Parallel File Storage

Service Overview

Parallel File Storage is a All NVMe-based high-performance parallel file storage that can process large amounts of data quickly and efficiently.
Storage that can be used in various fields such as AI/ML and big data analysis, distributes data across multiple storage nodes to improve data processing speed and reduce analysis time.

Features

• High Performance and Reliability: Distributes data across multiple NVMe-based nodes to provide high performance and reliability. High-performance processing is possible regardless of file size, and even if a single node fails, data is safely maintained through other nodes.
• Large-capacity volume: It can be reliably expanded while online, and its scalability is excellent, allowing use without capacity limits.
• Snapshot Backup: Through the image snapshot feature, recovery of changed and deleted data is possible. Recovery is performed by using the snapshot created at the point in time you want to recover.

Diagram

Provided Features

Parallel File Storage provides the following features.

• Volume Name: Users can set names for each volume.
• Capacity: Volumes can be created with capacities ranging from a minimum of 1TB to a maximum of up to 1,000TB.
• Connected Resource: Can be connected and used in a Multi-node GPU Cluster.
• Snapshot: Through the image snapshot feature, recovery of changed and deleted data is possible. Users select a snapshot created at the point in time they wish to recover from the list to perform the recovery.

Components

Volume

Volume (Volume) is the basic creation unit of the Parallel File Storage service and is used as data storage space. Users create a volume by entering a name and capacity, then connect it to one or more Multi-node GPU Clusters for use.
The volume name creation rules are as follows.

• It must start with a lowercase English letter and can be set to 3 to 21 characters using lowercase letters, numbers, and the special character (_).

Snapshot

A snapshot is an image backup of a volume at a specific point in time. Users can view the snapshot name and creation date in the snapshot list to select the snapshot they want to restore, and can recover data that was changed or deleted using that snapshot.
The notes for using snapshots are as follows.

Preceding Service

This is a list of services that must be pre-configured before creating the service. For details, refer to the guide provided for each service and prepare in advance.