I have done some internal benchmarking on OpenROAD (actually it's a central topic of my thesis). Due to the workload breakdown in OpenROAD-flow-scripts (and I assume OpenLane), more cores are extremely useful due to TritonRoute being multi-threaded. Things can always change in the future, but 16 cores and 64 GB RAM is what I'd recommend for ~200K insts, 128 GB for ~500K insts. Memory footprint can be reduced by reducing number of threads, but that would would of course reduce performance.
Linen is a search-engine friendly community platform. We offer integrations with existing Slack/Discord communities and make those conversations Google-searchable.
