Our catalog backup usually process's 800 gb of images daily(1.4-1.5 mill files). Lately through observation of the job in the java console, it has been noted that during the execution of the catalog backup, the throughput dip's significantly while processing the images for the SQL type backups. The catalog will start out and avg about 28-30 mb/s until it begins processing the SQL images, where is throughput dips to about 14-16 mb/s. After the catalog backup completes the SQL images, you will observe the throughput begin to increase again. I am aware that the KB/s is an average for the job, so I would think that the impact during the sql images backup would be lower than displayed in the console
Please provide a starting point or direction to go and help me find out why this is happening. thank you
