Hire more server huggers, make extra sure das-blinkenlights stay blinking, buying a second backup generator.
And- obviously enough- off site redundancy; multiple data centers. Which they talk about the difficulty of: 10Gbps pipes between DCs is not a lot compared to write throughput of a single computer's main memory, much less a DC's. For some compressible checkpoint logs there might be a way forward.
Perhaps we need a new *-cast technology: after any-cast we'll all be needing n-cast, which sends our traffic to the n-nearest DC's? :)
If it's a distributed network, data can be distributed across nodes using a RAID algorithm.
Here's the long answer to your question:
http://www.stanford.edu/~ouster/cgi-bin/papers/ramcloud-reco...
https://ramcloud.stanford.edu/wiki/display/ramcloud/Data+per...