Skip to content

prsyahmi/GpuRamDrive

master
Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?
Code

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
December 8, 2020 21:14
November 25, 2016 20:45
December 7, 2020 09:16

GpuRamDrive

Create a virtual drive backed by GPU RAM.

Screenshot

This application simply allocates a memory buffer inside GPU RAM and use it as virtual ram disk. The project is made possible by ImDisk and its proxy feature.

Using GPU RAM isn't as fast as host main memory, however it is still faster than a regular HDD. The result below is taken on my system with GTX 850M and i7-4710MQ. As IO operation is happening between the CPU and GPU, the GPU can become active from idle state and might causes system to draw more power. This merely just a PoC, user who search for this kind of solution is advised to upgrade the RAM or buy a faster storage.

Benchmark

People who interested in this might also want to check BadMemory: https://github.com/prsyahmi/BadMemory

Usage

  1. Install ImDisk Virtual drive (https://www.ltr-data.se/opencode.html)
  2. Download GpuRamDrive from https://github.com/prsyahmi/GpuRamDrive/releases
  3. Run GpuRamDrive_x64.exe or GpuRamDrive-cuda_x64.exe according to your platform

Compiling

To compile, open the solution, configure the target platform and build the project.

License

This project is licensed under MIT. See LICENSE.