You can install a Tetra Data Hub on a network server or virtual machine that meets the following requirements. The server or VM should be dedicated to TetraScience -- for example, it should be possible for TetraScience to remotely reboot the machine without causing any impact.
64-bit versions of the following platforms are supported:
Ubuntu 18.04 or above
RedHat 7.5 or above
CentOS 7 or above
**You need admin (sudo) access of the machine
2 GHz+ processor
8 GB RAM required, 16GB RAM recommended
50 GB+ available disk space
outbound internet access
minimum 1 port should be available (each connector should have a dedicated port)
TetraScience Data Hub uses AWS SSM Agent. The instance running the Data Hub/SSM agent needs to have outbound internet access.
All communication from the SSM agent is outbound HTTPS to the SSM and EC2 messages endpoints in the region where your instances are registered. The SSM agent doesn't require any inbound ports to be opened, thus enabling inbound internet access is not required.
Each Connector installed on the same Data Hub machine will use a different port to communicate with a TetraScience Agent. Thus to install Connectors, the minimum number of available ports should be 1. If you are going to use multiple Connectors but ports are limited on one Data Hub machine due to security reasons. You need to create multiple Data Hub machines and install Data Hub on them.
You can read more about SSM agent here
The outbound internet access connects AWS services endpoints
100 Mbps network (Required)
1000 Mbps (Preferred)
4 CPUs minimum recommended
The Data Hub requires a consistent and accurate time reference. You can use existing NTP sources or Amazon Time Sync Service
Data Hub supports proxy settings. Details on how and when to configure proxy settings are given in section https://developers.tetrascience.com/docs/proxy-settings
Updated over 1 year ago