General

The on-premise infrastructure is based on the Kubernetes framework, an open-source container-orchestration framework for automating deployment, and management of containerized applications. The infrastructure consists of high memory servers and GPU enabled servers backed by CEPH storage.

The platform is designed for Big Data analytics and a typical suite of analytical tools such as Jupyter, R, Python, SAS and Slurm are available.

The infrastructure is located at the Stanford Research Computing Facility (SRCF). SRCF is a Stanford data center located on the SLAC campus, designed specifically to host high density, power-consumptive computing equipment.

Cost for using Nero On-Premise

On-premise Nero is currently fully subsidized to users.

Prerequisites to Getting Started

Please complete the User Prerequisites before requesting a Nero on-premise account.

Primary Methods for Utilizing Nero Resources

Method Description
Jupyter Lab JupyterLab is an interactive development environment for working with notebooks, code and data
Interactive Session w/ Slurm    Access Nero via Linux Login Node to configure scripts and environment. Then request compute resources via the Slurm Job Scheduler

Base Storage Allocations

Type     Size     Replication     Path    
Home Directory 10GB yes /home/$SUNETID
PI Directory 1TB yes /share/pi/$PISUNETID

PI Directories are best suited for storing data collaboratively with other members of your group. Each Nero user is part of a PI-Group. View your membership with “id $SUNETID” and look for “$PISUNET-pi” for this information.

Hardware Technical Specs

Type Qty Details
Login nodes 1  
Compute nodes        16 Dual Core Intel Xeon Gold 6130 (16C 2.1GHz), 384 GB RAM
GPU nodes 2 Dual Core Intel Xeon Gold 6130 (16C 2.1GHz), 384 GB RAM + Dual P100 GPUs   
GPU nodes 6 Dual Intel Xeon Silver 4114 (10C 2.2GHz), 256 GB RAM, Quad NVIDIA Tesla V100
OS   Ubuntu 16.04