The center is furnished with high computing laboratory facilities costs more than 50 lakhs.
This lab is equipped with 2 PowerEdge servers and state of art computing facilities.
Servers : 2 PowerEdge Servers each with 2 NVIDIA V100 GPUs
Configuration Details : Vega 1 (Server 1)
Processor: 2 x Intel Xeon Gold 5220 2.2G, 18C/36T, 10.4GT/s, 24.75M Cache, Turbo, HT (125W)
Graphics Card: : 2 x NVIDIA Tesla V100 16G Passive GPU
Hard Drives: : 4 x 1.2TB 10K RPM SAS 12Gbps 512n 2.5in Hot-plug Hard Drive
Memory Capacity: : 6 x 32GB RDIMM, 3200MT/s, Dual Rank
Configuration Details : Vega 2(Server 2): -do-
Server 1:
Applications Installed:: Docker
Containers Pulled:: Tensorflow, mxnet, Pytorch
Operating system:: Ubuntu 18.04.05 LTS
Server 2:
Frameworks Installed: : keras, Tensorflow
Applications installed: Jupyter Notebook
Operating system: Ubuntu 21.04 LTS (GUI Enabled)
Desktop : 20 nos
OptiPlex 3070 SFF
Intel Core i7-9700 (8Cores)
16GB (1X16GB) 2666MHz DDR4 Memory
NVIDIA Developer Kits:
NVIDIA Jetson Nano: 2 nos
NVIDIA Jetson TX2: 2 nos
Open Command prompt from Windows or Terminal from Ubuntu
----------------------------------------------------------------------------------
ssh -L <portnumber>:localhost:<portnumber> user@172.20.33.92
Example: ssh -L 8089:localhost:8089 user@172.20.33.92
If you have already created conda environment, access via :
conda activate <your name>
Example: conda activate sumod
jupyter notebook --no-browser --port=<same port number>
Example: jupyter notebook --no-browser --port=8089
Upload files and RUN code
----------------------------------------------------------------------------------
Else, create new environment and run Jupyter Notebook as Above
conda create --name <your name> python=3.9
Example: conda create --name sumod python=3.9
source ~/anaconda3/etc/profile.d/conda.sh
conda activate <your name>
Example: conda activate sumod
conda install -c conda-forge cudatoolkit=11.8.0
pip install nvidia-cudnn-cu11==8.6.0.163
pip install --upgrade pip
pip install tensorflow==2.13.*