Object Storage

Hyperstack Object Storage provides scalable, S3-compatible storage designed for workloads that require reliable, cost-efficient, and flexible data access, such as AI/ML datasets, logs, media, and backup files. This guide covers the key concepts and functionality of Hyperstack Object Storage. It includes instructions on how to generate access credentials, interact with storage using S3-compatible tools, manage storage buckets, and understand relevant security considerations.

For complete documentation on Hyperstack's Object Storage APIs, see the Object Storage API Reference.

Overview

Hyperstack Object Storage is built on Amazon S3-compatible technology, offering a scalable, secure, and API-compliant solution for large-scale data handling. This storage option is suitable for workloads that include backups, long-term archives, AI/ML datasets, and media delivery.

Key Benefits

• Cost Optimization: Designed for high-volume storage with an efficient pay-as-you-go pricing model.
• S3 Compatibility: Supports standard S3 API operations via access keys, making it easy to integrate with existing tools and SDKs.
• Optimized for Unstructured Data: Object storage is designed for storing large-scale unstructured data such as media, logs, and datasets.
• Efficient Metadata Handling: Each object can include custom metadata, making it easier to manage, categorize, and retrieve content at scale.

What is an Access Key?

Access keys are credentials used to authenticate programmatic access to Object Storage. Each access key consists of:

• Access Key ID (public identifier)
• Secret Access Key (used to sign requests; shown once at creation)

What is a Bucket?

Buckets are the top-level containers where your data is stored in Object Storage. Each bucket can store an unlimited number of objects and includes configuration options like region, access level, and lifecycle settings.

Regional Availability

Hyperstack's S3-compatible Object Storage is currently exclusively available in the CANADA-1 region. This refers to the location of Hyperstack's own storage infrastructure and determines where your data is physically stored. Because the service is limited to one region, geographic replication is not currently available, and regional redundancy is not supported.

While Hyperstack regions are unique to Hyperstack, many S3-compatible tools still require an AWS region string during configuration. For example, Boto3 expects us-east-1 and S3cmd uses US, even when targeting non-AWS endpoints.

S3 Endpoint

To interact with Hyperstack Object Storage via any S3-compatible tool, use the following endpoint:

ca1.obj.nexgencloud.io

Billing

Object storage is priced at $0.000019397 per GB per hour, which is approximately $14.29 per TB per month.

You can view the latest pricing for object storage by using the Retrieve Pricebook API. The response includes a resource object with the name ObjectStorage, where the value field shows the price in USD per GB of storage used.

You can monitor object count, storage usage, and hourly costs directly in the Hyperstack Console by visiting the Buckets page.

---

Getting Started with Hyperstack Object Storage

You can interact with Hyperstack Object Storage using any S3-compatible client or SDK. This guide covers the popular tools: S3cmd, Boto3 Python SDK, and MinIO Client (mc), with step-by-step instructions for each.

Click the tab below that matches your preferred tool to get started.

---

Manage Access Keys and Buckets

Generate Access Keys

1. In Hyperstack, navigate to Object Storage > Access Keys
2. Click Generate Access Key, choose CANADA-1, and click Generate
3. Copy and securely store your Access Key ID and Secret Access Key (shown only once) which are used to authenticate with Hyperstack's Object Storage.

View and Manage Access Keys

1. In Hyperstack, navigate to Object Storage > Access Keys
2. View all keys associated with your account, including:

• Access Key ID
• Region
• Date of creation

Delete Access Keys

1. Click the ⋮ menu next to the key
2. Confirm deletion by entering the last 5 characters of the access key.

View and Manage Buckets

1. In Hyperstack, navigate to Object Storage > Buckets
2. View list of existing buckets and their basic info including:
   • Name
   • Creation date
   • Region
   • Storage size
   • Number of objects
3. Click on the name of a bucket or hover over the ⋮ menu and click More Details to see:
   • Creation date – the exact timestamp when the bucket was created.
   • Region – the Hyperstack region (e.g., CANADA-1) where the bucket is hosted.
   • Per hour running cost – current hourly cost for storing objects in the bucket, calculated based on storage size used.
   • Number of objects – total count of objects stored in the bucket.
   • Total Storage Used – cumulative storage consumption in human-readable units.
   • Endpoint – the unique S3-compatible URL to access the bucket (e.g.,
     https://ca1.obj.nexgencloud.io/<bucket-name> ).

Delete a Bucket

1. From the bucket list or detail page, click the ⋮ next to the bucket
2. Click Delete, confirm via dialog
3. A deletion confirmation email will be sent

S3 Reference Documentation

For comprehensive documentation on AWS S3 commands and libraries, refer to:

• AWS S3 API reference
• AWS S3cmd S3 reference
• Boto3 S3 reference
• MinIO Client S3 reference

---

Data Protection, Performance and Policies

Encryption and Data Security

All communication with the Hyperstack S3-compatible endpoint is encrypted using TLS 1.2 or higher, ensuring secure data transmission between clients and Object Storage.

Durability and Fault Tolerance

Objects are stored using a 4+2 erasure coding scheme, which divides data into four data blocks and two parity blocks. This configuration tolerates the loss of any two blocks without impacting data integrity, ensuring high durability and fault tolerance.

Multipart Uploads

Multipart upload is supported and recommended for large files. This feature:

• Enables more reliable uploads by retrying individual parts
• Delivers faster transfers through parallel uploads
• Uses less memory for handling large objects

See the Supported S3 Actions by Tool section for example Multipart Upload commands for your preferred tool.

Performance and Usage Limits

Performance depends on network conditions:

• Transfers over the public internet are limited by the region’s external uplink, currently ~40 Gb/s aggregate in the CANADA-1 region.
• Transfers within the same region (e.g., between VMs/Clusters and Object Storage in CANADA-1) benefit from faster internal networking.