Dan Dougherty

Under the same setup described in OP: Using fpsync -n 8, I did see better (33% faster) times than standard rsync -a Type Time real 23m19.199s user 2m43.576s sys 8m40.236s Using rsync -a Type Time real 35m31.824s user 0m35.508s sys 1m49.452s I'll post additional times for other n if possible later.

Thanks Ganael for your rapid response. Would there be a more optimal way of syncing across the 2 machines I descibed? Would using multiple ssh workers be expected to perform better for example? I have not tried multiple ssh workers, but could try. I'm doing a comparison now of rsync and fpsync -n 8 and will report what I find. One more question, I normally run rsync with the option -a -- is this implied when using fpsync? I assume so but wonder if additionally specifying that with the -o option to...

I'm wondering what the effect of number of cores on both source and destination machines is. In my case, I'm syncing files from machine2 onto machine1 using fpsync on machine1. Machine1 has 10 cores but machine2 only has 2 cores. I'm wondering, can I still expect a speed-up by virtue of machine1 having 10 cores or will the performance be restricted by having only 2 cores on machine2? Note that in my setup machine2 is mounted on machine1 via a CIFS mount. My syntax is (currently just using 8 of 10...

I'm wondering what the effect of number of cores on both source and destination machines is. In my case, I'm syncing files from machine2 onto machine1 using fpsync on machine1. Machine1 has 10 cores but machine2 only has 2 cores. I'm wondering, can I still expect a speeup by virtue of machine1 having 10 cores or will the performance be restricted by the 2 cores on machine2? Note that in my setup machine2 is mounted on machine1 via a CIFS mount. My syntax is (currently just using 8 of 10 cores) is...