Disks are block-level Elastic Block Storage (EBS) devices provided by Alibaba Cloud for Elastic Compute Service (ECS) and provide low latency, high performance, high durability, and high reliability. Cloud disks use a distributed triplicate mechanism to ensure 99.9999999% (nine 9's) data durability for ECS instances. If service disruptions due to hardware failures occur within a zone, data within the zone is copied to an available disk in another zone to ensure data durability and availability.
Disk categories
- Enhanced SSDs (ESSDs): ESSDs are based on the next-generation distributed block storage
architecture and utilize the 25 Gigabit Ethernet and remote direct memory access (RDMA)
technologies. Each ESSD has low latency and can deliver up to 1,000,000 random read/write
IOPS. For more information, see ESSDs.
We recommend that you use ESSDs for scenarios such as online transactional processing (OLTP) databases, NoSQL databases, and Elasticsearch, Logstash, and Kibana (ELK) distributed logs.
- Standard SSDs: Standard SSDs are high-performance disks that provide consistent high
random IOPS and high data reliability.
We recommend that you use standard SSDs for scenarios such as I/O-intensive applications, small and medium-sized relational databases, and NoSQL databases.
- Ultra disks: Ultra disks are cost-effective and provide medium random IOPS and high
data reliability.
We recommend that you use ultra disks as system disks or for scenarios such as development and testing.
- Basic disks: Basic disks are the previous generation of disks and are unavailable for purchase.
The following table describes the performance of disks of different categories.
Category | ESSD | Standard SSD | Ultra disk | Basic disk | |||
---|---|---|---|---|---|---|---|
PL3 | PL2 | PL1 | PL0 | ||||
Capacity range per disk (GiB) | 1261~32768 | 461~32768 | 20~32768 | 40~32768 | 20~32768 | 20~32768 | 5~2000 |
Maximum IOPS | 1000000 | 100000 | 50000 | 10000 | 25000 | 5000 | Several hundreds |
Maximum throughput (MB/s) | 4000 | 750 | 350 | 180 | 300 | 140 | 30~40 |
Formula for calculating the IOPS per disk | min{1800 + 50 × Capacity, 1000000} | min{1800 + 50 × Capacity, 100000} | min{1800 + 50 × Capacity, 50000} | min{1800 + 12 × Capacity, 10000} | min{1800 + 30 × Capacity, 25000} | min{1800 + 8 × Capacity, 5000} | N/A |
Formula for calculating the throughput per disk (MB/s) | min{120 + 0.5 × Capacity, 4000} | min{120 + 0.5 × Capacity, 750} | min{120 + 0.5 × Capacity, 350} | min{100 + 0.25 × Capacity, 180} | min{120 + 0.5 × Capacity, 300} | min{100 + 0.15 × Capacity, 140} | N/A |
Average single-channel random write latency in milliseconds (block size = 4 K) | 0.2 | 0.3~0.5 | 0.5~2 | 1~3 | 5~10 | ||
API parameter value | cloud_essd | cloud_ssd | cloud_efficiency | cloud |
For more information about disk performance, see EBS performance.
- System disks contain operating systems and can be created only along with instances. The lifecycle of a system disk is the same as that of the instance to which the system disk is attached.
- Data disks are used to store application data and can be created separately or along with instances.
Limits
A disk can be attached only to a single ECS instance within the same zone.
The following limits also apply.
Item | Limit | Adjustable |
---|---|---|
Permissions to create pay-as-you-go disks | To create disks within mainland China regions, you must first complete real-name verification. | N/A |
Pay-as-you-go disks | You can view the resource quota in the ECS console. For more information, see View and increase resource quotas. | N/A |
System disks on an instance | 1 | N/A |
Data disks on an instance | 16 | N/A |
Capacity of all pay-as-you-go ultra disks within an account | You can view the resource quota in the ECS console. For more information, see View and increase resource quotas. | N/A |
Capacity of all pay-as-you-go standard SSDs within an account | You can view the resource quota in the ECS console. For more information, see View and increase resource quotas. | N/A |
Capacity of all pay-as-you-go enhanced SSDs (ESSDs) within an account | You can view the resource quota in the ECS console. For more information, see View and increase resource quotas. | N/A |
Capacity of a single basic disk | 5 GiB~2000 GiB | N/A |
Capacity of a single standard SSD | 20 GiB~32768 GiB | N/A |
Capacity of a single ultra disk | 20 GiB~32768 GiB | N/A |
Capacity of a single ESSD | 20 GiB~32768 GiB | N/A |
Capacity of a single local SSD | 5 GiB~800 GiB | N/A |
Capacity of all local SSDs on an instance | 1024 GiB | N/A |
Capacity of a single system disk |
|
N/A |
Permissions to attach new local disks to instances that are equipped with local disks | New local disks cannot be attached to instances that are already equipped with local disks. | N/A |
Permissions to change configurations of instances that are equipped with local disks | Only bandwidth configurations of instances that are equipped with local disks can be changed. | N/A |
Mount points of system disks | /dev/vda | N/A |
Mount points of data disks | /dev/vd[b-z] | N/A |
Billing methods
For more information, see the Pricing tab on the Elastic Compute Service product page.
Disks support the subscription and pay-as-you-go billing methods. For more information, see Subscription and Pay-as-you-go.
- Disks created along with subscription instances or separately created for subscription instances use the subscription billing method.
- Disks created along with pay-as-you-go instances or separately created use the pay-as-you-go billing method. Storage capacity units (SCUs) can be used to offset the bills of pay-as-you-go disks.
After a disk is created, you can change its billing method. For more information, see Change the billing methods of a disk.
Related operations
The following table describes the operations that you can perform on disks.
Operation | References |
---|---|
Attach an idle pay-as-you-go disk to an ECS instance |
|
Create and use a data disk |
|
Encrypt data stored on a disk | For more information, see Encryption overview.
For more information, see the following topics:
|
Resize a system disk or data disk |
For more information, see Overview. For more information, see the following topics:
|
Replace the operating system of an instance | Replace the operating system of an instance by using a public image |
Back up data stored on a disk | |
Restore a disk to its initial state | Re-initialize a system disk |
Restore a disk to its state at a specific point of time | Roll back a disk by using a snapshot |
Detach a damaged system disk from an instance and attach the disk back to the instance after the disk is repaired | Detach or attach a system disk |
Release an instance but retain data stored on its system disk | |
Release a subscription disk that is no longer needed to minimize costs | |
Release a pay-as-you-go disk that is no longer needed to minimize costs |