Originally posted by varunsh-xilinx January 25, 2023
Having shared memory as a means to get data in and out of the inference server can be an enhancement for this use case where the C++ API is used locally to pass data from another process to the inference server and back. This kind of approach could be used between the inference server and the workers too (#117)
Originally posted by varunsh-xilinx January 25, 2023 Having shared memory as a means to get data in and out of the inference server can be an enhancement for this use case where the C++ API is used locally to pass data from another process to the inference server and back. This kind of approach could be used between the inference server and the workers too (#117)