Divide and conquer GC algorithms

Reading Richard's excellent book, I was surprised to observe that none of
the major GC algorithms seem to be hierarchical. Some algorithms like the
Train collector, Beltway and Immix amortize costs by grouping heap-allocated
blocks together but they don't seem to use trees to gradually refine the
granularity.

I've also been looking at "dynamic" graph algorithms for the first time.
These algorithms allow a small change in the graph (add/remove one
edge/vertex) to be converted into a small change in their output. The
concept of a topology tree, a data structure used in some dynamic graph
algorithms, caught my attention because this conveys information about the
topology of a graph. For example, you might be able to determine quickly
that a 1/4 partition of the heap contained pointers to another 1/4 partition
but not the other way around so there were no cycles spanning the two
partitions. Could a GC use a topology tree to do some of its work in O(log
n) time? Furthermore, what if a dynamic graph algorithm were used to
incrementally update the GC's topology tree as the write barrier mutates the
topology of the heap?

Do any GC algorithms use divide-and-conquer, trees or yield complexities
logarithmic in the number of pointers in the heap?  Can we estimate how big
a heap would have to be before hierarchical techniques would become
competitively performant?

This reminds me of the debate surrounding the use of trees in file systems
as opposed to flat collections...

On the exact opposite end of the spectrum, has anyone tried using GPGPU to
accelerate garbage collection, e.g. the mark phase?

-- 
Dr Jon Harrop, Flying Frog Consultancy Ltd.
http://www.ffconsultancy.com