This topic introduces the editions of Cloud Parallel File Storage (CPFS), compares them, and helps you get started.
Introduction to CPFS
CPFS is a fully managed and scalable parallel file system from Alibaba Cloud. It is designed for high-performance computing (HPC) workloads. CPFS provides a unified namespace that allows hundreds or even thousands of compute nodes to access data concurrently. It delivers tens of GB/s in throughput and millions of IOPS, all with sub-millisecond latency.
CPFS editions
CPFS is available in the following editions:
CPFS General-purpose Edition: Ideal for HPC workloads such as AI training, autonomous driving, genomics, film and television rendering, EDA simulation, oil and gas exploration, and meteorological analysis. This edition is optimized for high throughput, high IOPS, and handling massive numbers of files.
CPFS for Lingjun: Designed for intelligent computing scenarios such as AIGC and autonomous driving.
This edition can only be used with Lingjun resources, general computing resources of PAI, Lingjun resources of the single-tenant version, and Container Compute Service (ACS). It does not support access from ECS instances.
Differences between CPFS for Lingjun and CPFS General-purpose Edition
The following table details the differences between CPFS for Lingjun and the CPFS General-purpose Edition.
Category | Comparison criteria | CPFS for Lingjun (invitational preview) | CPFS General-purpose Edition |
Positioning | Target customers | Intelligent computing workloads | HPC workloads |
Product features | Ultra-high throughput and IOPS with an end-to-end RDMA network | High throughput and IOPS | |
Purchase method | Requires whitelisting for purchase on the Alibaba Cloud console. | Available for purchase on the Alibaba Cloud console. | |
Purchase | Available specifications |
|
|
Usage | Protocol | POSIX | Interoperability between POSIX and NFS v3 protocols |
Client | Linux | Linux | |
Region |
|
| |
Capacity specifications |
To create a file system up to 20 PiB, submit a ticket. |
| |
Performance | Throughput |
To increase throughput to 2 TB/s, submit a ticket. |
|
IOPS |
To increase IOPS to 30,000,000, submit a ticket. |
| |
Average latency for a single 4 KB read | 0.25 ms |
| |
Average latency for a single 4 KB write | 0.6 ms |
| |
Elasticity | Elastic scale-out | Supported | Supported |
Elastic scale-in | Not supported | Not supported | |
Scaling method | Manual | Manual | |
Scaling step size | 10 TiB | 1,200 GiB | |
Security | ACL-based access control | Not supported | Supported |
Server-side encryption | Not supported | Supported | |
Encryption in transit | Not supported | Supported | |
Stability | Subdirectory mounting | Not supported | Supported |
Dataflow | Supported | Supported | |
Backup | Supported | Supported | |
Locally redundant storage | Supported | Supported | |
Zone-redundant storage | Not supported | Supported only for the 100 MB/s/TiB Baseline | |
Cost | Billing method | Pay-as-you-go | Pay-as-you-go |
Get started with CPFS
Action | CPFS for Lingjun (invitational preview) | CPFS General-purpose Edition | |
Learn about CPFS | |||
Activate CPFS | |||
Mount a CPFS file system | |||
Use CPFS | Query file system details | ||
Scale out a file system | |||
Protocol service | Not supported | ||
Dataflow | |||
Monitoring and alerts | |||
Access control | Not supported | ||
Log management | Not supported | ||
Data backup | |||
Zone-redundant storage | Not supported | ||
Call the API | |||