Storage redundancy

更新时间:
复制 MD 格式

CPFS General-purpose Edition offers two storage redundancy types: locally redundant storage (LRS) and zone-redundant storage (ZRS). These options help you meet different requirements for data durability and service availability.

How it works

  • Locally redundant storage (LRS)

    LRS replicates your data across multiple devices at different facilities within a single availability zone (AZ). LRS protects your data against hardware failures. However, if a disaster such as a fire or power outage causes the entire availability zone to become unavailable, your data becomes inaccessible.

  • Zone-redundant storage (ZRS)

    ZRS synchronously replicates your data across multiple availability zones (AZs) within the same region. This design ensures that your data remains accessible for read and write operations even if an entire availability zone fails. This design provides higher availability.

    This feature is available only to allowlisted users. To use this feature, submit a ticket.

Choose a storage redundancy type

Comparison item

ZRS

LRS

Supported storage classes

100 MB/s/TiB Standard

  • 100 MB/s/TiB Standard

  • 200 MB/s/TiB Standard

Supported regions

China (Hangzhou), China (Beijing)

More regions will be available soon.

China (Hangzhou), China (Shanghai), China (Qingdao), China (Beijing), China (Zhangjiakou), China (Hohhot), China (Ulanqab), China (Shenzhen), China (Heyuan), China (Chengdu), and Singapore

Recovery Point Objective (RPO)

The maximum amount of data loss that a system can tolerate after a disaster.

0

Not recoverable

Recovery Time Objective (RTO)

The time required for a system to return to an operational state after a disaster.

Minutes

Not recoverable

Service Level Agreement (SLA)

99.99%

99.9%

Durability

99.999999999% (eleven 9s)

99.999999999% (eleven 9s)

Supported protocols

NFS v3.0

NFS v3.0, POSIX

Deployment AZ

Multi-AZ deployment

Single-AZ deployment

Mount target

Region-level mount target
(No AZ attribute, supports cross-AZ access)



AZ-level mount target (Has an AZ attribute and must be in the same AZ as the file system)

Service interruption during a single-node failure

≤ 70 s

≤ 70 s

Service interruption during a single availability zone failure

≤ 70 s

Read and write services become unavailable

Read latency for a single 4 KB I/O request

3 ms to 5 ms

  • 100 MB/s/TiB: 0.8 ms

  • 200 MB/s/TiB: 0.6 ms

Maximum read bandwidth per file system

8.75 GB/s

To request a higher bandwidth, submit a ticket.

  • CPFS-NFS:

    • 100 MB/s/TiB: 20 GB/s

    • 200 MB/s/TiB: 20 GB/s

  • CPFS-POSIX:

    • 100 MB/s/TiB: 30 GB/s

    • 200 MB/s/TiB: 45 GB/s

    A maximum throughput of 100 GB/s is supported. To request this throughput, submit a ticket.

Maximum read IOPS per file system (4 KB I/O)

350,000

To request a higher IOPS, submit a ticket.

CPFS-NFS:

  • 100 MB/s/TiB: 960,000

  • 200 MB/s/TiB: 960,000

CPFS-POSIX:

  • 100 MB/s/TiB: 3.6 million

  • 200 MB/s/TiB: 7.2 million

  • A maximum of 10 million IOPS is supported.

  • To request this limit, submit a ticket.

Minimum capacity

500 GiB

3,600 GiB

Maximum capacity

Approximately 88 TiB (89,600 GiB)

1 PiB

Scaling step

100 GiB

1,200 GiB

Maximum number of files or directories per file system

1 billion

4 billion

Feature support

  • Encryption at rest

  • Encryption in transit

  • Backup

  • NFS ACL

  • Encryption at rest

  • Encryption in transit

  • Data flow

  • Backup

  • NFS ACL

Create an LRS file system

To create a CPFS file system that uses LRS, see Create a file system.

Create a ZRS file system

Prerequisites

  1. You have created a Virtual Private Cloud (VPC) and a VSwitch in the target region.

  2. Ensure that your workloads are deployed in a region that supports ZRS.

Procedure

  1. Log on to the CPFS console.

  2. On the CPFS (pay-as-you-go) page, configure the parameters for the file system.

    Parameter

    Description

    Region

    Select the region where you want to create the file system.

    The file system must be in the same region as the ECS instances that will mount it. CPFS file systems and ECS instances in different regions cannot communicate with each other.

    Storage class

    100 MB/s/TiB Baseline. The throughput of the file system is calculated as: Baseline Throughput × Capacity. For more information, see Specifications.

    Storage redundancy type

    Zone-Redundant Storage.

    Capacity

    • Minimum capacity: 500 GiB

    • Maximum capacity: Approximately 88 TiB (89,600 GiB)

    • Scaling step: 100 GiB

    Select a capacity that meets your business requirements and ensures sufficient storage space.

    Encryption type

    The method for encryption at rest. You can set this parameter only when you create the file system, and it cannot be modified afterward.

    • Not Encrypted: Data is not encrypted. This option is suitable for development and test environments.

    • Managed key: Alibaba Cloud automatically manages encryption keys. No configuration is required. This option is suitable for standard encryption requirements.

    • KMS-managed key: Use KMS to manage your own keys. This option is suitable for scenarios with high compliance requirements.

    For more information, see Server-side encryption.

    Service-Linked Role

    This Service-Linked Role is required if you select KMS-managed key. The role allows the file system to access the KMS service.

    Key ID

    This parameter is required if you select KMS-managed key.

    Resource Group

    (Optional)

    Select a resource group for resource management. If you do not select a resource group, the file system is added to the default resource group. For more information, see Best practices for designing resource groups.

    If you need to create a new resource group, click Create Resource Group. In the Create Resource Group dialog box, configure the Resource Group Identifier and Resource Group Name, and then click OK.

    Virtual Private Cloud (VPC)

    Select the same VPC as your ECS instances.

    ZRS zones

    Select three different availability zones to implement zone-redundant storage. An availability zone is a physical location within a region that has independent power and network resources.

    ZRS VSwitches

    Select a VSwitch for each availability zone.

    Service-Linked Role

    CPFS requires a Service-Linked Role to configure network resources.

  3. Click Buy Now and complete the purchase as prompted.

    Creating the file system takes 15 to 20 minutes. Afterward, it is automatically configured for zone-redundant storage and provides the following capabilities:

    • Data is automatically and synchronously replicated across multiple availability zones.

    • Automatic failover occurs if a single availability zone fails, which provides an RPO of 0 and an RTO within minutes.

    • A Service Level Agreement (SLA) guarantees 99.99% service availability.

View the storage redundancy type

  1. Log on to the CPFS console.

  2. In the left-side navigation pane, choose File System > File System List.

  3. In the top navigation bar, select a region.

  4. On the File System List page, click the target file system.

  5. On the Basic Information page of the file system, view the redundancy type of the file system.

    image

Billing

  • LRS and ZRS are billed at different rates. For pricing details, see the CPFS Pricing page.

  • Pay-as-you-go billing is supported.

Related topics