Between DevOps, DataOps, MLOps and ModelOps, there are different "Ops" based on different environments. "Ops" generally is the shortened version of Operations. Check out some of the different ones in our current technological world. How many do you know?

Learning about DevOps

DevOps or Developer Operations refers to applying agile practices to deliver software continuously with better collaboration between software development and the IT operations team. DevOps has its roots in agile project management and adds continuous value through iterative delivery cycles.

What is DataOps?

Almost all businesses gather data. It can be different types and structures of data. Many customers within and outside the business consume the data with one universal goal of delivering value and insights at the right time.

Data Operations or DataOps is a practice that brings in an end-to-end data pipeline process to ensure every consumer from business users and Data Scientists can deliver results.

DataOps can orchestrate people, processes and technology that enables collaboration across the organisation and enhance agility, speed, and automation of processes.

How do DevOps and DataOps differ?

The key difference between DevOps and DataOps is while DevOps focuses on delivering valuable software to customers by deployments, DataOp s is focused on creating data streams that constantly deliver data.

As the name suggests, developer ops focus on the software development aspects of delivery. In contrast, data ops have a major delivery focus on getting data across, which inevitably means the practices, processes, and pipelines are significantly different.

Describing ModelOps

All models within an analytical lifecycle will have to pass through ModelOps when a production delivery stack is established within organisations. ModelOps is short for Model Operations, and this is a holistic approach to making predictive analytics and Machine Learning workflows operational. Model Ops is also about the governance and lifecycle management of AI and machine learning models.

Learn about MLOps

MLOps or machine learning operations is the DevOps for machine learning algorithms. This is all about operationalising machine learning models. MLOps is a collaboration between data scientists and operations teams that deploy those solutions and enable continuous delivery of ML models in production.

What is the difference between ModelOps and MLOps?

MLOps can be formed as a subset of ModelOps however, some aspects of ModelOps and MLOps can overlap or stay as distinct processes. MLOps can sit within the data science function whereas the ModelOps can expand across the enterprise and may sit within the IT function. In an ideal world, they complement each other.

In conclusion, all these operations help with different aspects of your data lifecycle. Getting the foundational platform will yield high-value results in governed efficient AI across the organisation.

Learn more about DataOps and ModelOps and how to build scalable data pipelines to deliver trusted, high-quality insights with SAS.


About Author

Prathiba Krishna

Data Scientist, SAS

Prathiba is an experienced Data Scientist with a rich background in the Insurance industry. With a Master’s degree in Operational Research with Applied Statistics and Risk, her passion takes form through seeing the varying applications of Machine Learning and AI techniques, and how they propel data scientists to build better models and solutions. Skilled in data analysis and modelling, she utilizes SAS software and Open Source to assess and address problems within enterprise organizations.

1 Comment

Leave A Reply

Back to Top