Back in August, we announced a new performance pod type - p2 - in public preview. Different from the existing p1 pods, p2 pods use a new graph-based index to provide blazing fast search speeds (up to 10x faster than p1) and higher throughput (up to 200 QPS) per replica.
This is a companion discussion topic for the original entry at https://www.pinecone.io/learn/pods-for-performance/