Best Practices for DevOps in the Cloud

February 17, 2016

  Despite being the latest buzzword in IT, the term DevOps still raises a lot of question marks any time it’s brought up. Simply put, DevOps is the combination of tasks performed by an organization’s systems operations, development and QA engineering teams in the entire service lifecycle, from design through the development process to production […]

DevOps Best Practices

 

Despite being the latest buzzword in IT, the term DevOps still raises a lot of question marks any time it’s brought up. Simply put, DevOps is the combination of tasks performed by an organization’s systems operations, development and QA engineering teams in the entire service lifecycle, from design through the development process to production support. However, DevOps is considered by many as more of a belief or cultural approach that aims to foster improved communication between development and operations teams as more elements of operations become programmable. DevOps has strong similitude with Lean and agile approaches. The need to break down the barrier between operations and development has been accelerated by cloud computing. DevOps and cloud computing are mutually reinforcing strategies for delivering business value through technology.

At the turn the century, enterprises began shifting their focus from efficiency and stability towards innovation and agility. In order to adapt to the changing face of the business market and increase delivery frequency, application delivery teams have to adopt concepts like experimentation, rapid iteration, collaboration, and Minimum Viable Product (MVP) deployment. DevOps successfully bridges this gap. A DevOps approach applies lean and agile thinking principles to all the stake holders who develop, operate or simply benefit from a company’s software systems, this includes partners, suppliers and customers. Cloud computing, whether on –premise or purchased as a service combines infrastructure, services and software to help organizations develop and deliver quality software at a much faster rate. The elastic properties of the cloud expedite scalability while DevOps streamlines and accelerates application releases; this is why the marriage of The Cloud and DevOps is the perfect partnership.

Best Practices for DevOps in the Cloud

DevOps practices and principles form the foundation that enables enterprises to fully utilize cloud-based computing and to address and mitigate the inherent risks associated with the cloud. Companies that are capable of reliably building their infrastructure, provision servers and deploy apps are in a better position to handle any challenge the cloud throws at them. However, IT professionals who practice DevOps in the cloud typically make mistakes due to a rudimentary understanding of the best practices and various deployment technologies.

A survey of 600 IT professionals conducted by Forrester on DevOps practices,  and where enterprises are in terms of the maturity of DevOps practices adoption revealed that roughly 33 percent of teams consistently deliver at cycles of one to three weeks and that the fastest teams generated higher business satisfaction than slower teams; a clear indication that quality is not sacrificed for fast delivery if the proper practices are implemented.

DevOps Team Assemble!

DevOps places a strong emphasis on the collaboration between development and operations. Assembling a team of developers who have more interpersonal, operational and communication skills than a regular head down developer is the best way to break down organizational silos and build a more agile approach to application development and deployment. Developers are responsible for selecting and implementing new technologies and features, and they should be able to quickly respond to, and address any issues that arise within existing systems. The operations team contributes the important expertise of how the technology behaves under live production conditions. If development and operation functions are separated, active collaboration is limited, leading to applications problems that subsequently delay deployment.

The DevOps scope is much larger than the operations and development teams. It also has to include other stakeholders from the organization and the service provider. In order to realize the success of the enterprise through DevOps, the key stakeholders within the organization need to participate in cloud and DevOps training.

Automated Performance Testing

In cloud deployments, application performance issues are typically a result of flawed application design. Most of these performance issues are missed and end up going into production where users eventually find them, which isn’t good. Performance testing is a crucial aspect that should never be overlooked by the DevOps stream. The development team should adopt automated regression testing as a common practice, and ideally extend it to test-first approaches like behavior-driven development (BDD) and test-driven development (TDD). This guarantees that the operations team receives a solution of sufficient quality before it’s approved for release into production. Shifting away from manual testing improves quality, delivery speed, and testing accuracy, thus dramatically reducing cost. Automated testing should combine existing accuracy and stability testing, as well as with existing testing for user interfaces and APIs.

Incorporate Containers into the Cloud Strategy

The easily manageable and portable nature of containers makes their integration one of the best practices for DevOps in the cloud. Containers provide a way to ‘componentize’ applications, simplifying every step from development to deployment. However, it is prudent to consider cluster management, governance, security and orchestration tools for applications that leverage containers.

Continuous Integration and Deployment

Continuous deployment and integration are effective techniques used in DevOps to eliminate unnecessary steps, delays and friction between steps to increase work flow. Cloud-based development can greatly benefit from automating deployments and frequently integrating changes. Continuous integration allows developers to safely create high-quality solutions in small, regular steps by providing immediate feedback on code defects; while continuous deployment allows them to minimize the time between a new feature being identified and being deployed into production. Continuous deployment and integration may increase operational risk if the development teams are not properly disciplined. For a continuous app delivery model to succeed, a strong management system must be put in place.

For the Cloud, by the Cloud

In order to take full advantage of the cloud, including platform as a service (PaaS) and infrastructure as service (IaaS), the applications have to be designed in a way that they are decoupled from physical resources. This is where the term ‘Infrastructure as code’ or ‘programmable infrastructure’ comes in play. DevOps places a strong emphasis on the ability to build and maintain essential infrastructure components with automated, programmatic features. So from a DevOps perspective, infrastructure as code (IAC) includes the ability to build middleware, provision servers, and install application code that makes up the core components of the system architecture. The use of loose architectural coupling within and between applications greatly reduces complexity and enables delivery in small increments. Considering the decoupled architecture in your design improves the overall utilization and efficiency of the cloud resources by up to 70 percent. Cloud computing subsequently helps in saving money and you only end up paying for the resources you use.

 

Author: Gabriel Lando

By Team FileCloud