Three Ways Enterprise Software is Changing Today's IT shops must grapple with analysis, cloud computing and DevOps
By Joab Jackson (IDG News Service) on 20 December, 2014 http://www.arnnet.com.au/article/562900/three-ways-enterprise-software-changing/?fp=2&fpid=1 Once upon a time, life in the enterprise IT shop was fairly simple, at least conceptually speaking. IT issued computers and laptops to employees, and maintained enterprise software, databases and servers that supported the company, which were mostly run in-house. These days, IT's basic firmament is giving way to a more breathtaking geography that the IT pro must traverse, based on pay-as-you-go cloud computing, building applications and performing deep data analysis. Perhaps more fundamentally, IT operations are moving from merely supporting the business to driving the business itself, which requires agility and making the most of resources. Here are three of the largest forces at work that will change enterprise software in 2015, and beyond: The Platform The idea of cloud computing has been around for a while, so it may be hard to think of it as a new force. Yet, after a few years of testing the cloud for running development projects and tangential applications, enterprises are now moving their more critical operations to the cloud. IDC expects that by 2017 organizations will spend 53.7 percent of their budgets on cloud computing, and the market for cloud computing software will be over $75 billion. Concerns about security and overall cost continue to fade as businesses face the upgrade costs of replacing data centers full of servers, or stare down the large up-front costs of implementing a complex in-house enterprise software system. Travel information provider Lonely Planet is one Web-facing company that has made the jump into cloud services, migrating all of its operations to Amazon hosted services when its data-center lease came to an end. "With Amazon, we could treat infrastructure as code," said Darragh Kennedy, head of cloud operations for Lonely Planet. Instead of worrying about how many servers to lease, the company could concentrate on perfecting its service, with Amazon quickly and easily providing however many servers are needed for seamless service. "Our product owners can stand up a new environment in under 10 minutes, and that really speeds up how quickly we can build new products," Kennedy said. Amazon Web Services got a head start in the cloud services space, but Microsoft is quickly catching up with its Azure service, according to Gartner. Other enterprise-focused IT companies quickly ramped up their cloud-computing operations this year. IBM and Hewlett-Packard, have each earmarked $1 billion to building out their cloud-computing services. Complicating the best laid cloud migration plans has been the sudden emergence of Docker, a new, lighterweight, form of virtualization that promises greater portability and faster performance. Launched in 2013, Docker has been downloaded over 70 million times. The major cloud service providers, including Google, IBM and Microsoft, all spun up their own, and sometimes proprietary, Docker-based services. While those CIOs who have already started down the path of cloud computing, perhaps by virtualizing some of their operations, may feel frustration at the potential of re-gearing with Docker, it provides one key element that they will need: swiftness. It has been said Docker is the first virtualization technology ready for the DevOps age. What is DevOps? You should know about that as well. The Software A decade ago, COTS (commercial-off-the-shelf) software was the way to go. Why go through the trouble of building your own software from scratch when Oracle, Microsoft and SAP could provide you with all (or at least most) of the capabilities? If employees grumbled about such software being sometimes difficult to use, well, they were getting paid to use it, right? These days, however, businesses are finding that enterprise software is no longer in a supporting role, but is central to businesses maintaining a competitive edge. In many cases, this means the organization must build its own software, at least for those parts of the operation that provide the crucial competitive edge for the company. Remaining competitive is a moving target, of course, as competitors are also busy sharpening their own products and services. Nowhere is this more pronounced than with large Internet-scale services such as Yelp, Facebook, or AirBnB, who live or die on beating their competition with more helpful, and easier to use, features. The days of asking users, or employees, to put up with fussy software are coming to an end. Such pressure has brought about a new operating paradigm called DevOps, which, in name and in spirit, combines software development and IT operations into one cohesive workflow. Tightly integrating the development cycle of an application with the subsequent operation of that application can cut the length of time required to update a customer-facing or internal application. About 60 percent of CIOs plan to use DevOps to manage their software, IDC has estimated. Microsoft has been filling out its portfolio of development software to support devops operations. IBM has set up a special consulting practice just for helping organizations get more into a devops-style workflow. One user of Microsoft's DevOps tools has been the business services division of French telecommunications company Orange, which develops systems and software for other organizations. "A few years ago, it was the norm to deliver good functionality on time and on budget," said Philippe Ensarguet, CTO at Orange Business Services. "Now, we have to deliver sooner and faster and better." One question that dogs the modern business is how to offer something unique in this global, hyper-competitive market. This is where new forms of data analysis could help. The Data Data analysis, once chiefly the provider of numbers for PowerPoint presentations and executive dashboards, is increasingly shaping the strategies and operations for many organizations. Of course, data-guided business decisions are nothing new. What is new is a new depth in the kind of insights that analysis can provide, as well a greater range of data that can be put to computerized scrutiny. IBM, among other companies, has been ambitiously pursuing the additional ways data can be parsed through cognitive computing, which harnesses techniques of machine learning, neural networks and other approaches to better mimic the ways humans intuit insight from data. And thanks to the open-source Hadoop data-processing platform, the use of which is growing in the enterprise, additional types of data can be mined for potential knowledge. Hadoop excels at churning through vast reams of unstructured data, data not stored in a relational database but captured in text files or log files--all the stuff IT staff used to largely ignore, then routinely delete once it filled its coffers. But e-mail, the Web surfing habits of customers or server log files can provide insight into long term trends, daily operations or heretofore undiscovered customer preferences. One such company that found a competitive edge with such big data, as it is often called, has been enterprise security services company Solutionary, which used a MapR-based Hadoop distribution to enlarge the set of services it offers for its customers. Solutionary uses Hadoop to store and analyze the security and events logs of its corporate customers, so they can be alerted when suspicious activity may be taking place on their systems. Hadoop allows the company to store more data, at a cost considerably less than if it were to be stored on a data warehouse. Using this additional data allows the company to offer a longer-view analysis to its customers about what is happening on their networks. It also allows them to perform predictive modeling on the data, potentially giving its customers earlier warning about security issues. Hadoop allowed Solutionary "to get off an architecture where you had to be careful about what to put into it, and to a model where you could store everything," said Scott Russmann, Solutionary's director of software engineering. -- Cheers, Stephen _______________________________________________ Link mailing list [email protected] http://mailman.anu.edu.au/mailman/listinfo/link
