Platform independent gpu profiles for more efficient utilization of gpu resources

A non-transitory computer-readable medium embodying program code executable in at least one computing device that, when executed by the at least one computing device, causes the at least one computing device to:

• identify a configuration of a virtual machine that indicates that a virtual graphics processing unit (vGPU) is to be utilized in an execution of the virtual machine, the configuration comprising a graphics computing requirement for the virtual machine, wherein the configuration is platform independent;
• identify a plurality of hosts available in the computing environment to place the virtual machine, wherein each of the plurality of hosts comprises at least one of a plurality of physical graphics processing units (GPUs);
• and
• identify one of the plurality of hosts to place the virtual machine based at least in part on the graphics computing requirement.

A method, comprising:

• identifying a configuration of a virtual machine that indicates that a virtual graphics processing unit (vGPU) is to be utilized in an execution of the virtual machine, the configuration comprising a graphics computing requirement for the virtual machine, wherein the configuration is platform independent;
• identifying a plurality of hosts available in the computing environment to place the virtual machine, wherein each of the plurality of hosts comprises at least one of a plurality of physical graphics processing units (GPUs);
• and
• identifying one of the plurality of hosts to place the virtual machine based at least in part on the graphics computing requirement.