There are three means to access GPUs as a student in Computer Science: Lab Machines, Timeshare Machines, and the Remote Workstation Service. It is recommended to start development on the Lab Machines, and to move on to either the timeshare machines or the remote workstation service as your work progresses. Taught students are not usually given access to the UCL CS HPC service, and exceptions are only made for very specific compute requirements.
Lab Machines
There are two teaching labs on the 1st Floor of Malet Place Engineering Building, each containing a number of workstations.
Lab 1.21: 30 machines, each with 32GB RAM and an RTX 4070 Super Ti GPU.
Lab 1.05: 24 machines, each with 128 GB RAM and an RTX 3090 GPU.
Timeshare Machines
There are three timeshare machines available for GPU processing: blaze, cream, and vanilla.
blaze: 4x Titan X cards, 64GB RAM
cream: 4x Quadro RTX 6000 cards, 375GB RAM
vanilla: 4x Quadro RTX 6000 cards, 375GB RAM
Remote Workstation Service
Bookable remote access workstations containing 4070 Ti SUPERs or 4090s.