In my opinion IP Hash often causes a more complicated setup in regards to the physical switches. Since the link aggregation standard defines that the aggregation channel must only be between two switches this makes things more challenging. In your case the two switches are one vSwitch and one physical switch, i.e. you can not by default connect a IP Hash NIC team to two physical switches, only one. Which means much lower fault tolerance.
In some cases you could configure the physical switches to handle this anyway with different vendor propriarity solutions. It might be possible on the HP 2910 to set up what is called a HP "DT-trunk", but I must check this. Note that this is a bit more complicated and could make the network less stable.
Note also that the standard vSwitch does not support any kind of LACP, only the Distributed vSwitch does - which you maybe can use depending on your vSphere license.
