We got to build a laboratory in a Virtual Environment, where you can deploy several machines to several purposes. We recommend some requirements for your laboratory.
Minimum requirements:
Recommended requirements:
Dedicated machine with graphic cards, at least one.
VMware ESX as Operating System in this machine. ESX is necessary to use GPU technology in the virtual environment.
CPU of server families.
More than 64Gb of RAM
More than 1Tb of SSD M2 disk.
In the first step, we recommend deploying a virtual machine with Ubuntu Desktop 20.04 LTS, and assign it all resources that you can.
It is important to know what resources are important for what process or use in Data Science and Machine Learning. In this case, the more resources, the faster.
Here, you have a small description of this:
CPU: The use of CPU is mainly for data preprocessing.
Storage: File transfer is blazing fast when in deep learning you are almost every time dealing with GBs of data.
RAM: A good amount of RAM should be there in a machine, but again if you have a lot of preprocessing to do, else 8 to 16
GB of it is fine.
GPU: This is the heart of your deep learning rig.