FCIP replication causing huge i/o load on host

[Ann Ayson]

I have two MXP110 FCIP gateways and two EVA 8100's (one at each site) that I'm trying to get set up for replication. The problem is that when I start up a replication job the LUN's being replication start hitting 100% disk i/o utilization. The replication job is going over controller A, port 1. We tried setting up the host so that all i/o for applications went over Controller B...but this made no difference. This happens in async and sync. The LUN I was attempting to replication was 300 GB in size (it took 30 hours to finish a full replication copy).

How about others out there using Replication via FCIP, do you see high i/o loads on your hosts?