Deepthought
From Robin
(Difference between revisions)
(Ny side: ''Deepthought'' is a desktop computer that can be useful for simulation work. It is quite powerful and sports a GPU that can be used for deep learning experiments. == Hardware == {| CLASS=…) |
(→Hardware) |
||
Line 4: | Line 4: | ||
{| CLASS="wikitable" | {| CLASS="wikitable" | ||
|- | |- | ||
- | | CPU || | + | | CPU || Intel i7-6700k |
|- | |- | ||
- | | GPU || | + | | GPU || nVidia GTX 1080 |
|- | |- | ||
| Motherboard || | | Motherboard || | ||
|- | |- | ||
- | | | + | | SSD || 450 GB |
|- | |- | ||
- | | RAM || | + | | RAM || 64 GB |
|- | |- | ||
| Case || | | Case || |
Current revision as of 12:59, 5 December 2016
Deepthought is a desktop computer that can be useful for simulation work. It is quite powerful and sports a GPU that can be used for deep learning experiments.
Hardware
CPU | Intel i7-6700k |
GPU | nVidia GTX 1080 |
Motherboard | |
SSD | 450 GB |
RAM | 64 GB |
Case |
Software
The computer runs default Ubuntu 16.04 with ROS installed. It can run docker containers.
Getting SSH access
In order to get SSH access you must first have a user on the machine. Once this is obtained you can log in, but only using RSA key exchange. Follow the steps below to enable this.
- Generate an SSH key with ssh-keygen
- This generates two files, a private and a public file. These can be use for asymmetric cryptography. In our case, the public part will be used to authenticate you on Deepthought.
- Copy the file with the .pub extension to Deepthought
- The file should be placed in your /home/username/.ssh/authorized_keys
You should now be able to login throught SSH.
If you have several SSH keys it can be wise to edit ~/.ssh/config on your own system adding Deepthought as a new entry:
Host ip.to.deepthought IdentityFile ~/.ssh/key_file1