diff --git a/docs/deployment-preparation/hardware-requirements.md b/docs/deployment-preparation/hardware-requirements.md
index 5fcd2dd5..951da1c9 100644
--- a/docs/deployment-preparation/hardware-requirements.md
+++ b/docs/deployment-preparation/hardware-requirements.md
@@ -14,11 +14,10 @@ network bandwidth, and free space on the boot disk.
| Node Type | vCPU(s) | RAM (GB) | Locally Attached Storage | Network Performance | Free Boot Disk | Number of Nodes |
|--------------|---------|------------------------|-----------------------------------|---------------------|----------------|------------------|
-| Storage Node | 8+ | 6+ DDR4 (1) | 2x dedicated NVMe (2) | 10 GBit/s | 10 GB | 3 (3) |
+| Storage Node | 8+ | 6+ DDR4 (1) | 2x dedicated NVMe (2) | 10 GBit/s | 10 GB | 3 (3) |
-1 Simplyblock highly recommends DDR5 memory on storage nodes for optimal performance.
-2 It is possible to test with only one dedicated NVMe, but this is not approved for production.
+2 Test setups require a minimum of 1 nvme. From 26.3., we support non-nvme device clusters (any SATA or SAS linux block device), this feature is still experimental.
3 The required number of nodes is only valid for erasure coding scheme 1+1.
@@ -49,16 +48,13 @@ simplyblock data plane (spdk_80xx containers) and the rest will remain under con
Simplyblock auto-detects NUMA nodes. It will configure and deploy storage nodes per NUMA node.
Each NUMA socket requires directly attached NVMe devices and NICs to deploy a storage node.
-For more information on simplyblock on NUMA, see [NUMA Considerations](numa-considerations.md).
+detected configuration. This file is later processed when adding the storage nodes to the storage host.
+Manual changes to the configuration are possible if the proposed configuration is not applicable.
-It is recommended to deploy multiple storage nodes per storage host if there are more than 32 cores available
-per socket.
+For more information on simplyblock on NUMA, see [NUMA Considerations](numa-considerations.md).
During deployment, simplyblock detects the underlying configuration and prepares a configuration file with the
recommended deployment strategy, including the recommended amount of storage nodes per storage host based on the
-detected configuration. This file is later processed when adding the storage nodes to the storage host.
-Manual changes to the configuration are possible if the proposed configuration is not applicable.
-
### Hyper-Converged Sizing Guidance
As hyper-converged deployments have to share vCPUs, it is recommended to dedicate 15%-20%, but not less than