Disk Types and Performance

EVS disks are classified based on the disk I/O performance. EVS disks differ in performance and price. Choose the disk type most appropriate for your applications.

EVS Performance

EVS performance metrics include:

IOPS: Number of read/write operations performed by an EVS disk per second
Throughput: Amount of data read from and written into an EVS disk per second
Read/write I/O latency: Minimum interval between two consecutive read/write operations on an EVS disk
Single-queue access latencies of different types of EVS disks are as follows:
- Common I/O: 5 ms to 10 ms
- High I/O: 1 ms to 3 ms
- Ultra-high I/O: 1 ms

**Table 1** EVS performance data
Parameter	Common I/O	High I/O	Ultra-high I/O
IOPS per GiB/EVS disk	2	8	50
Max. IOPS/EVS disk	2,200	5,000	50,000
Baseline IOPS/EVS disk	500	1,800	1,800
Disk IOPS	Min. (2,200, 500 + 2 x Capacity)	Min. (5,000, 1,800 + 8 x Capacity)	Min. (50,000, 1,800 + 50 x Capacity)
IOPS burst limit/EVS disk	2,200	5,000	16,000
Max. throughput	50 MiB/s	150 MiB/s	350 MiB/s
API name NOTE: This API name indicates the value of the volume_type parameter in the EVS API. It does not represent the type of the underlying hardware device.	SATA	SAS	SSD
Typical application scenarios	Business applications that require large capacity, normal read/write speed, and few transactions, such as enterprise office applications and small-scale testing. When higher I/O performance is required, it is recommended that you select ultra-high I/O or high I/O EVS disks.	Mainstream applications requiring high performance and high reliability, such as large-scale development and test environments, web server logs, and enterprise applications. Typical enterprise applications include SAP applications, Microsoft Exchange, and Microsoft SharePoint.	Read/write-intensive applications that require ultra-high I/O and throughput, such as distributed file systems used in HPC scenarios or NoSQL and relational databases used in I/O-intensive scenarios. Typical databases include MongoDB, Oracle, SQL Server, MySQL, and PostgreSQL databases.

Calculating Disk IOPS Limit

To calculate the IOPS limit of a disk, obtain the smaller value of the following two values:

Max. IOPS/disk
Baseline IOPS/disk + IOPS per GiB x Disk capacity

The following example uses an ultra-high I/O EVS disk with a maximum IOPS of 33,000.

If the disk capacity is 100 GiB, the disk IOPS limit is calculated as follows:
Disk IOPS limit = Min. (33,000, 1,500 + 50 x 100)

The disk IOPS limit is 6,500, the smaller value between 33,000 and 6,500.
If the disk capacity is 1,000 GiB, the disk IOPS limit is calculated as follows:
Disk IOPS limit = Min. (33,000, 1,500 + 50 x 1,000)

The disk IOPS limit is 33,000, the smaller value between 33,000 and 51,500.

Disk Burst Capability and Principles

EVS disks have burst capability, which allows a small-capacity disk to surpass its maximum IOPS within a certain period of time.

Disks with burst capability are well-suited for speeding up server startup. In most cases, system disks have small capacities. For example, the IOPS of a 50-GiB ultra-high I/O disk without burst capability can reach only 4,000 (1,500 + 50 x 50). If the disk has burst capability, its IOPS can burst up to 16,000.

The following example uses an ultra-high I/O EVS disk with the IOPS burst limit of 16,000.

If the disk capacity is 100 GiB, the disk has a maximum IOPS of 6,500, but it can burst to 16,000 IOPS in a certain duration.
If the disk capacity is 1,000 GiB, the disk has a maximum IOPS of 33,000. The disk maximum IOPS already exceeds its burst IOPS 10,000, and the disk does not use the burst capability.

The following describes the burst IOPS consumption and reservation.

A token bucket is used to handle burst I/O operations. The number of initial tokens in the bucket is calculated as follows:

Number of initial tokens = Burst duration x IOPS burst limit

In the following example, a 100-GiB ultra-high I/O EVS disk is used, and the fixed burst duration is 1800s. Therefore, the number of initial tokens is 28,800,000 (1,800 x 16,000).

Token production rate: This rate equals the disk maximum IOPS, which is 6,500 tokens/s.
Token consumption rate: This rate is calculated based on the I/O usage. Each I/O request consumes a token. The maximum consumption rate is 10,000 tokens/s, which is the larger value between the disk burst IOPS and maximum IOPS.

Consumption principles

When the token consumption rate is greater than the production rate, the number of tokens decreases accordingly, and eventually the disk IOPS will be consistent with the token production rate (the maximum IOPS). In this example, the disk can burst for approximately 3,032 seconds [28,800,000/(16,000 - 6,500)].

Reservation principles

When the token consumption rate is smaller than the production rate, the number of tokens increases accordingly, enabling the disk to regain the burst capability. In this example, if the disk is suspended for approximately 4,431 seconds (28,800,000/6,500), the token bucket will be filled up with tokens.

As long as there are tokens in the token bucket, the disk has the burst capability.

Figure 1 shows the token consumption and reservation principles. The blue bars indicate the disk IOPS usage, the green dashed line represents the maximum IOPS, the red dashed line indicates the IOPS burst limit, and the black curve indicates the changes of the number of tokens.

When the number of tokens is greater than zero, the disk IOPS can exceed 6,500 and has the capability to reach 16,000, the IOPS burst limit.
When the number of tokens is zero, the disk does not have the burst capability, and the maximum IOPS is 6,500.
When the disk IOPS is less than 6,500, the number of tokens starts to increase, and the disk can regain the burst capability.

Figure 1 Burst capability diagram
Click to enlarge