Install Aerospike SQL Powered by Starburst

Aerospike SQL Powered by Starburst provides a tightly-integrated SQL processing engine to the Aerospike Real-time Data Platform. Powered by Starburst Enterprise, Aerospike SQL provides an enterprise-grade implementation of the open source Trino (formerly PrestoSQL) that is installed, managed and supported by Aerospike.

The combination of the Aerospike Database and the Starburst SQL engine offers a rapid and cost effective way for SQL users to generate valuable insights in dashboards, ad-hoc queries, or Python applications.

Aerospike SQL runs Trino on a single machine, a cluster of machines, on-prem, or in the cloud. Aerospike SQL uses the Aerospike Trino connector to reconcile the data model differences which manifest when accessing a NoSQL database such as Aerospike (ASDB) using SQL with an SQL engine, such as Trino.

Overview

A deployment of Aerospike SQL consists of four components.

• Starburst Admin: This is a collection of Ansible playbooks for installing and managing Starburst Enterprise platform (SEP) or Trino clusters.
• Starburst Coordinator node and Starburst Worker node(s): More information about these here.
• Aerospike Connect for Trino.
• Aerospike Database Cluster.

This image illustrates the architectural view. This guide walks you through the setup and configuration of this deployment.

Installing Aerospike SQL broadly consists of the following four steps:

1. Install the Aerospike SQL package on Starburst Admin node.
2. Install the following components using Ansible scripts in the aerospike-sql package:
   • Starburst Enterprise
   • Aerospike Connect for Trino
3. Place the license files.
4. Start the deployment.

Prerequisites

• Aerospike cluster set up and running. See our install documentation for more information.
• Aerospike Enterprise Edition license key.
• Starburst Enterprise license key.
• Aerospike SQL package.

Aerospike SQL is supported on Debian, CentOS 7, and Red Hat Enterprise Linux (RHEL 8 and 9) with standard system utilities installed, including git, rpm, rsync, ssh, tar, wget, and zip/unzip.

danger

Before you proceed, verify passwordless SSH connectivity between the Starburst Admin to Starburst coordinator and worker(s).

tip

Some tips on installing and using ansible on RHEL 9:

dnf install -y ansible-core
ansible-galaxy collection install ansible.posix
yum install -y rsync

note

Run the .deb package without interactive dialogue.

DEBIAN_FRONTEND=noninteractive

Starburst Admin node requirements:

• Python3
• Java JDK 17
• Ansible

Starburst Coordinator and Worker node requirements:

• Python:
  • CentOS 7, RHEL (8 and 9): Python3
  • Debian: Python
• Java JDK 17
• jq command-line JSON processor
• unzip or zipinfo

Pre-Flight checklist

caution

Go through the following table and ensure you meet each requirement. The installation process will fail if the requirements are not met.

Location	Requirement
Starburst Admin node
	Python 3
	Java JDK 17
	Ansible
	Able to SSH without a password to and from the coordinator and worker nodes.
Starburst Coordinator and Worker nodes
	Python 3 (If using CentOS 7/RHEL 8 and 9) Python (If using Debian)
	Java 17
	jq command-line JSON processor
	unzip or zipinfo
Aerospike nodes	Aerospike database version compatible with your Trino connector. See this page for details.

1. Install the Aerospike SQL package

Download the Aerospike SQL package to the Starburst admin node.

1. Install the Aerospike SQL package:

• CentOS 7 and RHEL : sudo rpm -i aerospike-sql-1.3.0-1.noarch.rpm
• Debian: sudo dpkg --force-architecture -i aerospike-sql_1.3.0.noarch.deb

2. After the package is installed, update the configurations.

   a. Go to the installed Aerospike SQL directory:

   cd /opt/aerospike/aerospike-sql

   b. Edit the following lines in the playbooks/host file.

   [coordinator]
   COORDINATOR_IP ansible_user=STARBURST_USER ansible_ssh_private_key_file=~/.ssh/id_rsa
   
   [worker]
   WORKER_1_IP_ADDRESS  ansible_user=STARBURST_USER ansible_ssh_private_key_file=~/.ssh/id_rsa
   # Add all the worker nodes as applicable
   # WORKER_2_IP_ADDRESS  ansible_user=STARBURST_USER ansible_ssh_private_key_file=~/.ssh/id_rsa
   # WORKER_3_IP_ADDRESS  ansible_user=STARBURST_USER ansible_ssh_private_key_file=~/.ssh/id_rsa

   Replace:

   • COORDINATOR_IP with the private IP address of the Starburst coordinator node.
   • WORKER_1_IP_ADDRESS with the private IP address of the Starburst worker node(s).
   • [starburst_user] with your Starburst Admin username.

   c. Edit the following lines in the files/catalog/aerospike.properties file.

   connector.name=aerospike
   # Replace with your public IP for all aerospike nodes in the cluster
   aerospike.hostlist=AEROSPIKE_NODE_1_IP_ADDR:3000,AEROSPIKE_NODE_2_IP_ADDR:3000,AEROSPIKE_NODE_3_IP_ADDR:3000
   aerospike.split-number=8
   aerospike.strict-schemas=false
   aerospike.record-key-hidden=false
   aerospike.enable-statistics=true
   aerospike.insert-require-key=true
   #Change OS_USER as per your OS(eg: default user for CentOS/Fedora/RHEL on AWS is “ec2-user”)
   aerospike.table-desc-dir=/home/OS_USER/
   aerospike.clientpolicy.tls.enabled=false
   aerospike.case-insensitive-identifiers=true

   Replace:

   • AEROSPIKE_NODE_1_IP_ADDR, AEROSPIKE_NODE_2_IP_ADDR, and AEROSPIKE_NODE_3_IP_ADDR with the public IP for the Aerospike nodes.
   • OS_USER with the default username for your OS.

