--- title: "S3-Compatible Object Storage | CubePath" description: "S3-compatible object storage in Barcelona for backups, media and application data. Infrequent Access tier, public delivery through the CubePath CDN, pay per use." language: en source: https://cubepath.com/object-storage --- # Object Storage S3-compatible buckets in Barcelona for backups, media and application data. Use the tools you already have: just point them at eu.cubestorage.io. [Create Bucket](https://my.cubepath.com/deploy/object-storage) View Code Samples S3 ## Storage Tiers Pick a tier for each bucket. Tiers share the same endpoint and API; they differ in the disks behind them and the price. ### Infrequent Access HDD Available HDD storage with erasure coding for backups, archives, logs and media libraries. The lowest price per GiB. Storage $9.00/TiB-month $0.008789/GiB-month Egress $10.00/TiB $0.009766/GiB Class A requests $0.0060/1,000 Class B requests $0.0006/1,000 ### Standard SSD Coming soon SSD storage for data your applications read and write constantly. Endpoint https://eu.cubestorage.io Region eu, hosted in Barcelona. Path style and virtual host style both work. Free every month: 5 GiB-month of storage, 5 GiB of egress and 20,000 requests per organization and tier. Uploads (ingress) and DELETE requests are free. Traffic from a bucket to the CubePath CDN is free. Requests from the CDN edges are billed as class B requests and count towards the free tier. Any other egress, including to other CubePath services, is billed per GiB. [Full pricing](https://cubepath.com/pricing/object-storage) [Estimate your cost](https://cubepath.com/calculator) ## Built for Any Workload Object storage with the essentials to store, protect and deliver your data. ### S3 Compatible API Works with the AWS CLI, rclone, boto3 and any S3 SDK. Signature V4, multipart uploads and presigned URLs are supported. ### Public Delivery via CDN Connect a bucket to a CubePath CDN zone to serve it on cubecdn.io or your own domain. Traffic from the bucket to the CDN is free; requests from the CDN edges are billed as class B requests. ### Scoped Access Keys Create read only or read and write keys, limited to the buckets you choose and with an optional expiry date. Buckets are always private. ### Object Versioning Keep previous versions of your objects and recover from accidental overwrites and deletions. ### Erasure Coding Objects are spread across the disks and nodes of our Barcelona cluster, so they survive disk and node failures. There is no replication between sites. ### Lifecycle Rules Rules to expire old objects and versions automatically. Incomplete multipart uploads are already cleaned up after 7 days. Coming soon ## S3-Compatible API Use your favorite S3 tools and SDKs with the endpoint https://eu.cubestorage.io and the region eu. storage.py ``` 1234567891011121314151617 ``` ## Limits Clear numbers, so you can plan your uploads. 5 GiB Maximum object size in a single PUT request 10,000 Parts per multipart upload 5 MiB to 5 GiB Size of each part (the last part can be smaller) 1 TiB Storage per bucket 20 Buckets per organization 24 h Maximum lifetime of a presigned URL Uploads above the bucket quota are rejected with an InvalidRequest error. Need more buckets? Contact support to raise the limit. ## Frequently Asked Questions ### Is the API S3 compatible? Yes. Buckets speak the S3 API with Signature V4, so boto3, the AWS CLI, rclone, s3cmd and the AWS SDKs work once you set the endpoint to https://eu.cubestorage.io and the region to eu. Buckets and access keys are created from the dashboard, the API or cubecli. ### What is the maximum object size? A single PUT accepts up to 5 GiB. Larger objects use multipart uploads of up to 10,000 parts, each from 5 MiB to 5 GiB (the last part can be smaller), and every bucket holds up to 1 TiB. Most S3 tools switch to multipart automatically. ### How is my data protected? Objects are stored with erasure coding across the disks and nodes of our storage cluster in Barcelona, so a failed disk or node does not lose data. There is no replication between sites: keep a copy elsewhere if you need protection against the loss of a whole data center. Versioning protects against accidental overwrites and deletions. ### Can I make a bucket public or host a static website? Buckets are always private and there is no static website hosting. To serve files to the public, connect the bucket to a CubePath CDN zone and deliver them from cubecdn.io or your own domain. For private sharing, use presigned URLs, valid for up to 24 hours. ### Which tiers are available? Infrequent Access, on HDD, is available now in the eu region. Standard, on SSD, is coming soon. Each bucket belongs to one tier. ### How does billing work? Usage is billed hourly from your balance: average storage in GiB-month, egress in GiB and requests per 1,000 (class A for writes and listings, class B for reads). Uploads and deletes are free, and so is the traffic from a bucket to the CubePath CDN, although requests from the CDN edges are billed as class B. Any other egress, including to other CubePath services, is billed. ### Is there a free tier? Yes. Every organization gets 5 GiB-month of storage, 5 GiB of egress and 20,000 requests per month on each tier at no cost. Requests from the CDN edges count towards the free requests. ### Can I use my existing S3 tools? Yes. Any tool that supports the S3 API works, including the AWS CLI, rclone, Cyberduck, boto3 and the AWS SDK for JavaScript. ### Are lifecycle rules supported? Lifecycle rules that you manage yourself are coming soon. Today CubePath manages one rule for you: incomplete multipart uploads are aborted after 7 days, and they are not billed in the meantime. ## Ready to Store Your Data? Create a bucket and an access key from the dashboard, then connect any S3 client in minutes. [Get Started](https://my.cubepath.com/deploy/object-storage)