This week, most of my colleagues will be in San Francisco attending Dockercon 2015, joining everyone else consumed with building excellent cloud applications for the latest and greatest announcements, demos, and hands-on workshops from throughout the Docker and container ecosystem. As I am back on the East Coast, I didn’t want to feel left out of the excitement, so I’m going to talk about some of what we’re doing with containers in some of our client engagements and reference architectures, building on Bluemix and IBM Containers.
Mobile. Big Data & Analytics. Internet of Things. Cognitive. Hybrid. Integrated. Secured.
The majority of Cloud applications today aren’t just one “thing”. They are many, or have a roadmap to get to many “things”…
These are all the capabilities in new and existing applications that our clients are looking for today. These also happen to be all the capabilities that IBM Bluemix brings to market today. I want to cover two of the larger architecture styles in this post, some of which have the most overlap on the Venn Diagram of “Where are we now versus Where do we want to be” for our Enterprise clients — Mobile and Big Data. This overlap is made possible by the hybrid nature of containers and the IBM Container service on Bluemix.
The true power of Bluemix, IBM’s cloud platform, comes in its flexibility and applicability to a broad number of critical use cases for today’s Enterprise, with a similar experience for the user in both new and existing applications, which we’ll cover across the two different architectural styles.
Today’s Mobile applications are much, much more than just an entry in the Apple App Store or the Google Play Store. Many estimate that 70-80% of the effort of building a mobile app is not in developing the front end, but in all of the back-end integration, data storage, security, etc. At a minimum, you need a mobile application server or gateway for secure and managed engagement coupled with services for universal push, location, and data capabilities. In addition, API Management enables access to data in a reusable and scalable way. The entire Enterprise-grade Mobile architecture is documented here and can seem quite daunting at first.
IBM recognizes the increasing challenges our clients have to more quickly and easily build, deploy, and manage mobile applications. To help clients address these challenges, IBM intends to offer the MobileFirst Platform Foundation available for deployment on IBM Containers on Bluemix. IBM also plans to offer an evaluation of the MobileFirst Platform Foundation available for deployment on IBM Containers on Bluemix to help developers, IT professionals, and lines of business assess the offering.
IBM’s statements regarding its plans, directions, and intent are subject to change or withdrawal without notice at IBM’s sole discretion. Information regarding potential future products is intended to outline our general product direction and it should not be relied on in making a purchasing decision. The information mentioned regarding potential future products is not a commitment, promise, or legal obligation to deliver any material, code, or functionality. Information about potential future products may not be incorporated into any contract. The development, release, and timing of any future features or functionality described for our products remain at our sole discretion.
IBM MobileFirst Platform provides many of the core capabilities needed to quickly and efficiently build, manage, and secure your mobile app projects. Many of the functions of MobileFirst Platform are available as discrete services in Bluemix and on-premises software; however, core enterprise functions for enterprise security, integration and management were only available for on-premises or in VMs that could take special expertise to configure properly. The power of containers enables MobileFirst Platform customers to evaluate the MobileFirst Platform capability with the flexibility to move on-premises, if the business situation requires it. At Dockercon, we intend to demo an early preview of our full MobileFirst Platform Foundation running in the IBM Container service. The same exact code that runs in your data center, the same exact Apps & Adapters you’ve built, the same exact APIs you’re calling, all running in Bluemix. How easy is that?!
The same exact experience your Enterprise developers have been delivering your market-cornering Mobile applications to in recent years is available on Bluemix. And when your needs expand, Bluemix has you covered as well — integration with Cloudant, IBM’s NoSQL DB service, for offline data sync is as simple as
docker run to connect your container instance to Bluemix services, as well as integration with your current Systems of Record… which brings me to Big Data.
Big Data & Analytics
The problem with Big Data is it’s big… and the data is everywhere. To solve this problem, you need a truly hybrid application to access your data anywhere, anytime. There are two concepts in Big Data – taking the data to the compute (the most obvious) and taking the compute to the data (the not so obvious). Starting with taking your data to the compute seems pretty logical, since you have a centralized runtime, your killer app on Bluemix naturally, and you just need to collate your data from many different organizations inside your Enterprise.
To solve this problem, Bluemix provides a Secure Gateway capability to securely tunnel from Bluemix and connect to your Systems of Records, that you specify and control. Bluemix spits out a Docker command for you to run locally, which connects your on-premise System of Record to your Bluemix application via a Secure Gateway client running inside a Docker container. This gateway then provides you with a TLS-secured cloud host and port number for you to access your on-premise data from your application running in Bluemix. The power of the Secure Gateway, Docker, and containers makes it so easy to securely connect your hybrid application to your datacenters. We’ve just published a hands-on workshop that will walk you through this exact scenario!
But we’re not out of the woods yet. Why would you think about moving the compute to the data if the Bluemix Secure Gateway is so simple? Gigabytes of data are fine to transfer, as well as smaller amounts of Terabytes. But what about if you have Petabytes or Exabytes? You’re not going to want to spend the money, let alone time, to put that data on a wire and move it into your applications running on Bluemix.
I’d hazard a guess at saying you’re pretty efficient and comfortable with managing your data at that scale. However, there’s some hot new compute capabilities that other teams are leveraging in Bluemix and you don’t want to miss out. Easy enough… take the same application you were running in Bluemix, and move it into your datacenter — bringing the compute to the data. This is made possible by through the IBM Container service and the true hybrid portability of containers. Same application, same code, faster insights. How can you complain about that?