This blog series explains how to engineer applications for DevOps.
Topics covered in this blog series:
• Factors used to decide whether an application is a good candidate for DevOps.
• Practices to engineer designs for DevOps.
• DevOps applied to enterprise apps and software services.
• DevOps applied to COTS systems.
• DevOps applied to manufactured (embedded) systems software.
• DevOps applied to software services.
• Five levels of application maturity.
This blog covers the topic of DevOps applied to enterprise apps and software services; part three in the series. You can read part one here and part two here.
DevOps Applied to Enterprise Apps
An enterprise application assists an organization in solving business problems. Enterprise applications are typically designed to interface or integrate with other enterprise applications used within the organization and to be deployed across a variety of networks (internet, intranet and corporate networks) while meeting strict requirements for security and administration management.
Proprietary enterprise applications are usually designed and deployed in-house by specialized IT development teams within the organization. However, an enterprise may outsource some or all the development of the application and bring it back in-house for deployment.
The size and complexity of large enterprise applications, especially older monolithic-architected applications that are not designed in modules, are a challenge when it comes to implementing DevOps. Given the large scale and distributed nature of the people, process and technology components of enterprise applications, it is recommended that enterprises engage very experienced DevOps consultants to determine strategic requirements, roadmaps and plans for DevOps that fully consider all three dimensions (people, processes and technologies) and nine pillars of DevOps before embarking on a major DevOps transformation journey.
Despite the complexity, with the right strategy and plan, DevOps can deliver enormous ROI for enterprise applications. Amazon and Adobe are just two of many examples of large corporations that have benefited enormously from applying DevOps across the enterprise.
DevOps Applied to Software Services
I often hear people ask, “Does DevOps apply to software services as well as it does products?” These are usually people who have only worked in product-oriented organizations, such as companies that manufacture things that have embedded software.
Ironically, I often hear people ask the opposite question, “Does DevOps apply to products with embedded software as well as it does to services?” These are usually people who have only worked organizations in which software is consumed as a service over an intranet or internet for a price or for free. Examples of these organizations are IT departments (even the IT departments of manufacturers!), government agencies and enterprises such as banks, insurance companies, hospitals and financial trading companies.
You already know the answer to these two questions if you read the earlier parts of this blog series. The answer to both is clearly ‘Yes!’ Cloud services are doubly interesting to DevOps because not only are cloud services used by DevOps as a preferred infrastructure, as described in the next chapter, but cloud services also use DevOps in the process of creating and deploying software for the implementation of cloud components.
While we are on the topic of DevOps and services, I want to help clarify some confusion regarding whether “newer” DevOps renders “older” IT service management (ITSM) obsolete or less important. This confusion exists because ITSM has often been regarded as a gatekeeper run by Ops to prevent Dev from getting out of control. While some organizations may have, in fact, implemented ITSM that way, this is not the purpose of ITSM.
“DevOps tends to focus on the software delivery life cycle and on building and delivering software,” said Jayne Groll, president of the ITSM Academy. For its part, ITSM tends to have a broader focus; it is really looking at all aspects of customer engagement, from fulfilling standard services to ongoing support and operations activities. So, ITSM is “almost like the left and right ends of DevOps.”
Using ITSM With DevOps
Here are some recommended engineering practices for using ITSM with DevOps:
• Training to help DevOps professionals gain an understanding of ITSM, and vice versa, will help both teams find points of integration—not just in technology, but in process, as well.
• Find ways to automate service change approvals implementation of known, good changes to allow for agility, iteration and innovation. At the same time, hold back high-risk, high-impact changes that might need more than just an individual contributor to sign off on before they make it in front of a customer.
• Self-service requests, automation flows and IT orchestration are now the front door to IT service management actions. These can and should be happening in the background, automagically created, maintained and measured by our DevOps tools.
• Embrace Agile service management to take Agile values and apply Scrum-like development practices to process design and improvement. Aim to make more frequent incremental changes rather than design and implement a total end-to-end process in the traditional way.
What This Means
This blog is part three of a series to explain how to engineer applications to be most suitable for DevOps. This blog series explains how to engineer applications for DevOps including:
• Factors to decide whether an application is a good candidate for DevOps.
• Practices to engineer designs for DevOps.
• DevOps applied to enterprise apps and software services.
• DevOps applied to COTS systems.
• DevOps applied to manufactured (embedded) systems software.
• Five levels of application maturity.
This blog covered how DevOps can be applied to enterprise apps and software services. You can read part one here and part two here.
For more information regarding how to engineer applications for DevOps refer to my book Engineering DevOps.
