All GLU Analytics system runs on a separate server to the GLU.Engine. For testing purposes just a single server is needed however for Production use, an analysis of the the forecast transaction volumes, log levels and types and retention periods, system availability requirements etc. needs to be conducted in order for GLU to advise on the the sizing of the GLU Analytics Server/s.

The underlying enabling technology on which GLU Analytics runs is the Elastic Stack. Depending on client configuration choices, the three components of the Elastic Stack (ElasticSearch, Logstash and Kibana) could be deployed each to their own server.

For Testing purposes, all three components can be deployed to the same VM or server.

Server Specification (Initial)

16 GB RAM 

100 GB SSD disk (or high speed 15k RPM drives) 

1 Dual Core CPU (x86_64)

Operating System: Ubuntu (latest version 18.04 or later)

Java Virtual Machine (JVM) – Java 8

Java Prerequisites

Java installed – version 1.8 of java

$ java -version
openjdk version "1.8.0_201"
OpenJDK Runtime Environment (build 1.8.0_201-b09)
OpenJDK 64-Bit Server VM (build 25.201-b09, mixed mode)

If Java needs to be installed use this command (assuming yum is supported with the OS.)

$ yum install java-1.8.0-openjdk.x86_64

Network Prerequisites

The following ports will need to be opened.Port Server ELK serverice Area

Install Instructions for Elastic Stack

Please refer to the Elastic stack documentation for install instructions. 

ElasticStackInstallGuide

GLU.Engine logs have been tested with the following versions of Elastic Stack

Elasticsearch 7.8 [1]

Logstash 7.8 [2]

Kibana 7.8 [3]

filebeats : [4]