As suggested in earlier message, this pricing is only for single node. imo, this needs to be compared to cluster mode - any company with meaningful scale will need reliabilty and needs to run in cluster mode since you’d need your observability systems to be up and running when the rest of your systems are down.