@jj
Is there a way to dynamically load balance across many nodes to reduce a bottlenecked node? Parallelization introduces a lot of complexity maybe over engineering too early? I dunno maybe you all are thinking planet scale already
1 reply
0 recast
0 reaction