NVIDIA AI Enterprise with RAPIDS Accelerator Deployment Guide (Latest)

NVIDIA AI Enterprise

• Enterprise-Grade AI Software Platform
  
  • Enterprise-Grade AI Software Platform From NVIDIA
  • NVIDIA AI Software Supported
  • NVIDIA AI Workflows
  • Production-Ready Pretrained Models
  • Supported Hardware and Software

Overview

• Overview

Access the NVIDIA AI Enterprise RAPIDS Accelerator

• Access the NVIDIA AI Enterprise RAPIDS Accelerator

Google Cloud Dataproc

• Google Cloud Dataproc Deployment Guide
  
  • Overview of Steps
  • Prerequisites
  • Connectivity
  • Upload Jar to the Cloud (GCP Bucket)
  • Create the Cluster and Update Dataproc Jar
  • Validation
  • Cluster Cleanup

Databricks

• Databricks Deployment Guide
  
  • Overview of Steps
  • Prerequisites
  • Connectivity
  • Upload Jar and Initialization Script
  • Cluster Creation with Databricks Clusters API
  • Validation
  • Cluster Cleanup

Amazon EMR

• Amazon EMR Deployment UserGuide
  
  • Overview of Steps
  • Prerequisites
  • Connectivity
  • Create and Launch AWS EMR with GPU Nodes
    
    • Create a Key pair
    • Launch an EMR Cluster using AWS CLI
    • Create EMR Cluster with Specified Subnet (Optional)
  • Validation
    
    • Running an example join operation using Spark Shell
    • Spark Submit Jobs to a EMR Cluster Accelerated by GPUs
  • Spark History Server UI
  • Cluster Cleanup

Getting Started with RAPIDS and Kubernetes

• Getting Started with RAPIDS and Kubernetes
  
  • Prerequisites
  • Docker Image Preparation
  • Pulling Docker Image from NGC into Kubernetes
  • Running Spark Applications in the Kubernetes Cluster
    
    • Submitting a Simple Test Job

Appendix

• NGC API Key
• Additional Resources
• RAPIDS Accelerator for Apache Spark Configuration
  
  • General Configuration
  • Supported GPU Operators and Fine Tuning
    
    • Expressions
    • Execution
    • Commands
    • Scans
    • Partitioning
• RAPIDS Accelerator for Apache Spark Compatibility with Apache Spark
  
  • Ordering of Output
  • Floating Point
  • Unicode
  • CSV Reading
    
    • CSV Strings
    • CSV Dates
    • CSV Timestamps
    • CSV Floating Point
    • CSV ANSI day time interval
  • Hive Text File
    
    • Hive Text File Floating Point
    • Hive Text File Decimal
  • ORC
    
    • Push Down Aggregates for ORC
    • Limitations With RAPIDS
  • Parquet
  • JSON
    
    • JSON supporting types
    • JSON Floating Point
    • JSON Timestamps
    • JSON Schema discovery
    • JSON options
  • Avro
    
    • Supported Types
  • Regular Expressions
  • Timestamps
  • Windowing
    
    • Window Functions
    • Range Window
  • Parsing strings as dates or timestamps
    
    • CORRECTED and EXCEPTION timeParserPolicy
    • LEGACY timeParserPolicy
    • Formatting dates and timestamps as strings
  • Casting between types
    
    • Float to Decimal
    • Float to Integral Types
    • Float to String
    • String to Float
    • String to Date
    • String to Timestamp
    • Constant Folding
    • long/double to Timestamp
  • JSON string handling
  • Approximate Percentile
  • Conditionals and operations with side effects (ANSI mode)
• Supported Operators
  
  • General Limitations
    
    • Configuration
  • Key
    
    • Types
    • Support
  • SparkPlan or Executor Nodes
  • Expression and SQL Functions
  • Casting
  • Partitioning
  • Input/Output
  • Apache Iceberg Support

Support

• Support and Services

Notices

• Notices
  
  • Notice
  • Trademarks
  • Copyright