Paniikki: Computer Lab For Students

Paniikki is a cutting edge computer lab in the computer science department. It is located in T-building C106 (right under lecture hall T1). This documentation is a Paniikki cheatsheet.

alternate text

< The blue box at the entrance is Paniikki >

For more services directed at students, see Welcome, students!.

The name

Paniikki means “panic” in English which is a fascinating name as people in panic are in panic. I don’t know which comes first, the space or the emotion. Anyway, people experience the both simultaneously.

Access

Physical

You can access Paniikki in the T-building C106. It is right by the building’s main entrance (you can see it through the windows by the building’s main entrance).

Remote

You can ssh via the normal Aalto shell servers kosh and lyta. Going through them, you can then ssh to one of the Paniikki computers. Be warned, there is no guarantee that you get an empty one… if it seems loaded (use htop to check), try a different one.

You can find the hostnames of the Paniikki computers on aalto.fi.

Hardware

CPU properties

Spec

Model

Intel(R) Xeon(R) CPU E5-1650 v4 @ 3.60GHz

Architecture

x86_64

CPU(s)

12

Thread(s) per core

2

max MHz

4000.0000

Virtualization

VT-x

L1d cache

32K

L1i cache

32K

L2 cache

256K

L3 cache

15360K

Model

NVIDIA Quadro P5000

GPU properties

Spec

Core

GP104GL (Pascal-based)

Core clock

1607 MHz

Memory clock

1251 MHz

Memory size

16384 MiB

Memory type

256-bit GDDR5X

Memory bandwidth

320

CUDA cores

2560

CUDA compute capability

6.1

OpenGL

4.5

OpenCL

1.2

Near GeForce Model

GeForce GTX 1080

Memory properties

Spec

RAM

32GiB

Software

First thing first, you don’t have sudo rights on Aalto classroom machines and you can’t, because they are shared. We provide the most used SW and if you need more you could inquire via servicedesk@aalto.fi. We try to have a good base software that covers most people’s needs.

What?

How?

Python via Anaconda

module load anaconda

Python (system)

Default available

Tensorflow

in the Python environments, e.g. anaconda above

Modules

In short, module is a software environment management tool. With module you can manage multiple versions of software easily. Here are some sample commands:

Command

Description

module load NAME

load module

module avail

list all modules

module spider PATTERN

search modules

module spider NAME/ver

show prerequisite modules to this one

module list

list currently loaded modules

module show NAME

details on a module

module help NAME

details on a module

module unload NAME

unload a module

module save ALIAS

save module collection to this alias (saved in ~/.lmod.d/)

module savelist

list all saved collections

module describe ALIAS

details on a collection

module restore ALIAS

load saved module collection (faster than loading individually)

module purge

unload all loaded modules (faster than unloading individually)

There are some modules set up specifically for different courses: if you just load the environment with “module load”, you will have everything you need.

Read the details in Module environment page.

Example 1

Assume we are in Paniikki and wants to do our homework for CS-E4820 Machine Learning: Advanced probabilistic methods. In the course students use Tensorflow and Edward.

# Check available modules
$ module load courses/    # Tab to auto-complete

# Finally you will complete this
$ module load courses/CS-E4820-advanced-probabilistic-methods.lua

# Check the module you loaded
$ module list

Currently Loaded Modules:
        1) courses/CS-E4820-advanced-probabilistic-methods

# Check the packages
$ conda list    # You will see Tensorflow and etc.

# Launch Jupyter
$ jupyter notebook

# Do your homework

# You are done and want to un-load all the modules?
$ module purge

Example 2: General Python software

Need Python and general software? The anaconda modules have Python, a bunch of useful scientific and data packages, and machine learning libraries.

# Latest Python 3
$ module load anaconda

# Old Python 2
$ module load anaconda2

Example 3: List all software

You can check all other modules as well

$ module avail
alternate text

< Available modules in Paniikki as of 2018 March 8th >

You want to use Matlab?

$ module load matlab/2017b
$ matlab

Questions?

If you have any question please contact servicedesk@aalto.fi and clearly mention the Paniikki classroom in the message.