Enterprise MLOps in the Cloud

Work faster, deploy results sooner, scale rapidly, and reduce regulatory and operational risk

Domino provides a central system of record for data science activity across an organization. Domino orchestrates all data science artifacts, including AWS infrastructure, data, and services.

Code-first data science teams benefit from a flexible, collaborative, and reproducible research environment, with self-service access to powerful AWS infrastructure within the governance of IT.

Unleash Data Science in the Cloud

Enterprise MLOps on AWS

Domino’s Enterprise MLOps Platform runs on Amazon Elastic Kubernetes Service (EKS). Domino is an AWS Advanced Technology Partner, with both ML and Financial Services competencies.

Self-serve Infrastructure

Remove common DevOps barriers for data scientists with pre-built data connectors supporting Amazon S3, Redshift, EMR, and others. Self-serve access to Amazon EC2 machines (including GPUs and powerful servers).

Deploy Models Anywhere

Deploy Domino models in Amazon SageMaker for scalability and low-latency hosting from AWS. Deploy SageMaker models in Domino to monitor model performance across tools and teams.

Built for Data Science Teams

Provide project management and collaboration across data science teams while supporting the tools (Jupyter, R, SAS, MATLAB), packages, and compute frameworks (Spark, Ray, Dask, MPI) of choice. Compound knowledge instead of reinventing the wheel.

Demonstrated Expertise for ML Solutions on AWS

Domino Data Lab is a founding member of the AWS Machine Learning Competency Partners, validated since 2017.

AWS has identified Domino as one of the key innovators in helping organizations extract real value from their data scientist investments.

Domino Data Lab is also an AWS Financial Services Competency Partner due to Domino's large footprint in banking, insurance, and algorithmic trading.

Collaborative, flexible model development and deployment for teams.

Orchestrate AWS infrastructure, data, and services for data scientists.

Host on AWS: AWS-Certified

Enterprise MLOps Deployed on AWS

Domino is an AWS Machine Learning Competency Partner validated since 2017. Domino’s Enterprise MLOps Platform on Amazon EKS provides ready-to-deploy, field-tested deployment patterns for ease of management. Domino can run in a dedicated VPC in the AWS region of your choice, including GovCloud.

Kubernetes brings a whole new set of benefits to containerized applications, including efficient hardware/compute utilization. K8s has the capability to automatically scale up when capacity peaks and scale down again after a peak.

Scaling a cluster up or down is quick and easy because it’s a matter of just adding or removing virtual machines (VMs) to the cluster. This dynamic resource allocation is especially beneficial for data science workloads; the demand for high-powered CPUs, GPUs and RAM can be extremely intensive when training models or engineering features, but then the demand can scale down again very quickly.

Domino Documentation: How to Deploy Domino on Amazon EKS
Domino 4.3 Blog: Domino Support for Amazon EKS

Pre-built Connectors: Seamless AWS Data Access

Seamless Access to Amazon Redshift and Amazon S3

Domino Data Sources provides a mechanism to create and manage connection properties to external supported data services, such as Amazon Redshift, Amazon S3, and others. Connection properties are stored securely and there is no need to install data source-specific drivers and libraries.

Data Sources in Domino democratizes access to data by eliminating DevOps barriers related to driver installation, specific libraries, and more. Team collaboration is enhanced through data source connector sharing with colleagues. IT teams will be pleased that the capability supports per-user and service account credentials to maximize flexibility while maintaining the highest levels of data security.

Self-Serve Infrastructure: Automate DevOps

Flexible Tooling and Self-Serve Infrastructure - Governed by IT

Domino’s collaborative and flexible platform lets data scientists use the tools and packages of their choice, including Amazon SageMaker, Jupyter, RStudio, SAS, Anaconda, MATLAB, along with flexible compute frameworks (i.e., Spark, Ray, and Dask). Code-first data scientists get the benefit of project management and collaborative, reproducible research environments with flexible deployment options – maximizing productivity and business impact by compounding knowledge.

Abstract away the complexity of managing infrastructure and connecting to data sources, so data scientists can focus on innovation. Provide self-serve, easy access to Amazon EC2 machines, including GPUs and powerful servers, to run experiments faster and test more ideas to accelerate model development.

Self-Service Infrastructure Portal Product Page: Domino automates time-consuming DevOps tasks required for data science work at scale.
NVIDIA GPUs in AWS: Amazon EC2 instances powered by NVIDIA GPUs deliver the scalable performance needed for fast ML training, cost-effective ML inference, flexible remote virtual workstations, and powerful HPC computations.
Data Science Blog: Spark, Dask, and Ray: Choosing the Right Framework.

Models Anywhere: Flexible Deployment

Deploy and Monitor Models Anywhere

Models developed in Domino can be exported for deployment in Amazon SageMaker, giving customers the choice for AWS’ own scalable and low-latency hosting. Models developed in Amazon SageMaker and Amazon SageMaker Autopilot can be accessed inside Domino to support diverse business and operational requirements, then monitored for drift and prediction performance issues before the prediction loses its accuracy.

Amazon SageMaker Services/Features Complimentary to Domino Data Lab:

SageMaker Data Wrangler
SageMaker Batch Transform
SageMaker Data Labeling
SageMaker AutoPilot
SageMaker Neo
SageMaker Edge Manager
SageMaker Feature Store
SageMaker Clarify
SageMaker Serverless Inference
SageMaker Asynchronous Interface

More Resources

On-Demand Webinar: Model Management and Monitoring for the Enterprise (Deploy a SageMaker Autopilot Model in Domino).
AWS APN Blog: How to Export a Model from Domino for Deployment in SageMaker.
Github Project: How to Architect end-to-end development, monitoring, and maintenance of your models in AWS and Domino Data Lab.