Insights > Blog

Get to Know the Redapt ML Accelerator

By Paul Welch | Posted on June 24, 2020 | Posted in AI/ML

Machine learning (ML) is an idea that has taken the business world by storm in recent years. But for many enterprises, it has remained just that—an idea. In fact, by some estimates, 90% of ML models never make it to production.

While some of this failure rate can be blamed on the relative newness of the technology, a lack of skills with the specialized hardware—and software—necessary to deploy ML is only part of the story.

The bigger problem, especially for those enterprises actively working on machine learning projects, is a disconnect between data science teams and IT teams.

What’s driving this disconnect? One of the major culprits is the fact that many—if not most—data scientist ML models are developed on dedicated workstations or cloud instances that IT teams don’t actively manage.

Download Now: The CIO's Guide to Leveraging AI to Leap Ahead of the Market

Because of this, when it comes time to actually move an ML model from the workstation and into production, IT teams are often left scratching their heads about how to deploy the model at scale in a datacenter.  

Enter the ML accelerator

Designed to help enterprises of all sizes bridge the gap between data science and IT, our ML accelerator program can take you from zero to production ready with ML models in an accelerated time frame.

Our goal with the program is to help you increase your R&D production. To that end, we’ve leveraged our strong partnership with hardware providers like Dell EMC to assemble a hardware and software package to get you up and running with ML.

The package’s deliverables include:

  • Ready-to-use infrastructure installed and made operational at your datacenter
  • Engineering assistance with building and deploying your first ML model
  • A workshop focused on getting started on the platform, as well as a workflow plan taking you from dev to deployment
  • Best practices and knowledge transfer so you can build out your own ML models successfully


Under the hood

ML accelerator is designed to be adaptable to your enterprise’s unique needs. Included in the infrastructure are:

  • HA Kubernetes based on Rancher
  • Hardware to support ML and DL workloads, including a base model with 4xv100 GPUs
  • Workflow management with Kubeflow and ready built containers
  • Self-service Jupyter notebooks for data exploration
  • Integration with Nvidia RAPIDS and Spark
  • IT monitoring and alerting with Promethus and Grafana

Combined, these tools and other tools provide you with everything you need to get up and running with ML—all within a production-ready footprint that has been minimized to lower initial startup costs.

Benefits of ML accelerator

Beyond helping you actually take ML models from workstations to production, the ML accelerator program mitigates risk by providing you with experienced consultants with well-versed, proven best practices. Data science workloads will be moved to IT owned infrastructure using industry best practices for IT and software development.

This opens the door for your enterprise to employ ML not just to make smarter decisions, but to drive the creation of entirely new products and capabilities.

To learn more about crafting an ML and AI strategy, download our free eBook, The CIO’s Guide to Leveraging AI to Leap Ahead of the Market.