You can drag/drop, expand/collapse the panes to customize view of your

Offload Modeling

Summary.

Use this pane to view information about estimated acceleration of your code achieved by offloading.

View the estimated speed up of your code in elation to its original execution time in
Speed Up for Accelerated Code
pane.
View the estimated speed up achieved by offloading loops and functions that take the longest execution time in the
Amdahl's Law Speed Up
pane.
View the fraction of accelerated code relative to the total time of the original program in the
Fraction of Accelerated Code
pane.
View the number of offloaded functions/loops in the
Number of Offloads
pane.

Use this pane to view the information about host and target platforms and identify if your code is profitable to offload to a target platform. Compare Total Time of original code and Total Time of accelerated code.

Identify the factors that prevent your code from achieving better performance on a target device. The information is represented in a list and in a diagram that helps you visualize the results.

This pane enlists the top 5 functions and loops that are the most profitable to offload to a target device. Use this pane to do the following:

View top offloaded functions and loops in the
Loop/Function
tab and open them in Accelerated Regions window by double-clicking their names.
View the loop/function speed up on host and target devices in the
Speed Up
tab.
View the factors preventing the loop/function from achieving better performance in the
Estimated Bounded By
tab.
View data transfer overhead for the selected loop/function in the
Estimated Data Transfer
tab.

View the top five non-offloaded code regions, performance metrics, and the reason why they were not offloaded. Hover over a histogram in the Execution Time column to see time breakdown with color legend. Click a code region to open it in the

Non-Offloaded Regions

window. For details about reasons for not offloading and possible solutions view the

Why Not Offloaded

tab.