The ‘Ops’ concept is taking root in enterprise technology stores, but also new worries

DevOps – which promotes greater collaboration and automation in software delivery – is just the beginning of a new phase of technology management. Now we’re seeing a lot of spinoffs — DataOps, Machine Learning Operations (MLOps), ModelOps — and other Ops looking to add speed, reliability, and collaboration to the delivery of software and data across business channels. There’s even a DataOps Manifesto, which bears a striking resemblance to the Agile Manifesto created in 2001.

rubriks-cube-aug-2020-photo-by-joe-mckendrick.jpg

Photo: Joe McKendrick

None of this will happen overnight, however. Or even within a few months. As with any promising technological overhaul, a rethinking of processes and culture is essential.

Where are IT managers and professionals? How are they supposed to move forward with all these operations that promise a smoother and more responsive service? “An important part of the preparation is asking important questions about existing processes, both formal and informal,” said Alice McClure, director of artificial intelligence and analytics at SAS. “This helps determine where to focus first, what to update, and where bottlenecks exist.”

DataOps, for example, “provides a flexible approach to data access, quality, preparation and management — the entire data lifecycle, from preparation to reporting,” McClure says. “It brings greater reliability, speed, and collaboration to your efforts to operationalize data and analytic workflows. ModelOps is becoming an indispensable methodology for implementing scalable predictive analytics. It’s about getting analytics into production – iteratively moving models quickly through the lifecycle of move analytics while ensuring quality and enabling ongoing monitoring and governance of models over time.”

It’s all about bringing automation and architecture together, advises Amar Arsikere, CTO and co-founder of InfoWorks. “Implementing a system that automates the operation and orchestration of data, metadata, and workloads, rather than hand-coded, manual operations that require time, money, and specialized resources.”

xOps approaches become a necessity as manual adverse applications such as artificial intelligence and machine learning come to the fore. “Addressing these challenges is often an afterthought and ultimately comes down to DevOps and IT teams,” said Rahul Pradhan, VP Product and Strategy for Cloud Platforms for Couchbase. Emerging priorities such as continuous integration and continuous delivery, automation and real-time monitoring are putting pressure on these teams, he adds. “Not only are these teams being asked to do more, they are also being asked to be broader and full-stack. This highlights the need to eliminate operational, low-level tasks such as managing infrastructure and databases.”

Most operations “are heavily scripted or automated, but real success is achieved when the entire process is automated from start to finish,” agrees Patrick McFadin, VP of Developer Relations at DataStax. “This includes day two operations such as scaling. xOps can follow a similar path that site reliability engineers take for training and preparation as they deal with the same issues in cloud native applications.”

Contrary to popular belief, having a successful xOps effort doesn’t mean companies can cut their IT staffing levels — quite the contrary, it means stepping up their recruiting and retention games. Shortages of IT talent “can significantly hamper xOps initiatives,” Pradhan says. “Focusing more efforts on developer retention. Taking proactive steps to keep developers engaged and satisfied can help prevent digital transformation burnout.”

There is another key element to the success of xOps: time to implement and overcome old corporate cultures. A new ModelOps or DataOps methodology “cannot be implemented and built in a day,” emphasizes Pradhan. “It takes time to transform processes. Getting the right teams involved at the start of a project is critical and should include establishing measurable outcomes and a clear understanding of roles.”

The challenge is “to change the mindset of the teams to be organized around the goals and results of the business transformation,” Arsikere says. “Rethinking implementation by automating end-to-end processes rather than relying on manual manual coding or disparate point solutions.”

That’s where Ops methodologies “can help simplify things, increase business value while ensuring the best customer experience,” Pradhan urges. He urges a composable approach — akin to a Lego building block strategy — “to help reduce the tension that can arise as xOps capabilities and digital transformation strategies are built. The same blocks and strategy can be used over and over.”

Plus, it’s time to bring application and data infrastructure development and deployment under one roof, says McFadin. “Don’t cling to old methodologies,” he says. “I often see companies segregating application and data infrastructure using different methods and standards. Committing to one path for both code and data can open up a lot of opportunities. That means finding ways to cloud the data portion of the application stack. to make it native.”

Embracing cloud native for data “separates the teams that act quickly from those that don’t,” says McFadin. “That means using everything available in the Kubernetes ecosystem to their advantage. From CI/CD to observability, the goal is to create repeatable and trusted systems. DevOps has had an early lead with projects addressing a variety of issues. MLOps and DataOps are now rapidly catching up with new and emerging projects.”

Leave a Comment