An all-flash vSAN cluster contains a sufficient number of capacity devices and cache devices to support different RAID levels.
FTT=3(RAID-1 Mirroring) - This policy requires at least 4 capacity devices, where each object is mirrored across three copies of the object. This can be satisfied by the 4 nodes all-flash vSAN cluster.
FTT=1 (RAID-1 Mirroring) - This policy requires at least 2 capacity devices, where each object is mirrored across one copy of the object. This can be satisfied by the 4 nodes all-flash vSAN cluster.
FTT=2 (RAID-6 Erasure Coding) - This policy requires at least 6 capacity devices, where each object is striped across two sets of parity and data. This cannot be satisfied by the 4 nodes all-flash vSAN cluster.
FTT=1 (RAID-5 Erasure Coding) - vSAN does not support RAID-5 Erasure coding.

The RAID configuration that the administrator must use in this storage policy to achieve the best performance for the database virtual machines is RAID-6.
RAID-6 is an effective configuration for a vSAN cluster with eight hosts as it allows for two hosts to fail and still maintain protection of the data. RAID-6 is a data striping configuration with two parity bits stored across multiple disks, which improves read performance while also providing fault tolerance.
For more information, see the official VMware guide on vSAN Storage Policies: https://docs.vmware.com/en/VMware-vSphere/6.7/com.vmware.vsphere.vsan.doc/GUID-DAC07A3E-A077-4A93-A9D0-9CBB2F4F4B4D.html

The ESXi host's drives must be configured in RAID 1 to support Failures to Tolerate of 1. According to VMware's vSAN 6.7 Configuration Maximums guide, when deploying a vSAN 2-node cluster with a shared witness, the ESXi hosts must be configured in RAID 1 to support the Failure to Tolerate of 1. This is because a vSAN 2-node cluster with a shared witness will have a single point of failure, so it is important to ensure that the ESXi hosts are configured with redundancy to prevent data loss.