2. Use Ansible to install components

In the next stage, use the Ansible playbooks from the Aerospike SQL package to install Starburst Enterprise and the Aerospike Trino connector.

1. Install Starburst Enterprise with the following Ansible command on the Starburst admin node:

ansible-playbook   playbooks/install.yml

This downloads and installs Starburst Enterprise and Starburst Trino. This process should take less than 10 minutes.

When the installation successfully completes, the tail of the output looks like:

…
PLAY RECAP *****************************************************************************
172.31.2.13                : ok=14   changed=4    unreachable=0    failed=0    skipped=15   rescued=0    ignored=0   
172.31.5.45                : ok=14   changed=4    unreachable=0    failed=0    skipped=15   rescued=0    ignored=0   

For more information refer to the Starburst Trino configuration guide.

2. Use Ansible to Install the Aerospike Trino Connector

Run the following Ansible command on the Starburst admin node to install the Aerospike Trino Connector:

ansible-playbook   playbooks/aerospike.yml

This downloads and installs the Aerospike Trino connector.

When the installation successfully completes, the tail of the output looks like:

PLAY RECAP *****************************************************************************
172.31.2.13                : ok=9    changed=3    unreachable=0    failed=0    skipped=1    rescued=0    ignored=0   
172.31.5.45                : ok=9    changed=3    unreachable=0    failed=0    skipped=1    rescued=0    ignored=0  

For more information, see the Aerospike Connector configuration guide

3. Place the license files

In this stage of the deployment, place the Starburst and Aerospike license files as specified.

1. Starburst: Copy the Starburst license file (starburstdata.license) to the /etc/starburst directory on the Starburst coordinator server.

Contact your Aerospike Sales Representative to get the Starburst license file.

2. Aerospike EE: Follow the instructions in Providing the Feature-Key File to place the Aerospike Feature Key file on the Aerospike instance.

4. Start the Aerospike SQL deployment

a. Update the configurations

Use the following Ansible command on the Starburst admin node to generate the configuration file for the coordinator and workers, and distribute these files to all hosts:

ansible-playbook   playbooks/push-configs.yml

When this process is complete, the final output is:

PLAY RECAP *****************************************************************************
172.31.2.13                : ok=25   changed=17   unreachable=0    failed=0    skipped=0    rescued=0    ignored=0   
172.31.5.45                : ok=25   changed=17   unreachable=0    failed=0    skipped=0    rescued=0    ignored=0  

b. Start Aerospike SQL

Use Ansible on the Starburst admin node to start the Aerospike SQL deployment:

ansible-playbook   playbooks/start.yml

When this process is complete, the final output is:

PLAY RECAP *****************************************************************************
172.31.2.13                : ok=3    changed=1    unreachable=0    failed=0    skipped=1    rescued=0    ignored=0   
172.31.5.45                : ok=3    changed=1    unreachable=0    failed=0    skipped=1    rescued=0    ignored=0   

c. Verify the deployment on Coordinator

Use the following command to show the active Starburst process and verify it is running:

ps -ef | grep starburst

Expected output:

starbur+ 13120     1 34 05:30 ?        00:00:28 java -cp /usr/lib/starburst/lib/* -server -Xmx16G -XX:-UseBiasedLocking -XX:+UseG1GC -XX:G1HeapRegionSize=32M -XX:+ExplicitGCInvokesConcurrent -XX:+ExitOnOutOfMemoryError -XX:+HeapDumpOnOutOfMemoryError -XX:-OmitStackTraceInFastThrow -XX:ReservedCodeCacheSize=512M -XX:PerMethodRecompilationCutoff=10000 -XX:PerBytecodeRecompilationCutoff=10000 -Djdk.attach.allowAttachSelf=true -Djdk.nio.maxCachedBufferSize=2000000 -Dnode.launcher-log-file=/var/log/starburst/launcher.log -Dnode.environment=production -Dlog.enable-console=false -Dlog.levels-file=/etc/starburst/log.properties -Dlog.output-file=/var/log/starburst/server.log -Dnode.server-log-file=/var/log/starburst/server.log -Dcatalog.config-dir=/etc/starburst/catalog -Dconfig=/etc/starburst/config.properties -Dnode.data-dir=/var/lib/starburst -Dnode.id=1b176fe8-6f89-5bba-b826-df4bbdb4cf99 com.starburstdata.presto.StarburstTrinoServer
root     14054 13742  0 05:32 pts/0    00:00:00 grep --color=auto starburst

5. Visit the Aerospike SQL Query Editor Admin Panel

The Aerospike SQL Query Editor Admin Panel lets you view and edit your queries using a web browser.

1. Open a browser and go to the landing page: http://[coordinator's IP]:8080/ui/insights/ide

2. Create a user (trino in the following example), then log in with that account.

3. Select Query editor.

Query editor:

Manage the deployment

You can issue the following Ansible commands from the Starburst admin node to manage the deployment.

Stop services:

ansible-playbook playbooks/stop.yml   

Restart services:

ansible-playbook playbooks/restart.yml

Uninstall Starburst Enterprise:

ansible-playbook playbooks/uninstall.yml