Apparatus and method for scheduling graphics processing unit workloads from virtual machines
AssigneeIntel Corporation
InventorsDAVID J. COWPERTHWAITE, MURALI RAMADOSS, ANKUR N. SHAH, BALAJI VEMBU, ALTUG KOKER, ADITYA NAVALE
Life of this patent
Estimating the term…
Abstract
In an embodiment, a system includes a graphics processing unit (GPU) that includes one or more GPU engines, and a microcontroller. The microcontroller is to assign a respective schedule slot for each of a plurality of virtual machines (VMs). When a particular VM is scheduled to access a first GPU engine, the particular VM has exclusive access to the first GPU engine. Other embodiments are described and claimed.
Claims21
1-23. (canceled)
A system comprising:
- a graphics processing unit (GPU) comprising one or more GPU engines;
- and
- a microcontroller to assign a respective schedule slot for each of a plurality of virtual machines (VMs) to access a corresponding GPU engine, wherein when a particular VM is scheduled to access a first GPU engine of the GPU, the particular VM has exclusive access to the first GPU engine.
A method comprising:
- assigning a unique memory space to each of a plurality of virtual machines (VMs) including assigning a first memory space to a first VM and assigning a second memory space to a second VM;
- and
- executing first instructions issued by the first VM exclusively on a first graphical processing unit (GPU) engine of a GPU, and executing second instructions issued by the second VM exclusively on a second GPU engine of the GPU, wherein responsive to a first memory access issued by the first VM, memory access by the first GPU engine is restricted to the first memory space, and responsive to a second memory access issued by the second VM, memory access by the second GPU engine is restricted to the second memory space.
At least one machine readable storage including machine-readable instructions that, when executed by a processor, cause the processor to:
- assign a unique memory space to each of a plurality of virtual machines (VMs) including assignment of a first memory space to a first VM and assignment of a second memory space to a second VM;
- and
- execute first instructions issued by the first VM exclusively on a first graphical processing unit (GPU) engine of a GPU, and execute second instructions issued by the second VM exclusively on a second GPU engine of the GPU, wherein responsive to a first memory access issued by the first VM memory access by the first GPU engine is restricted to the first memory space, and responsive to a second memory access issued by the second VM memory access by the second GPU engine is restricted to the second memory space.
Dependencies are inferred from the claim text.