Running Parallel jobs to increase throughput
You should also consider a way to use a strategy to run more than one job in the same time frame by starting the replication from different points in the file structure which run in parallel at the same time.
For Example:
Create 3 jobs each to run at the same time IE: starting at 7PM each day and running every two hours. With an exclusion period that begins at 5:30 AM for 13 Hours and 20 Minutes. These jobs would be designed to start after normal office hours and stop before staff come in in the morning to prevent the servers and bandwidth from being too busy. The reason for the exclusion period starting at 5:30 is to prevent FRP from starting another job that would run past the start of your office hours at 8AM.
For instance lets say you have a folder structure as follows:
D:\datafiles\data1\data2\data3\data4\data5\data6\data7\data8\data9\data10
Create job #1 that begins at datafiles and put in an exclusion for *data4* in the job. It will replicate everything through data3
Create job #2 which begins at \data4 and put in an exclusion for *data8*
in the job. It will replicate all of \data4 down through \data7
Create job #3 with begins at \data8 this job will replicate \data8 and everything below it until all files below are replicated.
See the attached image for exclusion syntax.
In many cases this strategy will require you to increase the default memory available to FRP from 512mb to something higher. See the Wiki article on memory
-------
Yitzi 6-23-09