HCI compute

In this scenario, compute and storage run on every node, without object storage. The first three nodes are management nodes and also run the compute controller; every node runs Virtuozzo Storage and hosts VMs as a compute worker. This is a typical deployment architecture. The example below introduces the storage policies used throughout the other deployment scenarios.

Cluster layout

Compute worker	Compute worker	Compute worker	Compute worker	Compute worker
Compute controller	Compute controller	Compute controller
Virtuozzo Storage	Virtuozzo Storage	Virtuozzo Storage	Virtuozzo Storage	Virtuozzo Storage
Management	Management	Management
Node1	Node2	Node3	Node4	Node5
Bond 1 (Storage) 2 × 25 GbE LACP Non-routable, MTU 9000	Storage · Internal management
Bond 2 (Overlay + Public) 2 × 10 GbE LACP VLAN-tagged, public via NAT	VM private (VXLAN) · VM public · Admin panel · Compute API · SSH

Storage policy

VM volumes use a single NVMe all-flash tier for low latency. Each volume is protected by a storage policy, which combines the redundancy mode, storage tier, and failure domain.

Use erasure coding 3+2 with the host failure domain. EC 3+2 splits each fragment into three data pieces and two parity pieces, placed one per host, so the cluster survives the loss of any two hosts. This configuration requires five nodes, one per piece. Plan one additional spare node so the cluster can keep its redundancy and self-heal during an outage. Because this storage policy uses EC, the RAM values below include the per-node storage-client reservation described in CPU and RAM requirements. With fewer than five nodes, use replication with two or three replicas instead.

Per-node hardware

Resource	Management node		Worker node
	Minimum1 The management node minimum is based on the system reservation, storage reservation for five disks, erasure-coding storage-client reservation, and compute controller reservation, plus operating system and evaluation workload headroom. The worker node minimum is based on the system, storage, and erasure-coding storage-client reservations, plus workload headroom. See CPU and RAM requirements for the reservation values.	Recommended	Minimum	Recommended
CPU cores (physical)	16	32+	8	24+
CPU frequency	2.0 GHz	2.4 GHz+	2.0 GHz	2.4 GHz+
RAM (ECC, required)	32 GB	96 GB+	16 GB	64 GB+
System disk	120 GB SSD	2 × 480 GB SSD (RAID-1)	120 GB SSD	2 × 240 GB SSD (RAID-1)
Metadata (MDS)	On system disk (combined), first 5 nodes		On system disk (combined), first 5 nodes
Storage tier	NVMe (all-flash), ≥1 disk per node		NVMe (all-flash), ≥1 disk per node
Backend network	2 × 10 GbE (LACP)	2 × 25 GbE (LACP)	2 × 10 GbE (LACP)	2 × 25 GbE (LACP)
Frontend network	10 GbE	2 × 10 GbE (LACP)	10 GbE	2 × 10 GbE (LACP)