• Evaluation Guide
•  1. Introduction
  • Actian Vector, Hadoop, and Actian VectorH
  • Using the Actian Sites
•  2. Evaluation Process
  • Evaluation Steps
  • Functional Test Criteria
  • Non-functional Test Criteria
  • Success Criteria
  • Implementation Design
  • Disaster Recovery
•  3. Stage 1--Define and Create the Test Environment
  • Where to Undertake the Evaluation?
  • Evaluation Using Actian Vector
  • Which Hadoop Distribution to Use?
  • Cluster Environment: Hardware and Software
  •  Installation
    • Recommended Hadoop Settings
    • Pre-installation Tasks
    • Installing with root Access
    • Which DataNodes to Install VectorH On?
    • Installing with sudo Access
    • Installing Without Privileged Access
    • Installing Using a Response File
    •  How to Install on a Kerberos-enabled Cluster
      • Automatic Management of Kerberos Details
    • Resetting the Installation
    • Verifying the Installation
  •  Post Installation Tasks
    • Linux Configuration Settings
    •  Configuring Firewall Ports
      • TCP/IP Routing for Edge Nodes
    • Configuring to Run as a Service
    • Configuring Database Resources
    • Giving the Database User a Password
    • Add Environment Setup to the Login Script
    • Using the Correct Ethernet Connection
    • Enabling Short Circuit Reads
    • Running Under YARN
    • Controlling Checkpoint Disk Space
  •  Using VectorH
    • Stopping and Starting the Instance
    • Creating and Removing Databases
    • Stopping and Starting a Database
    • Users, Groups, Profiles, Permissions
•  4. Stage 2--Migrate the Schema, Data, and Queries
  • Migrating a Database Schema to Vector
  • Partitioning Tables
  • Loading Data
  • Generating Statistics
  • Routine Maintenance
• 5. Stage 3--Run a Single-user Test
•  6. Stage 4--Optimize the Database Schema
  •  Performance Considerations
    • Table Partitioning
    • Query Statistics
    • Ordered Data
    • Indexing
    • Nulls
    • Using Foreign Key Specifications
    • Using Correct Data Types
    • Query Profiling
•  7. Stage 5--Run a Multi-user Concurrency Test
  • How Do I Run a Concurrency Test?
  • Tuning for Concurrency
  • Small Data Volumes
  • How Busy Is My Cluster During the Test Run?
•  8. Stage 6--Run the Non-functional Tests
  • Non-functional Success Criteria
  • Testing Master Node Failure
  • Testing Slave Node Failure
  • Using Active Directory Authentication
•  9. Tools and Troubleshooting
  •  Tools
    • vwadmin Utility
  •  Troubleshooting and Log Files
    • Transparent Huge Pages and Defragmentation
    • Hortonworks and Hadoop Client Installation
    • Failure to Start or Slow Response
    • Data Optimization
    • Database/Instance Will Not Stop Due to Connections
    • Maintaining Error Log Size (Rotating the Log)
    • Enabling Disk Use for Query Operations
    • Ensuring Correct Database Configuration
    • Optimizer Ran Out of Memory
• 10. Running the Performance Test Kit
• A. References and Further Information
• B. Manually Running the DBT-3 Sample Performance Tests
• C. Create Table Statements
• D. Create Ordered Table Statements
•  E. Test Queries
  • Query 1
  • Query 2
  • Query 3
  • Query 4
  • Query 5
  • Query 6
  • Query 7
  • Query 8
•  F. Test Scripts
  • RemoteExec.sh
• G. Configuration Checklist

Evaluation Guide

G. Configuration Checklist