Welcome!

Java Authors: Dana Gardner, Elizabeth White, Carmen Gonzalez, Ivan Antsipau, Esmeralda Swartz

Related Topics: Cloud Expo, Java, SOA & WOA, Linux, Big Data Journal, SDN Journal

Cloud Expo: Blog Feed Post

@CloudExpo | Embrace #DBaaS to Speed Up Mobile App Development

The question for a company isn’t if they need to deploy mobile applications, but how to do it most effectively

By

As enterprises work to rapidly embrace the mobile revolution, both for their workforce and to engage more deeply with their customers, the pressure is on for IT to support the tools needed by their application developers. Mobile application developers are working with a massive variety of technologies and platforms, but one trend that stands out is the rapid adoption of NoSQL database engines and the use of Database-as-a-Service (DBaaS) platforms and services to run them.

Gartner has predicted that by 2017, 20% of enterprises will have their own internal mobile app store, meaning that enterprises are deploying both commercial and custom applications to their workforce at increasing speeds. There’s no denying the massive growth in mobile applications within the enterprise.

As for consumers, Portio Research claims that the worldwide mobile app user-base stood at almost 1.2 billion at the end of 2012 (and forecasts a nearly 30% CAGR through 2017). That growth is driving a many enterprises to put customer mobile experience front and center in their marketing and customer service plans. Users are now expecting to interact with a company through their mobile devices.

The question for a company isn’t if they need to deploy mobile applications, but how to do it most effectively. An on-premises Database-as-a-Service (DBaaS) platform, that supports a variety of NoSQL database technologies, can help your enterprise quickly respond to the challenge.

Mobile Causes Significant Issues for Data Services
Enterprise applications are largely about data: accessing data from anywhere and performing transactions that effect data. This is true for both consumer applications and workforce applications. While many of the lessons we have learned from our web-based systems apply to mobile apps, there are some interesting new challenges in the way that we manage our data.

Latency – First and foremost, mobile apps need to be designed for highly latent network environments. While some interactions may be over wifi, its very likely that your users will end up trying to complete some critical transaction at a time when their device is either struggling to get a cellular signal or dealing with traffic congestion on the carrier’s network.

Scaling – Next is the problem of scale. Mobile applications, especially customer-focused, are hopefully being deployed to increase customer engagement with your company. This increase in usage leads to increased demands on your backend systems. You need to be prepared for this by choosing architectures that will be able to rapidly scale as use grows. This is true up and down the stack, but particularly true for your database layer.

Agility – Successful mobile experiences require rapid feedback loops. The applications need to evolve as you receive feedback from your users, taking the form of both enhancements and whole new features. With that demand for agility comes a significant concern about how to handle rapid data model changes.

Multiple App Versions – If a mobile application is successful, mobile app developers quickly find that they are required to support multiple versions of the application. This is where the demands placed on your data model due to the need for agility are compounded. Especially for customers, it’s simply not acceptable to frequently drop support for the previous version(s) of your app. Your data management strategy needs to account for this multi-model reality from the start.

The Rationale for NoSQL Databases in a Mobile App World

Relational databases remain the most used database type within the enterprise today, but mobile application developers are frequently selecting a document-centric NoSQL database engine, such as MongoDB and Couchbase, to be the heart of their application’s backend infrastructure.

The reasons for this frequent choice is due to the architectural traits of these database systems. While there are always many ways to engineer around a particular problem, and each NoSQL database engine certainly has its own pros and cons to consider, some of the features common to many NoSQL database engines directly address the issues that mobile application developers are facing:

  1. Document-centric data access can reduce the effects of latency by reducing the time it takes for the back-end system to return a complex data structure
  2. The horizontally scaling architectures of these database engines can easily grow to match user demand
  3. Embracing lightweight data representations like JSON documents can make app dev agility much easier, especially if the document format can be easily consumed on a variety of platforms (different mobile development frameworks and even for the company’s standard website)
  4. Schema-less database design makes supporting multiple versions of your applications easier for the developers

The Operational Perspective

After a development team has made the choice of their preferred database engine, the next question for an organization is how they will operationally support this new data service. Operations teams are being challenged to quickly implement the required database environment in a way that will both meet the needs of the application development team and meets their own need to ensuring that the production environment lives up to the promise the company is making to it’s users.

Although these challenges are shared by non-mobile systems, mobile applications force operations teams to confront the following issues:

Operationalization of New Technologies – Many corporate IT departments simply don’t have experience in their DBA organization with NoSQL database engines, and the explosion in the number of applications they are supporting is continuing to grow.

Data Locality – The operational perspective for mobile applications needs to always consider questions of data locality. The deployment architecture needs to ensure that the data is in a secure location and that the application’s server-side systems are as close to the data storage as possible to help eliminate as much back-end latency as possible.

Resiliency and Availability – If mobile is about increasing user engagement, then there is an implicit promise being made to those users that the service will always be there when they need it. Operations teams need to ensure that they are able to achieve SLAs far greater than the typical internal system has to achieve.

How On-Premises (or Hybrid) DBaaS Helps

How does Database-as-a-Service (DBaaS) fit into all of this? DBaaS solves a number of challenges for both the operations team and the application developers. Further, a DBaaS platform that supports multiple database technologies and choice of infrastructure deployment targets gives both groups significant benefits.

A multi-Cloud, multi-engine DBaaS platform can provide:

Infrastructure Choice – Both the latency and data locality considerations can be addressed by a sufficiently robust DBaaS platform, by allowing application owners (devs and / or ops) to deploy database services into whatever infrastructure is most appropriate for that specific service. For example, developers could deploy dev/test systems into a public cloud environment, while the production version of the application can be deployed into a private cloud (or even bare-metal server farm).

Database Engine Choice – Giving application development teams a chance to select from multiple database technologies doesn’t have to be a point of stress for the operations team. DBaaS platforms enable this choice, while giving the operations team an automated provisioning and management tool that provides a consistent high quality operational experience for the diverse environment.

New Technology Operationalization – As new database technologies rise to prominence, and are embraced by developers, operations teams should be in a position to bring these new technologies into their environment easily. By using a DBaaS platform, the operations team gets the benefit of reducing their training burden for each new technology, and the developers benefit from getting access to new technologies faster than could be otherwise achieved.

Support for both NoSQL and Relational Databases – While many mobile applications will be built on top of NoSQL database engines, some will be built on relational databases. DBaaS can allow the organization to support both types of data services.

Enable Agility – DBaaS provides some powerful automation functions that can make life much easier for an application development and operations team. By aligning the database management processes with the application delivery model, teams will reduce friction in the app delivery pipeline.

More Stories By Bob Gourley

Bob Gourley, former CTO of the Defense Intelligence Agency (DIA), is Founder and CTO of Crucial Point LLC, a technology research and advisory firm providing fact based technology reviews in support of venture capital, private equity and emerging technology firms. He has extensive industry experience in intelligence and security and was awarded an intelligence community meritorious achievement award by AFCEA in 2008, and has also been recognized as an Infoworld Top 25 CTO and as one of the most fascinating communicators in Government IT by GovFresh.