База знаний CloupardFAQ
S3 object storage is a universal S3 object storage that is used for posting all kinds of data and quick access to them.
General information
Various types of data
In the object storage different types of data can be placed: backups, files, images and videos logs, etc
|
Compatibility with S3 API
To work with the storage and its objects, you can use API, CLI, SDK and file managers, such as CyberDuck and WinCSP
|
Protection and secure storage of the data
Data is stored in a secure way. Data safety is ensured by replication with preservation of current copies in two exemplars.
|
Versioning
For the objects you can turn on versioning and keep history of an object with the help of versions.
|
Speed
The speed of outgoing traffic is up to 500 Mbit/sec.
The speed of incoming traffic is up to 500 Mbit/sec.
|
How does an object storage work?
S3 works as a secure cloud storage with an available direct access to the objects through API via internet. It is also possible to turn on other functions that depend on the service provider.
Each S3 object consists of a unique identifier, metadata and content:
Unique identifier identifies an object and provides a direct link to access it.
Metadata represents object attributes, such as size, type and other user data.
Content represents data of any format, such as documents, multimedia or archives.
Objects can be united into buckets for a convenient organization.
Buckets also have unique identifiers.
|
Accesses and safety
Public access management
Public access to the buckets can be turned off/on for both reading and recording.
Which S3 access restriction mechanisms are there? White IP lists (preferably), firewall, IAM like in IAAS or something else?
The access is limited:
• by acces skeys
• anonymous access in three levels: no, yes, for reading, for reading and recording
|
Access to S3 service
Over HTTPS only. There are no plans to use http (without encryption).
|
Data encryption
There is data encryption (at rest / in transit).
|
Supported access policies
We support IAM/ACL/Bucket Policy access policies.
|
Parameters and restrictions
Possibility of private connection
There is no possibility of private connection (VPN, Direct Connect) to S3 in the service. This can be implemented, but it is not part of the S3 service.
|
The maximum number of files for one bucket
No restrictions
|
The maximum number of buckets
No restrictions
|
The maximum size of one bucket
We do not restrict. A client can set a limit himself or use an unlimited size
|
The maximum size of one file (during regular and multipart loading)
5 TiB for non-multipart upload
50 TiB for multipart upload
|
The limit on the number of requests per second GET, PUT, LIST, POST, DELETE, etc.
No limits
|
Replication
There is no replication yet.
|
Tariffs
Billing hourly
It means that every hour the exact amount of GB of data that is placed in the storage is tariffed.
Meanwhile traffic and requests are not tariffed.
|
Payment model
Pay as you go, for the actually consumed space, hourly billing
|
Payment for the use of space
Payment is only for the space that is actually occupied (pay-as-you-go).
The traffic and requests POST, GET, PUT, etc. Are not tariffed.
SLA and guarantees
99.99% service availability
99.99% object storage safety
|
Prices for outgoing and incoming traffic, GB
Not tariffed
|
Is inner traffic tariffed when a VM and storage are inside your infrastructure?
No
|
The price for requests GET, PUT, LIST, POST, DELETE, etc.
Not tariffed
|