How to spread data across two disks?

Using FoundationDB
1

I’m really new to FDB so sorry in advance if my question doesn’t make sense. Is there a way to use two disks on a single machine for FDB io/storage? I have a machine that has two ephemeral disks and io is limited to the disk size so I want to use both disks to use all io available on the virtual machine.
In the foundationdb.conf the datadir specifies the data location but not sure how to provide two locations in this case.

2

Can you use two fdb processes across the two disks? Otherwise one could stripe with LLVM. If you have enough CPU and memory on the machine, having two processes would likely improve throughput.

3

Inside /etc/foundationdb/foundationdb.conf you can add something along the line of:

[fdbserver.4500]
datadir = /mnt/disk1
[fdbserver.4501]
datadir = /mnt/disk2
4

Thanks @amirouche. I was able to do what you suggested (just for sotrage class) in my 5 nodes cluster but now get the following error when run fdbcli --exec "configure triple ssd":
ERROR: Not enough processes exist to support the specified configuration Type configure FORCE *’ to configure without this check`

Here is how 5 nodes processes are configured:
[["proxy", "proxy", "proxy", "log", "log", "log", "stateless","stateless"], ["storage", "storage", "storage", "storage", "storage", "storage", "storage", "storage","stateless"], ["storage", "storage", "storage", "storage", "storage", "storage", "storage", "storage","stateless"], ["storage", "storage", "storage", "storage", "storage", "storage", "storage", "storage","stateless"], ["storage", "storage", "storage", "storage", "storage", "storage", "storage", "storage","stateless"]]

Do you have any idea what is happening?

5

Triple redundancy (or double for that matter) require the transaction layer to have matching fault tolerance as well (you will need stateful processes across 3 nodes beyond storage for triple redundancy).