The Timeline Trace shows the data transfer from the host to the FPGA and back to the host as they appear. The Timeline Trace can be visualized so that the transfer from the host to the FPGA and the FPGA compute and transfer from the FPGA to host occur sequentially.
There is a sequential execution of operations starting from the data transferred from the host to the FPGA, followed by compute in the FPGA and transferring back the results from the FPGA to the host.
At any given time, either the host or FPGA has access to the DDR. In other words, there is no memory contention between the host and kernel accessing the same DDR.
Using a single buffer will create a kernel itself with lowest latency and most optimized performance.