NGC | Catalog
Welcome Guest
CatalogHelm ChartsVyasa Product Suite

Vyasa Product Suite

For versions and more information, please view on a desktop device.
Logo for Vyasa Product Suite

Description

Vyasa's collection of highly-scalable deep learning software for disparate and diverse biomedical datasets.

Publisher

Vyasa Analytics

Latest Version

0.2.72

Compressed Size

55.07 KB

Modified

September 15, 2021

Vyasa Analytics

Vyasa Analytics provides highly scalable deep learning software and analytics. We enable organizations to ask complex questions across large scale integrated biomedical data sets, thereby gaining critical insights to make better decisions.

Vyasa Product Suite

The Vyasa Product Suite includes Layar, Axon, Retina and Synapse, each of which is specifically designed for cutting-edge data analytics use cases. From data integration and search in Layar to deep learning image analytics with Retina to intuitive dynamic knowledge graph creation with Axon, experience the next generation of A.I. powered analytics capabilities with the Vyasa Product Suite.

What's included in the Vyasa Product Suite?

Layar - Layar is a secure, highly scalable, data fabric platform built specifically for enterprise analytics. Layar can be added to existing enterprise data architectures to augment analytics capabilities or can operate as a stand-alone data fabric for text, image, and data stream integration and analytics.

Axon - A knowledge graph application that enables derivation of dynamically generated knowledge graphs directly from integrated data and documents sources integrated in a Layar data fabric.

Retina - Retina is an image analytics application that offers a wide range of deep learning image-related tasks, including management, annotation and deep learning analytics on images.

Synapse - Synapse provides Smart Table Technology that directly connects a user's spreadsheet content to the analytical capabilities of Layar Data Fabrics.

System Requirements

Hardware

Please ensure your system meets the following hardware requirements:

  • 128 GB RAM
  • 16 CPUs
  • 500GB SSD disk
  • Four V100 or later generation GPUs

Software

Operating system:

  • Ubuntu 18
  • Ubuntu 20
  • DGX OS 5

Additional Software:

  • Ensure you have a functioning Helm v3 cluster (example set up guide).
  • Ensure you have a functioning NVIDIA Docker installation. The following command should show the available GPUs:
sudo docker run --rm --gpus all nvidia/cuda:11.0-base nvidia-smi

Installation

  1. Increase the operating system max_map_count:
echo "vm.max_map_count=262144" > /etc/sysctl.d/99-vyasa.conf
sysctl -p /etc/sysctl.d/99-vyasa.conf
  1. Install prerequisites to provide Ingress, Kubernetes CNI, and a default storage provisioner. Skip these steps if you have independently provided these prerequisites.
/usr/bin/kubectl apply -f https://docs.projectcalico.org/manifests/tigera-operator.yaml
/usr/bin/kubectl apply -f https://vyasa-static-assets.s3.amazonaws.com/layar/custom-resources.yaml
/usr/bin/kubectl apply -f https://vyasa-static-assets.s3.amazonaws.com/layar/nginx-ingress.yaml
/usr/bin/kubectl apply -f https://vyasa-static-assets.s3.amazonaws.com/layar/local-storage-provisioner.yml
/usr/bin/kubectl patch storageclass 'local-path' -p '{"metadata":{"annotations":{"storageclass.kubernetes.io/is-default-class":"true"}}}'
/usr/bin/kubectl taint nodes --all node-role.kubernetes.io/master:NoSchedule-
  1. Download and install the helm chart using the fetch command at the top of the page. Change MY_APP_URL to the DNS name or IP address of the system. This is the location you will browse to once Layar is started in order to use the software. Change X.Y.Z below to match the version used in the fetch commmand.
helm install layar ./layar-X.Y.Z.tgz --set APPURL=MY_APP_URL

Once launched you can monitor progress of stack startup by running:

kubectl get pods | grep vyasa

Once all pods have entered the Running state you should be able to visit the APPURL page defined above and get the Layar registration screen. It may take an additional few minutes for the screen to load even though the pods are all in the Running state. Be sure to specify http:// when connecting and not https://. SSL enablement is described further in the documentation.

Technical Support & Documentation

A README for the Vyasa Product Suite Helm Chart can be found here. Documentation on API endpoints is located here. An example application using these endpoints is available here.

For additional tutorials on the various applications and features, support documentation can be found in the Vyasa Knowledge Base.

License Information

By using the software you agree to fully comply with the terms and conditions of the SLA (Software License Agreement) located here.