How Squid selects which cache_dir to use when the response
	object will fit into more than one.

	Regardless of which algorithm is used the cache_dir min-size
	and max-size parameters are obeyed. As such they can affect
	the selection algorithm by limiting the set of considered
	cache_dir.

	Algorithms:

		least-load

	This algorithm is suited to caches with similar cache_dir
	sizes and disk speeds.

	The disk with the least I/O pending is selected.
	When there are multiple disks with the same I/O load ranking
	the cache_dir with most available capacity is selected.

	When a mix of cache_dir sizes are configured the faster disks
	have a naturally lower I/O loading and larger disks have more
	capacity. So space used to store objects and data throughput
	may be very unbalanced towards larger disks.


		round-robin

	This algorithm is suited to caches with unequal cache_dir
	disk sizes.

	Each cache_dir is selected in a rotation. The next suitable
	cache_dir is used.

	Available cache_dir capacity is only considered in relation
	to whether the object will fit and meets the min-size and
	max-size parameters.

	Disk I/O loading is only considered to prevent overload on slow
	disks. This algorithm does not spread objects by size, so any
	I/O loading per-disk may appear very unbalanced and volatile.