The slicing operation is consistently the most time-consuming part of the compressure process. AFAICT, it's not I/O bound, CPU bound, or memory bound. That makes me think it's just inefficiently written. Let's try to kick that off to a multiprocessing pool to see if we can get better throughput.
The slicing operation is consistently the most time-consuming part of the compressure process. AFAICT, it's not I/O bound, CPU bound, or memory bound. That makes me think it's just inefficiently written. Let's try to kick that off to a multiprocessing pool to see if we can get better throughput.