The Resource Pooling Principle
Damon Wischik, Mark Handley and Marcelo Bagnulo Braun.
Submitted to ACM/SIGCOMM CCR
[pdf]
Abstract.
Since the ARPAnet, network designers have built localized
mechanisms for statistical multiplexing, load balancing, and
failure resilience, often without understanding the broader
implications. These mechanisms are all types of resource
pooling, which means making a collection of resources behave
like a single pooled resource. We believe that the natural evolution
of the Internet is that it should achieve resource pooling
by harnessing the responsiveness of multipath-capable
end systems. We argue that this approach will solve the
problems and limitations of the current piecemeal approaches.
The Tower of Babel, by Brueghel.
The idea of resource pooling is to try to make a network perform
as well as if all its resources (links, web servers etc.) were together and
shared and interchangeable, so that "nothing will be restrained from the
users, which they have imagined to do."
Pooling capacity. The entire 36 Mb/s capacity in (left) can
be shared fairly, even though no source has access to more than two of the four links;
it is as though there were load balancing over the four links
as in (right).