Disaggregated computing for distributed confidential computing environment
An apparatus to facilitate disaggregated computing for a distributed confidential computing environment is disclosed. The apparatus includes a graphics processing unit (GPU) to: provide a virtual GPU monitor (VGM) to interface over a network with a middleware layer of a client platform, the VGM to interface with the middleware layer using a message passing interface; configure and expose, by the VGM, virtual functions (VFs) of the GPU to the middleware layer of the client platform; intercept, by the VGM, request messages directed to the GPU from the middleware layer, the request messages corresponding to VFs of the GPU to be utilized by the client platform; and generate, by the VGM, a response to the request messages for the middleware client.
1. An apparatus comprising:
a graphics processing unit (GPU) to:
provide a virtual GPU monitor (VGM) to interface over a network with a middleware layer of a client platform that is operated remote to the GPU, the VGM to interface with the middleware layer using a message passing interface, and wherein the client platform hosts userspace components of a GPU stack of the GPU, the userspace components comprising an application to utilize the GPU for acceleration of tasks of the application running on the client platform, a runtime corresponding to the GPU, a user mode driver of the GPU, and a kernel mode driver of the GPU;
expose, by the VGM, virtual functions (VFs) of the GPU to the middleware layer of the client platform;
process, by the VGM, request messages directed to the GPU from the middleware layer, the request messages originating from the application and corresponding to VFs of the GPU to be utilized by the client platform; and
generate, by the VGM, a response to the request messages for the middleware layer.
2. The apparatus of claim 1 , wherein the GPU virtualizes resources of the GPU and exposes the resources to the client platform, the resources comprising at least the VFs and memory of the GPU.
3. The apparatus of claim 1 , wherein the GPU is further to facilitate GPU attestation, GPU encryption, GPU integrity-protection, and verification of data and control messages at the GPU inside of a trusted execution environment (TEE) of the GPU.
4. The apparatus of claim 1 , wherein the runtime and the user mode driver are to prepare command buffers and data structures based on first instructions from the application.
5. The apparatus of claim 4 , wherein the command buffers and the data structures to initialize the GPU and to dispatch a workload of the application on the GPU based on second instructions from a command streamer of the GPU.
6. The apparatus of claim 1 , wherein the client platform comprises a GPU middleware layer to abstract details associated with a network connection between the client platform and the GPU, and wherein the GPU middleware layer to build a device model of the GPU based on information acquired from the GPU via the VGM.
7. The apparatus of claim 1 , wherein the VGM exposes a plurality of interfaces to the client platform, the plurality of interfaces comprises at least one of a management interface, a control interface, and a data interface.
8. A method comprising:
providing, by a graphics processing unit (GPU), a virtual GPU monitor (VGM) to interface over a network with a middleware layer of a client platform that is operated remote to the GPU, the VGM to interface with the middleware layer using a message passing interface, and wherein the client platform hosts userspace components of a GPU stack of the GPU, the userspace components comprising an application to utilize the GPU for acceleration of tasks of the application running on the client platform, a runtime corresponding to the GPU, a user mode driver of the GPU, and a kernel mode driver of the GPU;
exposing, by the VGM, virtual functions (VFs) of the GPU to the middleware layer of the client platform;
processing, by the VGM, request messages directed to the GPU from the middleware layer, the request messages originating from the application and corresponding to VFs of the GPU to be utilized by the client platform; and
generating, by the VGM, a response to the request messages for the middleware layer.
9. The method of claim 8 , wherein the GPU virtualizes resources of the GPU and exposes the resources to the client platform, the resources comprising at least the VFs and memory of the GPU.
10. The method of claim 8 , wherein the GPU is further to facilitate GPU attestation, encrypting, and integrity-protecting, and verifying data and control messages at the GPU inside of a trusted execution environment (TEE) of the GPU.
11. The method of claim 8 , wherein the runtime and the user mode driver are to prepare command buffers and data structures based on first instructions from the application.
12. The method of claim 11 , wherein the command buffers and the data structures to initialize the GPU and to dispatch a workload of the application on the GPU based on second instructions from a command streamer of the GPU.
13. The method of claim 8 , wherein the client platform comprises a GPU middleware layer to abstract details associated with a network connection between the client platform and the GPU, and wherein the GPU middleware layer to build a device model of the GPU based on information acquired from the GPU via the VGM.
14. The method of claim 8 , wherein the VGM exposes a plurality of interfaces to the client platform, the plurality of interfaces comprises at least one of a management interface, a control interface, and a data interface.
15. A non-transitory machine readable storage medium having stored thereon executable computer program instructions that, when executed by one or more processors, cause the one or more processors to perform operations to:
provide, by a graphics processing unit (GPU) of the one or more processors, a virtual GPU monitor (VGM) to interface over a network with a middleware layer of a client platform that is operated remote to the GPU, the VGM to interface with the middleware layer using a message passing interface, and wherein the client platform hosts userspace components of a GPU stack of the GPU, the userspace components comprising an application to utilize the GPU for acceleration of tasks of the application running on the client platform, a runtime corresponding to the GPU, a user mode driver of the GPU, and a kernel mode driver of the GPU;
expose, by the VGM, virtual functions (VFs) of the GPU to the middleware layer of the client platform;
process, by the VGM, request messages directed to the GPU from the middleware layer, the request messages originating from the application and corresponding to VFs of the GPU to be utilized by the client platform; and
generate, by the VGM, a response to the request messages for the middleware layer.
16. The non-transitory machine readable storage medium of claim 15 , wherein the GPU virtualizes resources of the GPU and exposes the resources to the client platform, the resources comprising at least the VFs and memory of the GPU.
17. The non-transitory machine readable storage medium of claim 15 , wherein the GPU is further to facilitate GPU attestation, encrypting, and integrity-protecting, and verifying data and control messages at the GPU inside of a trusted execution environment (TEE) of the GPU.
18. The non-transitory machine readable storage medium of claim 15 , wherein the runtime and the user mode driver are to prepare command buffers and data structures based on first instructions from the application, and wherein the command buffers and the data structures to initialize the GPU and to dispatch a workload of the application on the GPU based on second instructions from a command streamer of the GPU.
19. The non-transitory machine readable storage medium of claim 15 , wherein the client platform comprises a GPU middleware layer to abstract details associated with a network connection between the client platform and the GPU, and wherein the GPU middleware layer to build a device model of the GPU based on information acquired from the GPU via the VGM.
20. The non-transitory machine readable storage medium of claim 15 , wherein the VGM exposes a plurality of interfaces to the client platform, the plurality of interfaces comprises at least one of a management interface, a control interface, and a data interface.