Graphics Card Processing



What is a GPU and what does it stand for?

GPU is an abbreviation for graphics processing unit. Graphics processing unit is another name for the graphics card inside your computer.



What are the advantages of running World Community Grid on a computer's GPU?

GPUs can perform certain types of computations significantly faster than the central processing units (CPUs) that power computers. When the manufacturers of GPUs realized this, they extended the capabilities of some GPUs so that they could run tasks that were written in a specialized development language. As a result, applications that are modified to use this specialized development language–and can leverage GPUs' specialized capabilities appropriately–can potentially run many times faster on GPU than on the CPU. 

If you run a World Community Grid project that is modified to use GPU power on a device that has a GPU, and you adjust your settings accordingly, that device can complete work units more quickly than a device using just a CPU. 

However, GPU processing is different than CPU processing on World Community Grid. Specifically, GPUs don't have operating system capabilities that allow us to ensure that running work units doesn't interfere with your use of your computer. Many GPUs can run World Community Grid without any interference; however, some will cause your computer to be slow in responding to you. GPU usage for World Community Grid is therefore opt-in only; by default, World Community Grid will not use GPUs. You can opt in through the Device Profiles page - please refer to "How do I configure my settings so that World Community Grid uses my device's GPU?" for details.



Which World Community Grid research projects can run on GPUs?

For the latest on which World Community Grid research projects run on GPU, please refer to our system requirements.



Which GPUs can participate in OpenPandemics - COVID-19?

GPU work units for OpenPandemics - COVID-19 are designed to run on OpenCL version 1.2 and above. However, there are certain cards that still have issues due to having GPU drivers that aren't 100% compatible with OpenCL 1.2. Most of the issues are with cards that were released before 2016. Please check our GPU forum for a list of GPUs that are known to not work. 



How do I find out what type of GPU is installed in my computer?

To see which GPUs BOINC recognizes on your computer, please view the top lines of your Event Log by following these instructions:

  • Open the World Community Grid/BOINC Manager software application.
  • From the top main menu, select Tools and then Event Log and scroll to the top.

If your computer has been on for a long time, restarting it will allow you to view the top of the event log more easily. 



How do I configure my settings so that World Community Grid uses my device's GPU?

By default, World Community Grid will not use your device's GPUs. To allow World Community Grid to use your GPUs:

  • Go to Device Profiles and select the relevant profile. (If you're not sure which profile relates to the device on which you want to run GPU work, refer to your Device Manager list).
  • Ensure that the selected device profile is a Custom Profile.
  • Scroll through the Advanced Options to the 'Graphics Card Usage' section.
  • In order to get GPU work, select "Yes" for the type of GPU card (AMD/ATI, Intel or NVIDIA) you wish World Community Grid to use. You can set all three types (AMD/ATI, Intel, NVIDIA) as "Yes" if you do not know what card you have or wish to run GPU work on all cards available in a computer. 

Another option is "Do work on my graphics card while computer is in use?". This is an advanced option and is set to "No" be default, as in some cases choosing "Yes" can cause the device performance to appear sluggish while the GPU is used for mouse movements and other system operations.



How do I set my preferences so that World Community Grid will use my GPU while I am actively using my computer?

Once you have allowed World Community Grid to use your GPU, we only use your GPU while you are not actively using your computer. 

If you want World Community Grid to use your GPU at all times, you must manually enable this option by going to the Device Profile page under the Custom Options section. This option is labeled "Do work on my graphics card while computer is in use?". You may select either "Yes" or "No."



Will I be able to select to run work units only on my CPU, or only on my GPU, or both?

There are separate and independent settings for World Community Grid to run on either your CPU or on your GPU.

By default, World Community Grid is set up to only run work using your CPU and not using your GPU. The CPU preference is available on the Device Profiles page under the custom options section. This option is labeled "Allow research to run on my CPU?" You may select either "Yes" or "No."

To run World Community Grid work units on your GPU, refer to "How do I configure my settings so that World Community Grid uses my device's GPU?" for details.

 



If I have more than one GPU, can I limit which one is used?

Yes. In order to do this, you can make these changes in your Device Profile page by selecting which vendor they would like to use. You can choose between AMD/ATI, Intel, and NVIDIA.



If I have more than one GPU on my computer, how do I let World Community Grid run on all of my GPUs?

In order to do this, you have to use the <use_all_gpus> tag in the cc_config.xml file. Instructions to do this are available here.



What if my computer stutters while World Community Grid is using my GPU?

Due to the nature of running research applications, it is possible that running World Community Grid on your GPU while you are using your computer could interfere with the normal use of the device. If you encounter this, then we suggest that you try the option to only allow World Community Grid to use your GPU when you are not actively using your computer. This option is available on the Device Profile page under the custom options section. This option is labeled "Do work on my graphics card while computer is in use?". Select "No" and save.



How will statistics be updated for GPU participation?

World Community Grid will award run time based on the elapsed time from when a work unit is started and when it finishes running. Credit and result counts will be awarded in the same manner as CPU participation.



Will there be a badge for GPU participation?

No. Your contributions using your GPU will contribute to a specific research project and count towards earning a badge for that project.



Will you make GPU processing available for other research projects?

Not all algorithms are suited to general purpose computing on GPUs and may not receive performance improvements. Additionally, general purpose computing on a GPU is a complex process, requiring numerous modifications and optimizations to the underlying algorithms of an application to realize performance increases.

We will work with the research scientists to make projects available on GPU if feasible and applicable to advancing the science application. However, it is up to the research scientists to enable their application to utilize GPU technology.



Does the GPU application include a graphic screen saver showing what is running?

No, the GPU application does not include a screen saver. If you have selected to run work units on your GPU, which is the primary display card, we recommend that you select the least intensive screen saver available, such as the "blank" screen saver option.



What's the benefit of running OpenPandemics - COVID-19 work units for GPU?

AutoDock-GPU (AD-GPU) could increase the project's chances of finding a molecule with anti-viral properties even further. 

Compared to AutoDock 4 (the current CPU version of the software that is used for OpenPandemics - COVID-19), AD-GPU is much faster, which should provide a nice boost to the already amazing rate of docked results. 

Furthermore, AD-GPU has an improved search algorithm that exhibits a greater probability of finding strong interactions between the molecules and viral proteins, and is well suited to dock larger or more complex molecules. This means that scientists can use AD-GPU to not only screen more molecules, but also to enable the search of more complex molecules.