What are the features and scenarios of the Aliyun disk

　　 disk features and application scenarios

The disk is divided into local SSD disk and cloud disk, in which the cloud disk is divided into efficient cloud disk, SSD cloud disk and ordinary cloud disk. The following is an introduction to the characteristics and usage scenarios of these disks.

High Efficiency cloud Disk

SSD Cloud Disk

Common Cloud Plate

Local SSD Disk

High Efficiency cloud Disk

　　Product Features

The high efficiency cloud disk uses the solid state hard disk and the mechanical hard disk's mixed medium as the storage medium, has the following characteristic:

I/O performance: Provides up to 3000 random read and write IOPS, MBps throughput performance.

Data reliability: The common cloud Disk adopts distributed three-copy mechanism, which provides 99.9999999% data reliability.

　Performance Guidelines:

IOPS: Start 1000 ioPS, add 6 iops per GB, up to 3000;

Throughput: Start at Mbps, increase 0.1 Mbps per GB, maximum Mbps;

For example, GB-high efficient cloud disk, with 2500 of random Read and write IOPS, MBps throughput performance.

Maximum 32768 GB Capacity: a single block of efficient cloud disk provides 32768 GB storage space maximum.

Standalone mount: Efficient cloud disk supports mounting to an ECS instance of any cloud server in the same free zone.

　　Application Scenarios

MySQL, SQL Server, PostgreSQL and other small and medium relational database applications.

Medium and large scale development test applications with high and moderate performance requirements for data reliability.

SSD Cloud Disk

　　Product Features

The SSD cloud disk utilizes the distributed three-copy mechanism, can provide the stable high random I/O, high data reliability high-performance storage, has the following characteristics:

High performance: Provides up to 20000 random read and write IOPS, 256 MB/s throughput capability.

IOPS: IOPS per gigabyte: 30 random IOPS per gigabyte capacity, maximum 20000 random read IOPS performance. For example, GB of SSD cloud disk, provide 3000 IOPS performance, 334 GB SSD Cloud disk provides 10000 IOPS performance.

　　Note : The SSD cloud disk can only get the desired IOPS performance when it is mounted to an I/O optimized instance. You cannot get the desired IOPS performance when mounting to instances of non-I/O optimizations.

Throughput: SSD Cloud throughput Performance =min{50+0.5xdisk_size,256}mbps, starting at 50, increasing 0.5 Mbps per GB, with the upper limit of 256 Mbps throughput performance.

　　Performance Benchmark:

When the block size is 4kb/8kb, the IOPS can reach up to 20000;

When the block size is KB, the IOPS is up to 16300 and the throughput to the 256mb/s upper limit;

Block size of KB, IOPS Max 8150 or so, throughput to 256mb/s upper limit;

Block size is up to KB, IOPS maximum 4100 or so;

Analogy

Data reliability: SSD Cloud Disk adopts distributed three-copy mechanism, which provides 99.9999999% data reliability.

Maximum 32768 GB Capacity: Single-block SSD disk provides 32768 GB storage space maximum.

Separate mount: SSD Cloud disks support mounting to arbitrary ECS instances in the same free zone.

Application Scenarios

SSD cloud disk with stable high random I/O performance and high data reliability, suitable for the following scenarios:

PostgreSQL, MySQL, Oracle, SQL Server, and other large relational database applications.

Medium-and large-scale development test environments with high data reliability requirements.

Common Cloud Plate

Product Features

The ordinary cloud disk uses the mechanical disk as the storage medium, utilizes the distributed three copies mechanism, provides the high data reliability, has the following characteristics:

I/O performance: Provides hundreds of of random read and write IOPS capacity, maximum MB/s throughput.

Data reliability: The common cloud Disk adopts distributed three-copy mechanism, which provides 99.9999999% data reliability.

Maximum GB Capacity: a single block of ordinary cloud disk maximum supply GB storage space.

Separate mount: The ordinary cloud disk supports mounting to any ECS instance in the same free zone.

Application Scenarios

An application scenario that is suitable for data not being accessed frequently or with low I/O load, and is recommended for use with SSD cloud disks if the application requires higher I/O performance.

An application environment that requires low cost and random read/write I/O.

Local SSD Disk

Product Features

The local SSD disk takes the local SSD disk as the storage space on the physical server where the instance resides, with the following characteristics:

Low latency: typically provides microsecond-level access latency.

High random I/O: Maximum 12000 random IOPS performance.

High throughput: Provides the highest I/O throughput performance of up to MB/s.

Limited data reliability: Since the local SSD disk comes from a single physical server, the reliability of the data depends on the reliability of the physical server and the risk of single point of failure. It is recommended that data redundancy be done at the application layer to ensure the availability of data.

Max GB Capacity: a single block of local SSD disk provides up to GB of storage space.

Mount Uninstall is not supported: Separate mount uninstall is not supported due to the physical server from this site.

The CPU, memory, and the local SSD disk's lifting and lowering configuration are not supported after ordering.

Application Scenarios

The local SSD disk is characterized by low latency, high random I/O, and high throughput, suitable for redundant distributed I/O intensive applications, but not high in data reliability requirements, such as:

Distributed applications: NoSQL, MPP data warehouses, Distributed file systems and other I/O-intensive applications. Such applications themselves have distributed data redundancy capabilities, and local SSD disks can provide low latency, high random I/O, and high throughput I/O performance.

Large online Application log: Large online applications generate a large amount of log data, requiring high-performance storage, while log data does not require high storage reliability.

Swap partition as an instance: The swap space can be used in Linux when the required memory exceeds the actual allocated memory. When swap space is enabled, Linux systems can frequently swap the memory pages that are in use from physical memory to swap space, whether they are private partitions of existing file systems or swap files, and can free up space for memory pages that require fast access.