Parallel File Storage

Get Started

High-Performance Parallel File Storage for AI/ML Workload

Parallel File Storage is a high-performance parallel file storage that processes large volumes of data quickly and efficiently based on the All-NVMe system. Not only it has various use cases such as Al/ML and big data analysis, it uses a distributed storage system to store data across multiple storage nodes to achieve high performance data processing and analytics.

Overview

  • High Performance & Reliability

    Based on the All-NVMe system and a distributed storage of data across multiple nodes, it guarantees high performance and reliability. Regardless of the file size, it enables high performance data processing. Not to mention, it ensures reliability in case of a node failure by protecting data in another nodes.

  • Large Volume

    It can be stably expanded online, and it is highly scalable and can be used without capacity restrictions.

  • Snapshot Backup

    The image snapshot function enables recovery of modified or deleted data. Users can recover data to a specific point in time using generated snapshots.

01

04

Service Architecture

Samsung Cloud Platform
  • Multi-node GPU Cluster
      Parallel File Storage
    • Node
    • Node
    • Node
    • ...
    • Node
    • Node
    • Node

Key Features

  • Offered capacity
    1. Unlimited capacity with the minimum of 1TB
  • Snapshot backup
    1. Recover modified or deleted data using snapshots
    2. Recover data in a file format using snapshots generated at a specific point in time
  • Access control
    1. Allow access to only permitted resources in Multi-node GPU Cluster

Let’s talk

Whether you’re looking for a specific business solution or just need some questions answered, we’re here to help

Get in Touch