🎉 🎉 RAPIDMINER 9.10 IS OUT!!! 🎉🎉

Download the latest version helping analytics teams accelerate time-to-value for streaming and IIOT use cases.

CLICK HERE TO DOWNLOAD

Sizing Recommendation, Minimum Specification of node, vCPU, RAM, and Storage for Production

HanamianHanamian Member Posts: 1 Newbie
edited December 2021 in Help
Hi,
Do we need these Nodes to use RapidMiner in Production? and is that true recommendation value for minimum specification?

2 RapidMiner Server (Central Node) - 8vCPU - 32GB RAM - 1000GB Storage
2 Job Agent Node - 4vCPU - 32GB RAM - 1000GB Storage
1 Load Balancer (Sticky session) - Can use existing LB
1 Shared Database - Can use existing
1 Shared file system (NFS) - Can use existing
1 External ActiveMQ Broker - 4vCPU - 8GB RAM - 50GB Storage

Thank you.

Anyone can help?
Tagged:

Answers

  • aschaferdiekaschaferdiek Employee, Member Posts: 69   RM Engineering
    Hi @Hanamian . Standalone native requirements are listed in our requirements page. When you like to use docker as underlying deployment technology instead of deploying natively, then please consider looking at the Docker deployment section in addition to that.
    Furthermore, please have a look at the High Availability pages (here and here) when you consider such a setup.

    Generally speaking, the hardware requirements depend on your use-cases and goals, e.g. what you like to do, how many users , executins/jobs and data is expected. The obvious requirement is free disk space though: if you have a lot of different Projects or data, you should always have proper disk space left.
    If not using high availability setup, you don't necessarily need to externalize the ActiveMQ broker. In addition to that, assigning a lot of resources to the broker (CPU and memory like you proposed), might be too much. This also depends on your use-cases and how frequently jobs are scheduled and how many Job Agents are connected.

    Hope that helped.
Sign In or Register to comment.