Follow Datanami:
August 22, 2022

5 Steps to Achieve MLOps at Scale

A computer program beating the world champion of the game of Go? No way, right? Wrong. AlphaGo[1] machine learning (ML) and artificial intelligence (AI) technology beat the world champ. Five years later, AlphaFold is using AI/ML to solve some of the core challenges in biology.[2] Ironically, 85% of AI projects still fail[3] because they cannot deliver on their intended business promises. For ML models to be successful, data science projects must address three challenges:

  1. Trusted and governed data
  2. A multi-cloud ML strategy
  3. AI/ML automation

How MLOps Delivers Value

Machine learning operations (MLOps) is the process of streamlining ML models. MLOps focuses on data model deployment, operationalization and execution. This standard set of practices lets you enable the full power of AI at scale. It also lets you deliver trusted, machine-led decisions in real-time. MLOps combines model development and operations technologies. This is essential to high-performing AI solutions.

Many organizations follow the process of build, test and train ML models. But how can you provide continuous feedback? This is especially important once the models are in production. Data scientists can’t be responsible for the management of an end-to-end ML pipeline. What’s needed is a team with the right mix of technical skillsets to manage the orchestration. This establishes a continuous delivery cycle of models and form the basis for AI-based systems.

The benefits of implementing MLOps include the ability to:

  • Deliver business value for data science projects
  • Improve the efficiency of the data science team
  • Allow ML models to run more predictably with better results
  • Help enterprises improve revenue and operational efficiency
  • Accelerate AI/ML initiatives with high performing models

MLOps in 5 Steps with Informatica

Informatica Intelligent Data Management Cloud™ (IDMC), our comprehensive AI-powered data management platform, helps accelerate your data science initiatives by operationalizing ML models in 5 easy steps:

1: Identify the business problem and acquire data. Identify trusted data from various sources, including:
-Internet of Things (IoT) devices
-Relational databases
-Mainframe systems
-On-premises data warehouses

Then load them into a cloud data lake. Informatica’s Enterprise Data Catalog helps you identify trusted data. Informatica’s Cloud Mass Ingestion lets you ingest trusted data into the cloud data lake. Informatica’s AI-driven intelligent metadata discovery lets you discover your data assets. Then, apply them to a data pipeline so data engineers can search for inventory data.

2: Curate, cleanse and prepare data. Once ingested into a cloud data lake, you can cleanse and match your data and standardize rules. This ensures that your data is ready to consume. Informatica Cloud Data Quality has a drag and drop configuration so you can build, test and run data quality plans.

3: Build ML models. Data scientists can operationalize their learning models so they can build and test them. Data scientists can build their models using any framework and they can use Informatica’s Spark-based data integration engine or any application for consumption. For model development, Informatica runs on Advanced Serverless deployment. In fact, we are the industry’s first solution to run on advanced serverless deployment, which provides a pipeline for cleansed training data. CLAIRE®, our AI engine, applies the industry-leading metadata capabilities to speed up and automate your core data management. To run jobs at scale, you can apply auto-scaling and auto-tuning for better performance. Also available are innovations such as run time optimizations, advanced data management and elastic operations.

4: Deploy ML models. Informatica ModelServe enables data scientists and ML engineers to register and operationalize ML models developed in any workbench or framework. ModelServe helps accelerate AI/ML automation at an enterprise scale in minutes. This simplifies the process of deploying ML models to production without worrying about the underlying infrastructure and scaling the infrastructure.

5: Model monitoring. DataOps teams can monitor model performance using Informatica ModelServe service in IDMC for continued value delivery. They can also leverage Informatica’s built-in monitoring and alerting capabilities. These automate monitoring and management of your models. From there, you can automate ML model operationalization.

Learn More

MLOps is vital to operationalize data science use cases, drive business value and speed up digital transformation. Informatica’s end-to-end MLOps works on any platform and any cloud, multi-cloud and multi-hybrid.

Learn how to operationalize your ML models or contact us to explore next steps.





 machine learning (ML) and artificial intelligence (AI) technology beat the world champ. Read more…

" share_counter=""]