Expansion of ellipses and choice of erasure sets based on this expansion is an automated process in MinIO. Here are some of the details of our underlying erasure coding behavior.
- Erasure coding used by MinIO is [Reed-Solomon](https://github.com/klauspost/reedsolomon) erasure coding scheme, which has a total shard maximum of 256 i.e 128 data and 128 parity. MinIO design goes beyond this limitation by doing some practical architecture choices.
- Erasure set is a single erasure coding unit within a MinIO deployment. An object is sharded within an erasure set. Erasure set size is automatically calculated based on the number of disks. MinIO supports unlimited number of disks but each erasure set can be upto 16 disks and a minimum of 4 disks.
- We limited the number of drives to 16 for erasure set because, erasure code shards more than 16 can become chatty and do not have any performance advantages. Additionally since 16 drive erasure set gives you tolerance of 8 disks per object by default which is plenty in any practical scenario.
- Choice of erasure set size is automatic based on the number of disks available, let's say for example if there are 32 servers and 32 disks which is a total of 1024 disks. In this scenario 16 becomes the erasure set size. This is decided based on the greatest common divisor (GCD) of acceptable erasure set sizes ranging from *4, 6, 8, 10, 12, 14, 16*.
- *If total disks has many common divisors the algorithm chooses the minimum amounts of erasure sets possible for a erasure set size of any N*. In the example with 1024 disks - 4, 8, 16 are GCD factors. With 16 disks we get a total of 64 possible sets, with 8 disks we get a total of 128 possible sets, with 4 disks we get a total of 256 possible sets. So algorithm automatically chooses 64 sets, which is *16 * 64 = 1024* disks in total.
- In this algorithm, we also make sure that we spread the disks out evenly. MinIO server expands ellipses passed as arguments. Here is a sample expansion to demonstrate the process.
- Choosing an erasure set for the object is decided during `PutObject()`, object names are used to find the right erasure set using the following pseudo code.
```go
// hashes the key returning an integer.
func crcHashMod(key string, cardinality int) int {
Input for the key is the object name specified in `PutObject()`, returns a unique index. This index is one of the erasure sets where the object will reside. This function is a consistent hash for a given object name i.e for a given object name the index returned is always the same.
- Write and Read quorum are required to be satisfied only across the erasure set for an object. Healing is also done per object within the erasure set which contains the object.
- MinIO does erasure coding at the object level not at the volume level, unlike other object storage vendors. This allows applications to choose different storage class by setting `x-amz-storage-class=STANDARD/REDUCED_REDUNDANCY` for each object uploads so effectively utilizing the capacity of the cluster. Additionally these can also be enforced using IAM policies to make sure the client uploads with correct HTTP headers.
- MinIO also supports expansion of existing clusters in zones. Each zone is a self contained entity with same SLA's (read/write quorum) for each object as original cluster. By using the existing namespace for lookup validation MinIO ensures conflicting objects are not created. When no such object exists then MinIO simply uses the least used zone.
*There are no limits on how many zones can be combined*
```
minio server http://host{1...32}/export{1...32} http://host{5...6}/export{1...8}
```
In above example there are two zones
- 32 * 32 = 1024 drives zone1
- 2 * 8 = 16 drives zone2
> Notice the requirement of common SLA here original cluster had 1024 drives with 16 drives per erasure set, second zone is expected to have a minimum of 16 drives to match the original cluster SLA or it should be in multiples of 16.
Following pseudo code returns the correct least used zone index to upload an object.
```go
func getAvailableZoneIdx(ctx context.Context) int {
zones := z.getZonesAvailableSpace(ctx)
total := zones.TotalAvailable()
// choose when we reach this many
choose := rand.Uint64() % total
atTotal := uint64(0)
for _, zone := range zones {
atTotal += zone.Available
if atTotal > choose && zone.Available > 0 {
return zone.Index
}
}
// Should not happen, but print values just in case.
panic(fmt.Errorf("reached end of zones (total: %v, atTotal: %v, choose: %v)", total, atTotal, choose))
