mirror of https://github.com/minio/minio.git
fb96779a8a
This PR implements an object layer which combines input erasure sets of XL layers into a unified namespace. This object layer extends the existing erasure coded implementation, it is assumed in this design that providing > 16 disks is a static configuration as well i.e if you started the setup with 32 disks with 4 sets 8 disks per pack then you would need to provide 4 sets always. Some design details and restrictions: - Objects are distributed using consistent ordering to a unique erasure coded layer. - Each pack has its own dsync so locks are synchronized properly at pack (erasure layer). - Each pack still has a maximum of 16 disks requirement, you can start with multiple such sets statically. - Static sets set of disks and cannot be changed, there is no elastic expansion allowed. - Static sets set of disks and cannot be changed, there is no elastic removal allowed. - ListObjects() across sets can be noticeably slower since List happens on all servers, and is merged at this sets layer. Fixes #5465 Fixes #5464 Fixes #5461 Fixes #5460 Fixes #5459 Fixes #5458 Fixes #5460 Fixes #5488 Fixes #5489 Fixes #5497 Fixes #5496 |
||
|---|---|---|
| .. | ||
| README.md | ||
README.md
Introduction 
This feature allows Minio to combine a set of disks larger than 16 in a distributed setup. There are no special configuration changes required to enable this feature. Access to files stored across this setup are locked and synchronized by default.
Motivation
As next-generation data centers continue to shrink, IT professions must re-evaluate ahead to get the benefits of greater server density and storage density. Computer hardware is changing rapidly in system form factors, virtualization, containerization have allowed far more enterprise computing with just a fraction of the physical space. Increased densities allow for smaller capital purchases and lower energy bills.
Restrictions
- Each set is still a maximum of 16 disks, you can start with multiple such sets statically.
- Static sets of disks and cannot be changed, there is no elastic expansion allowed.
- ListObjects() across sets can be relatively slower since List happens on all servers, and is merged at this layer.