DevOps and IBM Z: You've Come a Long Way
IBM's 3-pronged DevOps approach helps organizations leverage the IBM Z platform, tools and open-source strategies.
Image by Charles Williams
By Karen E Lewis01/02/2020
Application development operates in open, cloud-native environments. “Using open source in a seamless and supported manner—alongside IBM Z* tooling and capabilities is a vastly different experience,” says Nathan Dotson, hybrid cloud offering manager for IBM Z. DevOps and the IBM Z platform are evolving quickly to meet the rising demand for cloud-native development, striving to empower developers, system administrators and architects to build and deliver applications using common tooling across the development process and varying environments.
A 3-Pronged Approach to DevOps
The convergence of open source and cloud adoption is causing a shift in how technology companies and developers design, deliver and use software. “Incoming cloud-native developers expect to apply their development skills wherever they land—regardless of the platforms used within these organizations,” says Sanjay Chandru, director for Enterprise DevOps, IBM. IBM’s three-pronged DevOps approach is helping organizations harness the power of an integrated development experience (IDE), cutting across platforms, open source and IBM tools. The combination of these capabilities opens all sorts of possibilities for skills acquisition, talent retention and responsive application development. This approach entails:
1. Enterprise-wide standardization
The premise behind DevOps is continuous improvement based on continuous feedback—in any environment. For example, in a heterogeneous multicloud environment, where hybrid cloud development occurs, it’s imperative to standardize processes and tools across the enterprise. It enables developers to work efficiently, leveraging combinations of skills to work in a cloud-based development environment across different platforms—including IBM Z.
2. CI/CD pipeline standardization
Modern development practices embrace continuous integration/continuous delivery (CI/CD) that’s standard across any platform. Clients want to leverage CI/CD that doesn’t require developers to do anything special for any platform. Through these practices, it’s easy to embrace true parallel development using software configuration management like GIT, along with generic artifact repositories like Artifactory and Nexus for developing, building and deploying COBOL, PL/I, Assembler code and more on IBM Z.
The more you can automate, the more efficient your development processes become. Automation and machine learning allow organizations to eliminate the need for specialized skills and extra testing. Applying these capabilities drives speed, reduces costs, minimizes errors and removes bottlenecks—which all contribute to a more streamlined DevOps process.
More Freedom Without Compromise
Embracing open technologies, enterprise standardization, test automation, deployment automation and build automation while using modern development practices puts organizations in the
driver’s seat when making decisions about which platform or cloud to use. “When you use hybrid multicloud, what should define the path for application deployment is the business need, not skills. Suddenly you have a choice—and resource restraints aren’t driving that decision; there’s no need to compromise,” explains Dotson.
The IBM Z server provides the flexibility to deploy workloads both on- and off-premises, making the same trusted data protection, availability and reliability expected from the platform available in the cloud. “Based on individual needs per workload, resources, time and business requirements, organizations can choose to develop cloud-native applications in private cloud, public cloud or a combination of the two,” explains Minaz Merali, program director, IBM Z DevOps. IBM Z security and reliability are combined with the cloud's agility and ease of access to provide a unique set of capabilities.
Leveraging Available Skills
Cloud-native development has particular characteristics. “If you’re developing in a container environment, you’re still driving efficient development using DevOps practices. The ability to leverage practices easily enables organizations to embrace cloud-native development more naturally—resulting in faster transformation,” says Dotson. Standardizing processes and tools give organizations the ability to take advantage of the different skills and knowledge their people bring to the table. For example, people within an organization must know why and how an application was designed, and what the requirements were that merited the level of security, availability, and reliability that drove its deployment on IBM Z.
Accelerate Knowledge Transfer
Long-term IBM clients often employ individuals who possess decades of domain expertise. As these employees retire, they’re leaving with vital knowledge. Organizations facing the brain drain must embrace changes quickly and transition the invaluable domain expertise from the retiring workforce to the recruits entering the workplace.
According to Chandru, knowledge transfer has two aspects. The first is the human dimension; the second involves tools to address the inherent challenges of how to transition tacit knowledge. “Domain experts are people who have invested their entire life into working on mission-critical applications. They’re the employees who have kept their organization running—it’s been their lifelong passion. To them, these applications are like babies, and they feel a deep sense of pride in ownership. They want to see their applications continue to thrive and deliver for the business,” says Chandru.
IBM Application Discovery and Delivery Intelligence (ADDI) is designed to accelerate the transfer of domain expertise, as well as the understanding of the application landscape (ibm.co/2NUzDIR). The tool allows you to run both static analysis and dynamic analysis around an organization’s existing enterprise application landscape. For new developers entering the workforce, the analysis provides a quick on-ramp to understanding the code. As a result, they can immediately speak more intelligently about the code, making their engagement with the organization’s domain experts more fruitful.
