For batch workloads, cheap enough compute makes latency irrelevant
"Hashing It Out #70 - Akash Network - Greg Osuri" (Hashing It Out (The Bitcoin Podcast Network))
“It really comes down to price performance. So when you have a price performance metric, where the cost is insignificant, latency becomes insignificant as well… when you do batch optimality… the requirement really is driven by cost. So our thesis is the cost is exponentially lower, about like eight times, nine times lower, latency becomes less important for batch optimal workloads.” — 00:18:21
Context: Answering the HPC-background host’s challenge about distributed clusters lacking fast interconnects (quote spans into the re-transcribed 00:19:00 block) — the argument that ML/batch jobs would tolerate decentralized placement, which presaged Akash’s ML-first demand profile.