of progress from a public cloud standpoint, where they have primarily been used for development, testing and experimentation. I predict that PaaS will come out of this paradigm for production purposes. While very few applications are in production right now in public clouds, I think we'll see a lot of companies begin to take applications into production via private clouds.

A word of caution, though. Organizations need to shift their ways of thinking about governance models. Developers have much more responsibility and power than they did in the on-premises world. There was always a development process, and then an application went into operations, and operations took it into production. Today, we have the idea of DevOps, which is designed to empower the developer to not only work in the development phase, but then take the finished application into production and even manage it there. This model redefines governance because companies will need to think about what kind of responsibility and authority they give to staff members that may have not had those roles before. Ultimately, I think there will be a lot more flexibility, productivity and freedom, but companies will need to rethink their governance model to avoid running into problems.

In the near term, I predict the primary themes will remain the top priorities moving forward: cloud computing and PaaS, especially on-premises, will be important; OpenStack will play a big role in terms of basic cloud infrastructure for IaaS; and the industry will see continued movement in the mobile area. Today, most enterprise mobile activity has been around the client application with little backend service considerations, but I think we'll see a huge movement in iPhone or Android development, even natively, but requiring backend services such as security, transactions, enterprise integration, and persistence. As a result, I expect to see a lot of technology being developed that allows these mobile applications to fully participate in backend services and applications, and that will be a breakthrough for the industry.

In terms of big data, I hope to see a surge of technologies that seek to allow a broad set of applications, both analytical applications and transactional applications, become more involved. Whether this is using data in-memory, such as NoSQL solutions, or data management solutions like Hadoop with virtualized storage, I predict that it will ultimately lead to more data-driven applications. The foundation for this is already in place with MapReduce and Hadoop, and NoSQL solutions like Mongo and our own Red Hat JBoss Data Grid, but I think the interfaces required for participation from a broad set of applications still need to be built out over the next few years. Companies will look to NoSQL and big data solutions as a replacement for traditional relational databases or to reduce the dependency on them, and they will use them for a variety of transaction and analytical apps. I expect middleware to evolve to support this.

Long term, I think the nature of an enterprise application will change, especially as it relates to creating higher levels of abstraction, allowing non-technologists and business users to participate in the creation of those applications and mobile applications. Over the next two to five years, business users, consumers and partners will use tools like BPM and rules management that offer this level of abstraction to develop and adjust applications based on needs. There is some discussion regarding intelligent BPM (iBPM) now, but it still seems to be largely under the radar. The industry will likely hear much more discussion about these sophisticated and end user-friendly technologies, both in the cloud and in traditional uses, in the years to come.

Undoubtedly, the future holds a number of surprises for the enterprise IT industry. While my thoughts and observations here are based on some existing trends, all of these things are driving innovation. What's more, cloud, mobile and big data are all driving the need for innovation. Small pieces of innovation are happening every day in communities at the developer level. It's exciting to see and be a part of these innovations as they emerge, take hold in the enterprise and make their mark on the world.