Welcome!

Cloud, Big Data, and the Internet of Things

Automic Blog

Subscribe to Automic Blog: eMailAlertsEmail Alerts
Get Automic Blog via: homepageHomepage mobileMobile rssRSS facebookFacebook twitterTwitter linkedinLinkedIn


Related Topics: Application Performance Management (APM), DevOps Journal

Blog Feed Post

How to Smooth the Software Delivery Process By @RobinBloor | @DevOpsSummit #DevOps

Why DevOps automation is your runway from software development to bullet-proof operational deployments

How to Smooth the Software Delivery Process
By Robin Bloor

The path of an application from software development to bullet-proof operational deployment has rarely been a smooth one. This is true whether we're talking new applications or maintenance releases of established applications. There's no deep mystery to why this is. The fact is that the software development environment has always been a "moving target."

Consider the history. Time travel back to a world where centralized software engineering teams were creating monolithic applications that married C to Java, C++ and other development toolkits. They were disrupted by the advent of Service Oriented Architecture (SOA) and Web Services as organizations realized their IT systems were business assets that needed to be leveraged across the Internet. And this world was disrupted itself by the advent of virtualization, Software-as-a-Service (SaaS) and Cloud enabling IT not only to think about where they developed new apps and systems but also extra options when sourcing and consuming compute power.

Spare a thought for the poor software developers that had to grapple with this fast changing Internet-centric world, where content (what you could see on the web page) and code (how you programmed it) took part in a complex and guilty embrace, which gave birth to fortnightly web maintenance releases.

Let's also note that, as time marches on, the world of software development is continuing to be transformed - mostly recently by the advancing of DevOps - in an era where billions of mobile devices are online 24/7, the sudden emergence of Hadoop and the nascent world of the Internet of Things, which is currently putting its shoes on.

An Incomplete Inventory

By the way, this is not a complete inventory of the various disruptions that have worked away at the corporate software development environment. The complexity increases when you add in different software testing environments, different programming languages, different job control and scripting environments, different databases and so on.

And all of that would not be so difficult to manage were it not for the fact that we moved from stand-alone applications to client server, to three-tier and then multi-tier software arrangements. The subsequent complicating factor, in that respect, was the move to SOA which opened up almost every application interface and thus multiplied the complexity of software connectivity and effectiveness of software testing.

It is fair to say that over the space of about a decade, we moved from silo systems to increasingly connected systems and that put extra obstacles on the DevOps path, that runs as it always ran from software development to operational deployment.

Because of the various consultancy work I get involved in I get to visit a variety of different IT user sites. While it is true that some sites are a little chaotic, the general rule is that the operational side of the IT equation is usually well buttoned down in respect of operational discipline and the best practices that go with it. There are well understood rules and procedures that are firmly implemented in an effort to prevent the disruption that inevitably occurs if anything goes amiss when an application passes from development to operations. The cost of this is a good deal of manual effort and checking, on top of whatever automation exists for this process.

DevOps Automation

The case for DevOps automation of the kind that can be achieved through the enlightened use of Automic's release automation capabilities is easy enough to make. Yes, you will save labor. Yes there will be fewer errors and less wasted time or wasted resources. Yes, there will be an increased capability to provide bespoke capability in particular environments - for example in the area of data center to cloud integration. That's fine. It is the kind of thing that Automic is good at. But for the full impact of what can be achieved, and maybe what should be achieved, it may be a good idea to review the whole DevOps activity and design, and then implement, a global solution for it. The point is this...

Consider a fairly simple DevOps environment. Let's say that your business started its computing life in the PC client-server era. Maybe you built your first applications using Microsoft tools and deployed applications from the desktop to the server. Maybe (in fact almost certainly) you implemented software packages. Your company and its IT department stayed true to that one environment for as long as it could and thus the DevOps situation was stable and uncomplicated.  But as the company grew, the environment became more diverse. A database project mandated a Unix server and Linux crept into the picture and, when Amazon's AWS emerged it was quickly taken advantage of, and then it became necessary to build mobile applications, and suddenly the whole Hadoop ecosystem began to appeal for particular Big Data and BI applications.

Imagine if you had had a global DevOps environment from the get go, with automation from Automic embedded. Any change to the DevOps environment would mandate a change to the global DevOps system. But that system would be created using only one "development capability" and that capability would prove constant across different operating environments, across different development environments, across different testing regimes, across different virtualization technology and across time. It could and would become the foundation on which a successful DevOps initiative was built.

So if you're an Automic user, take a look at how you manage the DevOps environment. Effective though it might be, it's possible that it could become even smoother.

Read the original blog entry...

More Stories By Automic Blog

Automic, a leader in business automation, helps enterprises drive competitive advantage by automating their IT factory - from on-premise to the Cloud, Big Data and the Internet of Things.

With offices across North America, Europe and Asia-Pacific, Automic powers over 2,600 customers including Bosch, PSA, BT, Carphone Warehouse, Deutsche Post, Societe Generale, TUI and Swisscom. The company is privately held by EQT. More information can be found at www.automic.com.