Having a basis of understanding about the code bridges the knowledge gap between the new developers and the domain experts. It also makes it easier for domain experts to drive the day-to-day development activities while helping their colleagues get up to speed.
Innovate With Shift Left
IBM has taken a giant leap forward, innovating its test automation capability with a Shift Left approach. “Today, there are a lot of vendors who claim they’re doing Shift Left on IBM Z. In reality, they’re still using a waterfall approach, only enabling the people doing the testing in the QA phase to be more efficient,” explains Merali. Through customer research, IBM has honed in on the real meaning of Shift Left, taking the unit testing capability apart and redesigning it from the ground up.
“Through customer feedback, we’ve evolved the IBM Z Unit and Z Open Unit Test (ibm.co/2o4tDpU). Typically, vendors make assumptions about customer needs. We stepped back and said, ‘Let’s find out what they need. Let’s leverage our customer experiences and redesign how test automation is done,’ ” explains Chandru. Now, organizations can unit test in the build phase of the DevOps lifecycle. The impact is enormous; the earlier you find a problem, the less expensive it is to fix. Catching issues in the development phase versus when it’s in production improves the quality of your software and enables you to save thousands of dollars per defect.
Immediate Cost Reduction and Productivity Gains
When developers quickly understand application complexity and quality across platforms, environments and languages, they’re in a better position to anticipate and thwart potential issues. “If your organization normally takes a month to go from development into testing and production, adopting DevOps practices can reduce that month to three weeks. The more ways you find to optimize and streamline processes throughout the lifecycle, the more productivity gains you’ll experience. It’s a multiplier effect,” says Merali. Learn more about adopting modern DevOps practices in “Accelerate Your Transformation,” left.
To break this down by the numbers:
- Catching mistakes and problems earlier in the lifecycle can cut costs by up to 25% (ibm.co/2BvBpMG)
- According to Forrester, leveraging DevOps for an integrated application development approach results in 48% shorter development/delivery cycles, 41% faster response to business needs and 45% improved collaboration and communication between silos (bit.ly/2N2yTD8)
- An IDC whitepaper titled "The Business Value of the Transformative Mainframe" indicates organizations using DevOps and Agile with IBM Z deliver 64% more code releases more frequently and need 44% less time per release (ibm.co/2pJSrUy)
And, as organizations continue to improve their development processes, the benefits rise substantially.
Overcome Barriers and Remove Bottlenecks
By leveraging an ecosystem of capabilities built around standard tools, organizations can overcome barriers. Using tools like ADDI allows developers to understand the impact of changes being made continuously to the code. This kind of insight enables organizations to transition to Agile DevOps practices and continue to thrive after the transition.
Accelerate Your Transformation
Adopting modern DevOps practices can help enterprises manage application moves successfully while making the transition to Agile DevOps practices.
Embracing open-source automation servers like Jenkins—to automate the non-human part of the software development process, and a true parallel development software configuration manager (SCM) like GIT—represents the tip of the spear. When you start using GIT and the software configuration management capabilities that it brings to the table, you’re equipping your organization with integrated tools and processes to standardize across the enterprise.
Removing silos operating outside of enterprise-wide processes means new talent coming into the organization can become immediately productive. There’s no learning curve. They can leverage what they know directly while gaining an understanding of the application landscape, getting a leg up on engaging with incumbent domain experts.
Modern DevOps practices and tools can help your organization:
Facilitate knowledge transfer
Using tools such as ADDI ensures everyone within the organization understands what the application landscape is. Having this insight enables you to opportunistically and strategically pick the portions of the applications you might want to componentize, create microservices from and expose them to the outside world—allowing you to refactor, creating more consumable applications gradually.
Move to a modern SCM
Moving away from legacy library managers to a modern SCM like GIT is essential. When you make a move, you inherit a robust set of best practices to implement. As a result, you can componentize and handle these applications easily. And, the ongoing development of these applications is far more efficient.
Tap into the DevOps acceleration team
IBM’s DevOps acceleration team (ibm.co/2o76oeU) can help clients establish best practices, set up GIT, and use modern parallel development SCMs. The team collaborates with clients, mentoring them through their DevOps transformation. So, not only can IBM provide organizations with powerful DevOps tools, but they can also assist with the best practices throughout their transition.
z/OS / Linux on IBM Z / z/VM / z/VSE / Article / Cloud / Security / Systems management / DevOps / Modernization / Cloud strategy / IBM Z / Data security / Open source on IBM Z / Automation / High availability
Karen E. Lewis is a global market segment manager with the IBM Systems and Technology Group. She has more than 20 years experience.
Sponsored ContentAchieve Compliance Without Impacting Productivity
Post a Comment
Note: Comments are moderated and will not appear until approvedcomments powered by Disqus