gpuDeviceTable

Table of properties of GPU devices

Syntax

tbl = gpuDeviceTable

tbl = gpuDeviceTable(props)

Description

tbl = gpuDeviceTable returns a table of properties of all GPU devices detected in your local machine. The table displays the value of the Index, Name, ComputeCapability, DeviceAvailable, and DeviceSelected properties of each GPU device detected in your system. Each row of the table contains the properties of a single GPU device.

tbl = gpuDeviceTable(props) returns a customized table of properties. Each element of props must be one of the properties returned by gpuDevice. Use this syntax to query and compare specific properties of the GPU devices in your system, such as AvailableMemory.

Examples

Compare Properties of GPU Devices

Use gpuDeviceTable to query and compare the properties of all GPUs in your system at a glance.

tbl = gpuDeviceTable

tbl =

  2×5 table

    Index           Name           ComputeCapability    DeviceAvailable    DeviceSelected
    _____    __________________    _________________    _______________    ______________

      1      "NVIDIA RTX A5000"          "8.6"               true              true      
      2      "NVIDIA RTX A5000"          "8.6"               true              false       

Both devices are available for use in this MATLAB^® session. The selected GPU device, with index 1, has a higher compute capability than the device with index 2.

Alternatively, you can determine how many GPU devices are available, inspect some of their properties, and select a device to use from the MATLAB® desktop. On the Home tab, in the Environment area, select Parallel > Select GPU Environment.

Compare Specific Properties of GPU Devices

Use gpuDeviceTable to query and compare the specific properties of all GPUs in your system.

Compare the compute capability, total memory, multiprocessor count, and availability of the GPU devices in your system.

tbl = gpuDeviceTable(["Index","ComputeCapability",...
   "TotalMemory","MultiprocessorCount","DeviceAvailable"])

tbl =

  2×5 table

    Index    ComputeCapability    TotalMemory    MultiprocessorCount    DeviceAvailable
    _____    _________________    ___________    ___________________    _______________

      1            "8.6"          2.5295e+10             64                  true      
      2            "8.6"          2.5295e+10             64                  true      

Input Arguments

props — GPU device properties

string array | cell array

GPU device properties, specified as a string array or a cell array of character vectors. Each element of props must be one of the properties returned by gpuDevice.

The variables of the output table are the properties specified by props, in the same order as provided in props.

Example: ["Name","ComputeCapability","AvailableMemory"]

Data Types: char | string | cell

Output Arguments

tbl — Table of GPU device properties

table

Table of GPU device properties, returned as a table.

The default variables of tbl are Index, Name, ComputeCapability, DeviceAvailable, and DeviceSelected. If you specify the props argument, then the output table contains only the specified properties.

Tips

• To query properties of remote GPUs, start a parallel pool using your cluster, then call gpuDeviceTable inside an spmd block.

  cluster = parcluster("myRemoteGPUCluster");
  pool = parpool(cluster);
  
  spmd
    tbl = gpuDeviceTable
  end

  For more information about using remote GPUs, see Work with Remote GPUs.

Extended Capabilities

Thread-Based Environment

Run code in the background using MATLAB® backgroundPool or accelerate code with Parallel Computing Toolbox™ ThreadPool.

The gpuDeviceTable function fully supports thread-based environments. For more information, see Run MATLAB Functions in Thread-Based Environment.

Version History

Introduced in R2021a