Pay-as-you-go

To purchase a pay-as-you-go CLB instance, go to the buy page.

Billable items

The total amount that you must pay for a CLB instance is the sum of fees for all billable items. The fees that you are charged vary based on network types and instance types of CLB instances. The following table describes the billable items.

Pay-by-bandwidth

Before you purchase pay-by-bandwidth CLB instances, take note of the following rules:

• Instance fee = Instance unit price × Usage duration

  The usage duration of a CLB instance is calculated from the time when the CLB instance is created to the time when the CLB instance is released.

• Bandwidth fee= Highest maximum bandwidth for the day × Bandwidth unit price × Usage duration

  Bandwidth fees are charged on an hourly basis, and bills are generated on a daily basis. If you use a CLB instance for less than 1 hour, the usage duration is rounded up to 1 hour. If you use a CLB instance for less than one day, you are charged for the CLB instance based on the usage duration. In this scenario, the fee of the CLB instance is calculated by multiplying the hours by the unit price of the highest maximum bandwidth for the day.

  The bandwidth fee is charged based on the maximum bandwidth you specify when you create an CLB instance. However, if you change the bandwidth within a billing cycle, the bandwidth fee is charged based on the highest maximum bandwidth within that billing cycle.

• Specification fee: You are charged a specification fee for a high-performance CLB instance based on the actual usage. The fee is calculated based on the specification actually used by the CLB instance, no matter which specification you select.
  Note This rule is not applicable to specifications higher than slb.s3.large, which are available only for accounts that are included in the whitelist.

Instance fees

If you use an Internet-facing CLB instance, you are charged for the public IP address assigned to the CLB instance.

Instance fees for Internet-facing CLB instances are calculated by using the following formula:

• Instance fee = Instance unit price × Usage duration

  The usage duration of a CLB instance is calculated from the time when the CLB instance is created to the time when the CLB instance is released.

• Instance fees are charged on an hourly basis. Fees are deducted in real time. If you use a CLB instance for less than 1 hour within a billing cycle, the usage duration is rounded up to 1 hour.

The prices in the following table are only for reference. The actual prices on the buy page shall prevail.

Data transfer fees

If you use an Internet-facing pay-by-data-transfer CLB instance, you are charged for the actual data transfer.

Data transfer fees for Internet-facing CLB instances are calculated by using the following formula:

• Data transfer fee = Unit price of data transfer × Usage duration

  You are charged only for outbound data transfer (data transfer initiated from internal). You are not charged for inbound data transfer (data transfer initiated from external).

• Instance fees are charged on an hourly basis. Fees are deducted in real time. If you use a CLB instance for less than 1 hour within a billing cycle, the usage duration is rounded up to 1 hour.

  The prices in the following table are only for reference. The actual prices on the buy page shall prevail.

  Region	Data transfer fee (USD/GB)
  China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Hohhot), China (Shenzhen), and China (Heyuan)	0.125
  China (Qingdao)	0.113
  China (Hong Kong)	0.156
  US (Silicon Valley) and US (Virginia)	0.078
  Malaysia (Kuala Lumpur)	0.112
  Singapore (Singapore), Indonesia (Jakarta), India (Mumbai), and Thailand (Bangkok)	0.117
  Japan (Tokyo)	0.087
  Germany (Frankfurt) and UK (London)	0.070
  UAE (Dubai)	0.447
  Australia (Sydney)	0.096
  South Korea (Seoul)	0.123

Specification fee

The following section describes the metrics of high-performance CLB instances. The performance guaranteed varies based on the specification. For more information, see FAQ about high-performance CLB instances.

• Maximum number of connections

  The maximum number of concurrent connections supported by a CLB instance. When the number of existing concurrent connections reaches the upper limit, new connection requests are dropped.

• Connections per second (CPS)

  The maximum number of new connections established per second. When the actual number reaches the upper limit, new connection requests are denied.

• Queries per second (QPS)

  The maximum number of HTTP or HTTPS requests that can be processed per second. This metric is specific to Layer 7 listeners. When the actual number reaches the upper limit, new requests are denied.

  Note A CLB instance uses eight proxy servers to listen for connection requests at Layer 7 and distributes requests to backend servers. Requests are evenly distributed to each proxy server. Therefore, the QPS limit of each proxy server is calculated by using the following formula: QPS per proxy sever = QPS/(8 - 1). For example, the QPS limit of your CLB instance is 1,000. The QPS limit of each proxy server is calculated by using the following formula: 1,000/7 ≈ 143.

The specification fee of a high-performance CLB instance is charged based on the actual usage no matter which specification you choose. If the actual performance of a CLB instance is between two specifications, you are charged based on the higher specification.

• The maximum number of concurrent connections processed by the CLB instance is 90,000, which is higher than the 50,000 limit defined in Standard I (slb.s2.small) and lower than the 100,000 limit defined in Standard II (slb.s2.medium). In this case, Standard II (slb.s2.medium) is used for the hour.

• The maximum number of new connections processed by the CLB instance per second is 4,000, which is higher than the 3,000 limit defined in Small I (slb.s1.small) and lower than the 5,000 limit defined in Standard I (slb.s2.small). In this case, Standard I (slb.s2.small) is used for the hour.

• The maximum number of requests processed by the CLB instance is 11,000, which is higher than the 10,000 limit defined in Standard II (slb.s2.medium) and lower than the 20,000 limit defined in Higher I (slb.s3.small). In this case, Higher I (slb.s3.small) is used for the hour.

  The QPS value is the highest among the preceding dimensions. Therefore, the specification fee for the hour is calculated based on the price of Higher I (slb.s3.small).

The hourly specification fee is calculated based on the preceding method. The following figure shows an example.

Therefore, the billing of high-performance CLB instances is flexible. The specification that you select when you purchase a CLB instance represents the upper performance limit of the CLB instance. For example, if you select Higher II (slb.s3.medium), the performance cannot be higher than Higher II (slb.s3.medium).