[ Skip to the content ]

Institute of Formal and Applied Linguistics Wiki


[ Back to the navigation ]

This is an old revision of the document!


Install marian

- only titan1 has g++-6 and gcc-6, install it from there

- make sure you use CUDA 9.0 and Boost >= 1.56.1

- export CC=`which gcc-6`

- export CXX=`which g++-6`

- follow marian's instructions (git clone, cmake, make)

Servers with GPU units

GPU cluster gpu.q at Malá Strana:

machine GPU type cc GPUs GPU RAM Speed category Comment
iridium Quadro K2000 cc3.0 1 2 GB terribly slow
titan-gpu GeForce GTX Titan Z cc3.5 2 6 GB medium
twister1; twister2; kronos Tesla K40c cc3.5 1 12 GB slow
dll1; dll2 GeForce GTX 1080 cc6.1 8 8 GB best
titan GeForce GTX 1080 cc6.1 1 8 GB best
dll3; dll4; dll5 GeForce GTX 1080 Ti cc6.1 10 11 GB best dll3 has only 9 GPUs since 2017/07
dll6 GeForce GTX 1080 Ti cc6.1 3 11 GB best

Desktop machines:

machine GPU type cc GPUs GPU RAM Comment
victoria; arc GeForce GT 630 cc3.0 1 2 GB desktop machine
athena GeForce GTX 1080 cc6.1 1 8 GB Tom's desktop machine

[ Back to the navigation ] [ Back to the content ]