Overview
The cloud server ECS disks can be used separately and in combination to meet the needs of different scenarios. Mainly divided into local SSD disk and cloud disk.
After reading this chapter, you can choose to use the appropriate data storage options, depending on your requirements.
For operational steps on how to mount a disk, see mount a data disk.
Cloud Disk
The cloud disk provides data block level data storage for ECS instances, and uses the distributed mechanism of three replicas to provide 99.9999999% data reliability guarantee for ECS instances.
According to the performance of different, the cloud can also be divided into ordinary cloud disk, efficient cloud disk and SSD and other types of cloud disk.
Common cloud disk for low I/O load scenarios, providing hundreds of IOPS of I/O performance for ECS instances.
High-efficiency cloud disk for moderate I/O load applications, providing an ECS instance with up to 3000 random IOPS storage performance.
SSD Cloud disks provide stable, high random IOPS performance for I/O intensive applications.
Local SSD Disk
The local SSD disk comes from the local storage of the physical machine where the instance resides. This class stores a block-level data access capability for instances with low latency, high random IOPS, and high throughput I/O capabilities.
Using local SSD disks, note the following issues:
Because storage space is from the server's local SSD disk, there is a single point of failure risk. It is recommended that data redundancy be done at the application layer to ensure the availability of data.
The CPU, memory, and the local SSD disk's lifting and lowering configuration are not supported after ordering.
Disk parameter comparison
Parameters |
SSD Cloud Disk |
High Efficiency cloud disk |
Common Cloud Plate |
Local SSD Disk |
Maximum capacity |
32768 GB |
32768 GB |
GB |
GB |
Max IOPS |
20000 |
3000 |
Hundreds |
12000 |
Maximum throughput |
256 MBps |
MBps |
20-40 MBps |
250-300 MBps |
Performance Calculation formula |
iops=min{30* capacity, 20000} Throughput =min{50+0.5* capacity, 256}mbps |
iops=min{1000+6* capacity, 3000} Throughput =min{50+0.1* capacity, 80}mbps |
does not apply |
does not apply |
Access Time delay |
0.5-2 ms |
1-3 ms |
5-10 ms |
0.5-2 ms |
Data reliability |
99.9999999% |
99.9999999% |
99.9999999% |
Physical machine reliability only, no SLA guarantee |
API Name |
Cloud_ssd |
Cloud_efficiency |
Cloud |
Ephemeral_ssd |
Price |
1.0 Yuan/gb/months |
0.5 yuan/gb/months |
0.3 Yuan/gb/months |
0.8 yuan/gb/months |
Typical application Scenarios |
- I/O intensive applications
- Medium and large relational databases
- NoSQL Database
|
- Small and medium sized databases
- Large scale Development test
- Web server Log
|
Scenarios that are not frequently accessed or have low I/O load |
Hadoop, NoSQL and other distributed applications, the application itself has a very high reliability, the need for low latency, high I/O storage. |
* The price of Hangzhou region in China is more detailed price
Below is a comparison of several important parameters for the cloud disk (excluding the local SSD disk).
Disk Performance test Methods
To test random write IOPS:
Fio-direct=1-iodepth=128-rw=randwrite-ioengine=libaio-bs=4k-size=10g-numjobs=1-runtime=1000-group_reporting- Name=/path/testfile
To test Random Read ioPS:
Fio-direct=1-iodepth=128-rw=randread-ioengine=libaio-bs=4k-size=10g-numjobs=1-runtime=1000-group_reporting- Name=/path/testfile
To test write throughput:
Fio-direct=1-iodepth=64-rw=randwrite-ioengine=libaio-bs=64k-size=10g-numjobs=1-runtime=1000-group_reporting- Name=/path/testfile
To test read throughput:
Fio-direct=1-iodepth=64-rw=randread-ioengine=libaio-bs=64k-size=10g-numjobs=1-runtime=1000-group_reporting- Name=/path/testfile