In today’s fast-paced technology landscape, staying agile and having the capability to deploy applications and services quickly is essential to being competitive. In order to do this, more and more enterprises are adopting a set of software development tools and processes commonly referred to as DevOps.
One example is Lifescript, a digital media publisher that shares information on women’s health. A key component of its business entails sending up to 30 million emails a day to subscribers across six daily email newsletters. It runs complex data analytics to identify relevant, tailored information to include in these emails. In need of a framework to make sharper content and business decisions at a faster rate, the company turned to DevOps.
Lifescript is not the only business that has gone down the DevOps route to keep ahead of the pack. A Salesforce study reveals that the majority of customers expect companies to provide new products and services more frequently than ever before. This comes as 66 per cent of customers say that it also takes more for a company to impress them with new products and services than ever before. With such mounting pressures, it’s no wonder why DevOps has emerged as the key for innovation and launching new products fast.
Opting for DevOps
Although DevOps has existed for a decade now, its upturn in recent years is evident. IDC has found that the Southeast Asian DevOps software market is set to reach US$131.1 million by 2022, representing a compound annual growth rate of 29 per cent.
The rising popularity of DevOps makes sense from a technical standpoint. Conventionally, software and products are sequentially developed and launched in versions after long release cycles. Instead, businesses now build and progressively update new applications in a continuous, service-oriented process.
The Continuous Integration/Continuous Deployment (CI/CD) pipeline comprises a series of repetitive tasks, such as infrastructure provisioning, carrying out small tests and pushing code out to repositories. DevOps automates these tasks so they do not bog down the process of developing and continuously updating software. Furthermore, DevOps emphasises practices that break down barriers between developers and operations teams, so they can collaborate better for shorter development cycles and accelerated product and service delivery.
When implemented properly, DevOps can bring tremendous improvements in productivity and engagement. According to the 2018 State of DevOps report, high-performing DevOps teams are able to deploy code 46 times more frequently than lower-performing peers and have 2,556 times faster lead time from commit to deploy. These teams also have seven times lower change failure rate, pointing to the benefits that businesses stand to gain when they rapidly deliver applications and updates.
For many digital businesses today, speed is the name of the game as they seek to win customers over with fresh and innovative products. Although they are building applications more prolifically than ever before, IT teams also need to navigate more complex enterprise environments, thanks to the proliferation of virtualisation and cloud-native workloads.
From public cloud to on-premise storage, it can be challenging to standardise the environments that host applications. Developers, operations and infrastructure teams often work within their own systems, effectively locking themselves into silos. The result is that new code is often tested only at the point of production, resulting in a huge number of bugs.
These obstacles can slow down processes and product delivery considerably, even for those who have adopted DevOps.
Squashing bugs and wait times
In DevOps, deploying multiple iterative updates help contain the size of any potential errors, allowing them to be rectified sooner. Similarly, the CI/CD pipeline within a DevOps framework is often powered by virtual machines (VM) or containers. Although this isolates and distributes hardware, workloads and applications to provide a greater degree of flexibility, the infrastructure team can be slammed by a flood of storage provision and API configuration requests to support the stream of updates that developers are working on. These updates can also balloon from a few kilobytes to several gigabytes, adding to the frequency of requests.
Legacy storage platforms struggle to cope with the scale of these requests. For these reasons, all-flash data platforms have become the weapon of choice within the DevOps framework. With all-flash, the infrastructure team can reduce provisioning from hours to seconds. Developers can also speed up application testing and development by automating custom settings with APIs.
Whatever the environment for CI/CD, the right all-flash data platform should be self-running and agile enough to withstand repeated tests and troubleshooting. For example, manual failure analyses traditionally take days, but can now be slashed to seconds with near real-time analytics. It should also be flexible, providing native integration with all toolsets, whether it be Kubernetes, Docker, or OpenStack. Supporting infrastructure as code tools like Ansible, Puppet, Terraform or SaltStack, businesses can deploy faster by automating repetitive processes for continuous delivery. They can also free up developers and operations teams to focus on coding and using data to drive business growth.
Fast-forward to the next innovation
Businesses that are backed by a highly performant storage infrastructure have begun to unlock the full potential of DevOps.
Online gaming company Betfair offers a wide range of products to nearly a million customers across the world. As part of its DevOps strategy, the company deploys continuous delivery with up to 300 VMs on a daily basis, and these machines must be available on-demand to push through with code deployment. Upgrading its data infrastructure with an all-flash solution, Betfair was able to deploy stateful applications to get new products and improvements in front of customers at a rapid pace.
Speed is at the heart of DevOps. One of the goals of the methodology is to generate business value by maximising innovation and productivity across the organisation, so it follows that the framework must be supported by an all-flash data platform that enables agility and flexibility. Powered by a robust storage infrastructure, organisations can eliminate bottlenecks and inconsistencies across tools and practices. Instead, they can keep their eye on the prize: sprinting ahead to the next innovation.