Is there any risk in using Garage at a large scale? #812

Closed
opened 2024-04-18 09:14:13 +00:00 by zhwenh · 1 comment

Is there any risk in using Garage at a large scale? Currently, I'm using Garage to manage 5 servers node, each with 1TB of data per node, and haven't encountered any issues. However, I'm planning to use Garage to manage 300 servers node, each with around 22 disks totaling about 250TB capacity per server node. The data will be replication_factor =2, and there will be a large number of 4k small files. Can Garage support such a large scale? What are the potential risks?

Is there any risk in using Garage at a large scale? Currently, I'm using Garage to manage 5 servers node, each with 1TB of data per node, and haven't encountered any issues. However, I'm planning to use Garage to manage 300 servers node, each with around 22 disks totaling about 250TB capacity per server node. The data will be replication_factor =2, and there will be a large number of 4k small files. Can Garage support such a large scale? What are the potential risks?
Contributor

There's a max node limit right now of 256:

33c2086d9e/src/rpc/layout/mod.rs (L46)

Besides that; there's risk involved with everything in life - you'll have to test the deployment at a small scale, test your workload, gather metrics, and evaluate and repeat with larger scale.

There's a max node limit right now of 256: https://git.deuxfleurs.fr/Deuxfleurs/garage/src/commit/33c2086d9e985059574302b42560e769f14587de/src/rpc/layout/mod.rs#L46 Besides that; there's risk involved with everything in life - you'll have to test the deployment at a small scale, test your workload, gather metrics, and evaluate and repeat with larger scale.
Sign in to join this conversation.
No Milestone
No Assignees
2 Participants
Notifications
Due Date
The due date is invalid or out of range. Please use the format 'yyyy-mm-dd'.

No due date set.

Dependencies

No dependencies set.

Reference: Deuxfleurs/garage#812
No description provided.