Usage Limits

1. S3 API Limits

The following standards strictly follow S3 protocol standards for specification.

Item	Specification
Maximum object size	5 TiB
Minimum object size	0 B
Maximum object size for single PUT operation	Non-multipart upload: 500 GiB; Multipart upload: 5 TiB
Maximum number of parts per upload	10,000
Part size range	5 MiB to 5 GiB; last part can be 0 B to 5 GiB
Maximum number of parts returned per list parts request	10,000
Maximum number of objects returned per list objects request	1,000
Maximum number of multipart uploads returned per list multipart uploads request	1,000
Maximum length of bucket name	63 characters
Maximum length of object name	1024 characters
Maximum length of each / separated object name segment	255 characters
Maximum number of versions per single object	10,000 (configurable)

---

2. Erasure Coding Limits

EC parameters are configured based on Reed-Solomon matrix EC algorithm. Subject to actual EC parameter configuration.

Item	Specification
Maximum number of servers per cluster	Unlimited
Minimum number of servers	1
When server count is 1, minimum number of drives per server	1 (for single-node single-drive deployment, cannot provide additional reliability or availability)
When server count is 2 or more, minimum number of drives per server	1
Maximum number of drives per server	Unlimited
Read quorum count	N/2
Write quorum count	(N/2) + 1

---

3. Object Naming Limits

File System and Operating System Limits

Object names in RustFS are primarily limited by the underlying operating system and file system. For example, Windows and some other operating systems restrict the use of certain special characters such as ^, *, |, \, /, &, ", or ;.

Please refer to relevant documentation for a complete list of restrictions based on your specific operating system and file system.

RustFS recommends using Linux operating systems based on XFS file systems in production environments for better performance and compatibility.

Naming Conflict Handling

In RustFS, applications must assign unique and non-conflicting keys to all objects. This includes avoiding creating objects whose names might conflict with parent object or sibling object names. RustFS will return an empty set when performing LIST operations at locations where conflicts occur.

For example, the following operations would cause namespace conflicts:

PUT data/hello/2025/first/a.csv
PUT data/hello/2025/first # Conflicts with existing object prefix

PUT data/hello/2025/first/
PUT data/hello/2025/first/vendors.csv # Conflicts with existing object

Although you can perform GET or HEAD operations on these objects, naming conflicts will cause LIST operations executed at the hello/2025/first/ path to return empty result sets.