Calculating the Graph Throughput Using Graph Output - Calculating the Graph Throughput Using Graph Output - 2025.2 English - XD100

From the trace information in the run_summary in the Vitis Analyzer, navigate to the output port for which you want to calculate the throughput (upscale kernel in this case). Add a marker at the start of the first output sample. Click the Go to last time icon, and observe the cursor moves to the end of the last iteration. Now, click the previous transition icon to go the start of the last iteration. Add one more marker at the end, and observe the time difference as 2282.320 ns.

The number of bytes transferred is 128 samples * 4 Bytes * 7 iterations = 3584 bytes.

Throughput = 3584/2282 * e-9 ~= 1.5 GBPS.

Theoratically, AI Engine transfers four bytes per cycle (in this case, 0.8 ns). Transferring 3584 bytes of data requires 896 cycles (896 * 0.8 = 716 ns). So, the theoratical throughput is 5 GBPS.