Hi Stas,
It is possible to have more horizontal-scale replication - i.e. adding servers - of course, at the cost of hardware which inevitably raises the question of budget -
Since the number of non-empty TPFs per dataset is finite, just more caching should do, depending of course on the change frequency of the data.
Some of the failures maybe because of parallel connection limits, I'm not sure how many parallel requests the JS client produces - it uses web workers but I haven't found how the parallelism is controlled.
I believe browsers have a politeness limit; the client code itself is limited by default to maximum 10. Might be different with HTTP/2.
Ruben