sheepkiller / presto-marathon-docker

On demand presto cluster with mesos, marathon and docker.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

CPU and Memory setting

premal opened this issue · comments

Not really an issue, but a question. What is the logic behind setting cpu = 2 and memory = 2G for the worker nodes?

Development was made on VMs with 4 vCPUs and 6GB of RAM, so those default settings.
Please note that presto has changed and you need at least 4GB of RAM for workers (2GB are allocated by default for internal use). Current version I use to deal with it is unsatisfactory and I hope to release an update soon to address those issues