1 / 6

Title Member Names

Title Member Names. Give an overview of your design; include the final flow charts of your algorithms, and discuss how parallelism is exploited in them. Design Overview.

trubin
Download Presentation

Title Member Names

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. TitleMember Names

  2. Give an overview of your design; include the final flow charts of your algorithms, and discuss how parallelism is exploited in them. Design Overview

  3. Describe in more detail the kernels and functions you created. Trace the data flow through each kernel, describing any data marshaling where applicable. Implementation

  4. What is your testing procedure to make sure your implementation is robust and correct? For cases where the accuracy may be compromised in order to use the single-precision hardware and/or to achieve higher level of parallelism, provide an acceptance test for your output. Verification

  5. Did you achieve your performance goals? What went well with CUDA in terms of the performance gained? What hindered performance? Given enough development time, what is the best performance you think you could ultimately achieve with this application on the current generation of GPUs? What are the features in the toolset or hardware that would allow you to increase performance even more? Performance

  6. CONCLUSIONS

More Related