On a normal block device you can change the i/o scheduler like this:
# cat /sys/block/hda/queue/scheduler
noop anticipatory deadline [cfq]
# echo anticipatory > /sys/block/hda/queue/scheduler
# cat /sys/block/hda/queue/scheduler
noop [anticipatory] deadline cfq
How can I control the setting for a NFS volume or do I just have to set it globally in the boot config with
elevator=noop
or similar?
No, this is not possible.
NFS is providing a network mount and is outside of the block-device scheduler settings.