IPv6 can use flow labels to "sort" traffic into separate flows according to the label number. According to RFC 1809, the notion is that by simply looking up the Flow Label in a table, the router can decide how to route and forward the datagram without examining the rest of the header. Routers can cache a labeled route for a given destination host, regardless of how many different sources are sending to that same host. The danger of caching labeled routes like this is that if too many TCP connections enter the network at the same time, with each one possessing its own cache entry, there's the potential of a cache "explosion". Discuss different methods for dealing with this type of cache explosion that is associated with flow labels. What happens if a sending device crashes, leaving behind "stale" flow labels on packets within the network? How can flow labels be used to optimize traffic flow within the network?