Machine-Learning

Machine Learning Stack

Let data science teams build, train and deploy machine learning models efficiently and at scale on Kubernetes. Real-world algorithms require real-world compute.

Machine learning is a multi-step process. For beginners, it can be a daunting endeavor. For seasoned professionals, it can be a challenge, as well. Our SuperHub connects the software for a fully functional machine learning pipeline in minutes. We deliver unmatched levels of automation and ease-of-use for your machine learning initiatives. Data scientists get back time to focus on the modeling tasks and not infrastructure operations.

A value we bring to you is the curation and composition of elements needed in the standard ML pipeline:

Data preparation / ETL
Model training and testing
Model evaluation and validation
Deployment and versioning
Production and monitoring
Continuous training/reinforcement learning

An effort to create a sufficiently scalable solution could take an engineer months. We can deploy your machine learning stack through our automation platform in under an hour. Achieving your company's strategic AI initiative is now available in a safe, easy, and reliable platform. The Machine Learning Stack incorporates open, standard software for machine learning: Kubeflow, TensorFlow, Keras, PyTorch, Argo, and others. At the core of Machine Learning Stack is the open source Kubeflow platform, enhanced and automated using AgileStacks' own security, monitoring, CI/CD, workflows, and configuration management capabilities.

MLControlPlane_1467x871_OSXframe_True24

ML Platform

Provisioned as our opinionated preference for ML workflows running on a highly scalable software infrastructure.

Kubeflow, Kubernetes

ML Frameworks

Select your machine learning and deep learning framework, toolkit, and libraries.

TensorFlow, Keras, Cafee, PyTorch

Storage Volume Management

Choose from software and tools for storage to meet your high performance ML needs

Local FS, AWS EFS, AWS EBS, Ceph (block and object), Minio, NFS, HDFS

Container Image Governance

Choose from software and tools that register, secure and manage the distribution of container images.

AWS ECR, Harbor, GitLab

Workflow Engine

Provisioned by default to govern scheduling and coordination of jobs

Argo

Model Training

Include collaboration tooling and interactive model training as part of your template

JupyterHub, TensorBoard, Argo workflow templates

Model Serving

Pick the tool to expose trained models to business applications.

Seldon, tf-serving

Model Validation

Set by default, models will be evaluated against test data as part of your ML pipeline.

Argo workflow templates

Data Storage Services

Choose from storage options befitting the performance of other ML services.

Mini, AWS S3, MongoDB, Cassandra, HDFS

Data Preparation & Processing

Select your tooling to manage the data processing event of your ML pipeline

Argo, NATS, workflow application templates

Infrastructure Monitoring

Elect which reporting and dashboarding tool gives you the better optics into your stack performance.

Prometheus, Grafana

Model Monitoring

Find and choose the appropriate tool to watch model accuracy over time.

Prometheus, Grafana, Isto

Load Balancing & Ingress

Determine the appropriate tool to expose cluster services broadly to other application services.

ELB, Traefik, Ambassador

Security

Find the right tooling for you to manage certificates, passwords and secrets tuned for RBAC across all hybrid-cloud environments.

Okta, Hashicorp Vault, AWS Certificate Manager

Log Management

Make logging easier by choosing pre-integrated tools for ingest, analysis and reporting.

ElasticSearch, Fluentd, Kibana

Machine Learning Stack

Let data science teams build, train and deploy machine learning models efficiently and at scale on Kubernetes. Real-world algorithms require real-world compute.