Updated on 2023-08-03 GMT+08:00

File System Types

SFS provides SFS Turbo file systems, which are further classified into SFS Turbo Standard, SFS Turbo Standard – Enhanced, SFS Turbo Performance, and SFS Turbo Performance – Enhanced.

The following table describes the features, advantages, and application scenarios of these file system types.

Table 1 Comparison of file system types

File System Type

Storage Class

Feature

Highlight

Application Scenario

SFS Turbo

SFS Turbo Standard

  • Maximum bandwidth: 150 MB/s; maximum IOPS: 5,000
  • Latency: 2 to 5 ms; maximum capacity: 32 TB
  • It is suitable for services with massive small files and services that require low latency.

Low latency and tenant exclusive

Workloads dealing with massive small files, such as code storage, log storage, web services, and virtual desktop

SFS Turbo Standard - Enhanced

  • Maximum bandwidth: 1 GB/s; maximum IOPS: 15,000
  • Latency: 2 to 5 ms; maximum capacity: 320 TB
  • Enhanced bandwidth, IOPS, and capacity

Low latency, high bandwidth, and tenant exclusive

Workloads dealing with massive small files and those requiring high bandwidth, such as code storage, file sharing, enterprise office automation (OA), and log storage.

SFS Turbo Performance

  • Maximum bandwidth: 350 MB/s; maximum IOPS: 20,000
  • Latency: 1 to 2 ms; maximum capacity: 32 TB
  • With optimized features, it is suitable for services with massive small files and services that require low latency and high IOPS.

Low latency, high IOPS, and tenant exclusive

Workloads dealing with massive small files, and random I/O-intensive and latency-sensitive services, such as high-performance websites, file sharing, and content management

SFS Turbo Performance - Enhanced

  • Maximum bandwidth: 2 GB/s; maximum IOPS: 100,000
  • Latency: 1 to 2 ms; maximum capacity: 320 TB
  • Enhanced bandwidth, IOPS, and capacity

Low latency, high IOPS, high bandwidth, and tenant exclusive

Workloads dealing with massive small files, and latency-sensitive and bandwidth-demanding workloads, such as image rendering, AI training, and enterprise OA.