Skip to content

Latest commit

 

History

History
56 lines (43 loc) · 2.36 KB

getting-started.md

File metadata and controls

56 lines (43 loc) · 2.36 KB
Raw

Getting Started

Prerequisites

  • Hardware Requirements
    • At least 8G of memory
    • SSD Hard Drive
  • Software Requirements
    • Java 17
    • HTTP Proxy software recommended to provide SSL (e.g. Nginx or Apache 2)
    • Maven 3 (optional - to rebuild the software)
  • Other Resource Requirements

More on Memory Requirements

As a minimum Snowstorm should have 2G and Elasticsearch should have 4G of memory in order to import a Snapshot and perform ECL queries. Elasticsearch will work best with another 4G of memory left free on the server for OS level disk caching.

Setup

Install Elasticsearch

Download and install Elasticsearch version 8 (tested against 8.11.1).

When running snowstorm locally, you can disable security in Elasticsearch by setting xpack.security.enabled to false in the config/elasticsearch.yml file:

xpack.security.enabled: false

See Elastic upgrade guide, if you need to migrate an existing Elasticsearch 7.x cluster to version 8.x.

Update the configuration file config/jvm.options with the memory options -Xms4g and -Xmx4g. See Set JVM Options.

Get Snowstorm Application Jar

Download the latest release jar from the releases page.

Alternatively if you would like to build Snowstorm from source code:

  • Docker must be running because unit tests use the TestContainers framework. The latest version of Docker is recommended.
  • Then use maven: 
    mvn clean package
  • Maven creates the jar file in the 'target' directory.

Start Snowstorm

First start Elasticsearch from wherever it has been installed.

./bin/elasticsearch

On the first run of Snowstorm the SNOMED CT data may need to be loaded. Follow instructions here.

On subsequent runs just start Snowstorm (in read only mode).

java -Xms2g -Xmx4g -jar target/snowstorm*.jar --snowstorm.rest-api.readonly=true