August 19, 2017

News media moves legacy backend systems off the back burner in a bad way.

The online news media often highlights the bad penny that is legacy backend systems in a bad way. Stories relate how the legacy backend’s development lifecycle is too slow to keep up with that of the modern app that enterprises are rapidly releasing software (weekly, daily, even many times a day) using DevOps.

If you build a new mobile app that needs data from your traditional ERP system or some complex legacy system that’s been around a long time, it will be difficult to build new abilities for that app on top of that legacy system if you only update it once a year or once a quarter, says Brad Schick, CTO and VP of Engineering, Skytap.

“Some enterprises building newer, more modern apps that are small and nimble, web-focused or mobile-focused, and run on Amazon or Azure  platforms are creating them in a vacuum,” says Schick. Though these apps will have dependencies on some kind of system, they may not be prepared to count on the legacy backend system, which also may not be ready to support these apps or to grow with them.

Enterprises seeking to innovate at the speed of DevOps find themselves and their shiny new apps constrained by how fast they can move these backend systems while conversely companies with fewer legacy or traditional technology investments are able to move faster, says Schick. This is how the former companies fall behind their competition.

Good News on the Horizon?

To meet this challenge, organizations are trying to move towards DevOps and Agile software development in the traditional application, towards breaking up monolithic applications into microservices, and / or towards integrating modern technologies like containers or cloud services that help speed up software development, according to Schick.

One cloud company is targeting the problem using containers and the cloud. “We provide virtual environments that are essentially virtual data centers that mimic what the enterprise can do on premise,” says Skytap’s Schick. Skytap’s as-a-Service model supports containers and copying environments to and within the cloud to create a parallelized sandbox for faster experimentation, collaboration, and development of backend systems using a quick and nimble infrastructure.

Instead of having to develop their current production backend systems inside the limiting walls of available on premise technology resources, enterprises can add, say, four copies of their working systems in the cloud and count on scalability to give them room (capacity) to develop them. As you know, parallelizing work saves time, and this speeds the backend development process. At least this is Skytap’s goal.

Other Skytap clients use the cloud to containerize services for the first time. These are the folks seeking to turn monolithic software blobs into multiple, agile microservices.

Container Role Case Study

“We have a customer that’s running a large custom-built application that’s got an Oracle www.oracle.com RAC (Real Application Cluster) backend,” says Schick. Skytap enables the customer to bring the large traditional database cluster application in the cloud where it will work just as it did on-site. “The failover cluster mode will behave as expected,” says Schick.

This backend supports a large monolithic application that in turn serves the customer’s entire website. Skytap enables the customer to make copies of this humongous app so that every developer can have and work on their own copy simultaneously with other developer work, in parallel.

Skytap also helps them to isolate, iterate, and improve individual functions such as the shopping cart function and then to rebuild that into a standalone service that can run inside its own container. “They can redesign and re-architect the application, and communicate with components inside containers over a message bus,” says Schick.

A Word of Caution

Despite the obvious advantages that appear on the face of the Skytap offerings, it seems risky putting multiple copies of what could be your entire store of valuable data into the cloud. Do your due diligence, collect ample proofs of cloud security, and then make a decision.

 

David Geer

David Geer’s work has appeared in ScientificAmerican, The Economist Technology Quarterly, CSO & CSOonline, FierceMarkets, TechTarget, InformationWeek, Computerworld, Byte.com, ITWorld.com, IEEE Computer Society’s Computer magazine, IEEE Distributed Systems Online, Government Security News, Laptop, Smart Computing, Technical Support, The Hosting Standard (Canada), TechWorld.com (UK), SIGnature, Processor, and the Engineering News-Record. David served as a technician at CoreComm in Cleveland, OH prior venturing into writing.