Monthly Subscription Cluster
Cluster Upgrade
Cluster upgrade follows a daily differential charge model, where the upgrade fee is calculated as follows: Upgrade fee = (New specification price post-upgrade - Original specification price pre-upgrade) x Number of days for configuration adjustment/(365/12) x Applicable discount.
Number of days for configuration adjustment: Resource expiration time - Current time.
Applicable discount: The discount in effect on the live network.
Configuration upgrade doesn't affect the resource expiration time.
The configuration upgrade fee can be paid with your promo vouchers and platform-gifted balance (free credit).
Example:
Assuming a user purchased 5 units of TCHouse-D Cluster Standard 4-core 16 GB machines in Singapore on January 1, 2023, at a unit price of 122.8 USD per month. On October 10, 2023, the configuration was upgraded to 8-core 32 GB, with a unit price of 245.6 USD per month, and a discount of 20%. The upgrade fee calculation process is as follows:
Monthly upgrade price difference: The price difference per machine is 245.6 - 122.8 = 122.8 USD/month.
Number of days for upgrade: 21 days
Applicable discount: 20%
Upgrade fee = (245.6 - 122.8) x 21/(365/12) x 0.8 x 5 = 399.13 USD.
Cluster Downgrade
Cluster downgrade follows the principle refund first, then repurchase: Cluster downgrade refund amount = cluster refund fee - new specification repurchase fee, where the cluster refund fee calculation rules can be found in the Refund Instructions. If the refund amount is greater than 0, after the downgrade is implemented, the refund will be returned to your Tencent Cloud account via the original payment method.
If the refund amount is less than or equal to 0, after the downgrade is implemented, the refund amount will be considered 0, and no additional charges will be applied for any excess fees.
If discounts or promo vouchers were used during the purchase, they will not be returned.
Pay-as-You-Go Cluster
Billing rules remain unchanged after configuration adjustment, and fees are settled hourly based on the cluster configuration at the time of each deduction.