vCPU 384 512 640 768 1024 1280 1536 2048 2560 3072 3584 4096 8192 10240 12800 Number of vCPUs (AMD EPYC ROME based) in your Cluster.

Memory 768 GB 1.5 TB 2 TB 2.5 TB 3 TB 3.5 TB 4 TB 5 TB 8 TB 10 TB 15 TB 20 TB 25 TB Gigabyte of Memory in your Cluster

Persistent NVME Storage None 8 TB 16 TB 30 TB 60 TB NVME Flash redundant PVC Storage (rook.io)

Interconnect Speed 20 Gbit/s 200 Gbit/s Interconnect speed between worker nodes (always redundant)

Internet Uplink Capacity 20 Gbit/s 40 Gbit/s 100 Gbit/s Combined Tier-1 Transit and Exchange Outbound Bandwidth Capacity

Internet Uplink Commitment 1 Gbit/s 2 Gbit/s 3 Gbit/s 4 Gbit/s 5 Gbit/s 10 Gbit/s 15 Gbit/s 20 Gbit/s 30 Gbit/s 40 Gbit/s 50 Gbit/s Bandwidth Commited in 95% billing / monthly

AWS DirectConnects None 10 Gbit/s 2x 10 Gbit/s AWS DirectConnect Uplinks to next close proximity AWS Location

IPSec VPN Throughput 500 Mbit/s 1 Gbit/s 2 Gbit/s 3 Gbit/s 5 Gbit/s IPSec Uplink guaranted throughput for remote links to datacenters and aws/google/azure

Expense Type CAPEX (own hardware) OPEX (1 year) OPEX (2 year) OPEX (3 year) Wether you want to buy the hardware and only rent rackspace/bandwith or rent everything with commitment