The paper is:
L. Wang, Y. Xiao, B. Shao and H. Wang, "How to partition a billion-node graph," 2014 IEEE 30th International Conference on Data Engineering, Chicago, IL, 2014, pp. 568-579.
Here's the (potentially paywalled) link to IEEExplore: http://ieeexplore.ieee.org/document/6816682/
Mods, maybe swap the links?
Traversing Trillions of Edges in Real-time:
Graph Exploration on Large-scale Parallel Machines
Originally these graph partitioning algorithms were largely designed to solve that problem.
What's the problem? Machines with 32G RAM are commonplace, you can probably just fit it all into memory. The important bits at least.
There are two assumptions implicit in your statement. First, that you have hardware available with large amounts of RAM (32GB would not be enough for many of the graphs I work with, though in fairness 128GB+ almost always would be). Second, you assume that RAM is the graph application's only bound (rather than CPU, network, etc...).
I'm not personally convinced that either assumption is sound (but then, as a PhD student nearing the end of his thesis investigating graph partitioning ... check my bias :-) ).
You might fit quite a bit into a beefy 3TB RAM server, but it might not be the best ram-to-cpu ratio for graph algorithms.
I would imagine real biological data isn't quite there yet, but it's certainly not for want of trying. Simulations might get pretty close. Blue Brain was aiming for 100M neurons and (presumably) a few orders of magnitude more connections and I think there's another simulation floating around with 10^11 neurons (but presumably less biophysical accuracy).
But yes, assuming your data is "big data" when it can easily be processed on a dev workstation or beefy server is very much in vogue right now.