Does mongo cluster need to do raid??

does your mongoDB sharding+replication cluster do Raid??

our cluster has 6 ssd.

each ssd starts a mongod, and then makes a replication with the same disk on the other two servers.

personal understanding of mongoDB is that the high reliability of data is solved through mongo"s own replication.
read and write speed is resolved through sharding.

it would be a waste for mongo to raid, the disk.

Oct.30,2021

simple answer: official suggestion is RAID10.

For optimal performance in terms of the storage layer, use disks
backed by RAID-10. RAID-5 and RAID-6 do not typically provide
sufficient performance to support a MongoDB deployment. Avoid RAID-0
with MongoDB deployments. While RAID-0 provides good write
performance, it also provides limited availability and can lead to
reduced performance on read operations, particularly when using
Amazon's EBS volumes.

how to understand? Although it has something to do with the actual stress situation, for software such as a database, the bottleneck usually appears first on Icano. Depending on rereading or rewriting, RAID0 (limited read speed) or RAID1 (limited write speed) has its own limitations, so RAID10 is the best choice.
Sharding and Replication also provide functions similar to RAID0 and RAID1, but they are upper-level implementations, and their performance is lower than that of low-level implementations such as RAID0/1. After all, there will be additional overhead such as network. The weakness of the latter is that it cannot scale horizontally by adding more servers.
where possible, vertical scaling is achieved first through RAID10, which is the best choice for efficiency or reliability and maintenance complexity. After all, why should it be divided into several machines for things that can be done by one machine? It is only necessary to consider horizontal expansion when a server cannot be satisfied for reasons such as cost.

Menu