Multi-user multi-GPU render server apparatus and methods

1. A method for rendering images comprising:

a) executing on a server a render server program, where the server includes a server digital data processor and a graphics resource including one or more graphics processing units (GPUs);

b) receiving a render request at the server from one of a plurality of clients, where the render request is one of a plurality of render requests, where the render request is a request to render a data set including an image, where the plurality of render requests require a plurality of rendering time slices;

c) if the render server program determines that one or more of the plurality of rendering time slices would exceed a graphics resource limit then breaking down one or more of the plurality of render requests corresponding to the one or more of the plurality of rendering time slices which exceed the graphics resource limit, where the graphics resource limit depends on a parameter selected from the group consisting of a rendering mode of the image, a resolution of the image, a size of the image and the time required for rendering a time slice;

d) scheduling on the server one or more interleaved commands that correspond to the plurality of render requests;

e) rendering a plurality of images based on the scheduling of the one or more interleaved commands; and

f) sending from the server one or more of the plurality of images to one or more of the plurality of clients.

2. The method of claim 1 , where if the rendering mode is maximum intensity projection, then the voxels intersected by each viewing ray which intersects a broken down render request are assigned the maximum grey value of any of the voxels which the viewing ray corresponding to the screen pixel intersects.

3. The method of claim 1 , where if the rendering mode is a volume rendering technique, then for each pixel of a resulting image, a color and/or opacity is computed from sub images blended over each other in back to front order, one after the other using the formula c_result=(1−a_front)*c_back+a_front*c_front, where, a_front denotes the opacity of a front picture and c_front denotes the color of the front picture, and c_back denotes the color of a back picture.

4. The method of claim 1 , where the graphics resource limit is exceeded when a rendering time slice of the plurality of rendering time slices is greater than 100 milliseconds.

5. The method of claim 1 , where the graphics resource limit is exceeded when a rendering time slice of the plurality of rendering time slices is greater than a few hundred milliseconds.

6. The method of claim 1 , where the graphics resource limit is exceeded when a rendering time slice of the plurality of rendering time slices is significantly stalled.

7. The method of claim 1 , where the graphics resource limit is exceeded when the screen update/second rate is less than 25 per second.

8. The method of claim 1 , where the graphics resource limit is exceeded when the screen update/second rate is less than 5 per second.

9. The method of claim 1 , where the scheduling results in the plurality of render requests being processed in an alternating fashion.

10. The method of claim 9 , where multiple render requests are issued to the same graphics board simultaneously.

11. The method of claim 1 , where the scheduling of the plurality of render requests enables concurrent execution of two or more render requests.

12. A method for rendering images comprising:

a) executing, on a server a render server program, where the server includes a server digital data processor and a graphics resource including one or more graphics processing units (GPUs);

b) receiving at the server from a first client selected from a plurality of client digital data processors one or more first render requests, where the one or more first render requests require one or more first rendering time slices;

c) receiving at the server from a second client selected from the plurality of client digital data processors one or more second render requests, where the one or more second render requests require one or more second rendering time slices;

d) if the render server program determines that one or more first rendering time slices or one or more second rendering time slices would exceed a graphics resource limit then breaking down one or both one or more first render requests and one or more second render requests corresponding to the one or more first rendering time slices and/or one or more second rendering time slices which exceed the graphics resource limit, where the graphics resource limit is exceeded when a rendering time slice of one of the one or more first rendering time slices or the one or more second rendering time slices is greater than 100 milliseconds;

e) scheduling on the server one or more interleaved commands that correspond to the plurality of render requests;

f) rendering a plurality of images based on the scheduling of the one or more interleaved commands; and

g) sending from the server one or more of the plurality of images to one or both of the first client and the second client.

13. The method of claim 12 , where the graphics resource limit is exceeded when the screen update/second rate is less than 5 per second.

14. The method of claim 12 , where the scheduling of the plurality of render requests results in the processing of the plurality of render requests in an alternating fashion.

15. The method of claim 14 , where multiple render requests are issued to the same graphics board simultaneously.

16. The method of claim 12 , where the scheduling of the plurality of render requests enables concurrent execution of two or more render requests.

17. A method for rendering images comprising:

a) executing, on a server a render server program, where the server includes a server digital data processor and a graphics resource including one or more graphics processing units (GPUs);

b) receiving at the server from a first client selected from a plurality of client digital data processors one or more first render requests to render at least a first image, where the one or more first render requests require one or more first rendering time slices;

c) receiving at the server from a second client selected from the plurality of client digital data processors one or more second render requests to render at least a second image, where the one or more second render requests require one or more second rendering time slices;

d) if the render server program determines that one or more first rendering time slices or one or more second rendering time slices would exceed a graphics resource limit then breaking down one or both one or more first render requests and one or more second render requests corresponding to the one or more first rendering time slices and/or one or more second rendering time slices which exceed the graphics resource limit, where the graphics resource limit is exceeded when a rendering time slice of one of the one or more first rendering time slices or the one or more second rendering time slices is greater than a few hundred milliseconds;

e) scheduling on the server one or more interleaved commands that correspond to the plurality of render requests, where the scheduling results in the plurality of render requests being processed in an alternating fashion;

f) rendering a plurality of images based on the scheduling of the one or more interleaved commands; and

g) sending from the server one or both the first image to the first client and the second image to the second client.