1.1.1 FAQs for Billing
What Payment Methods Are Supported?
At present, eSurfing Cloud Distributed Message Service (Kafka) supports two billing modes, subscription-based billing (annual and monthly subscription) and pay-as-you-go.
l Subscription-based billing
You can subscribe resources by month or by year. This offers a larger discount than the pay-as-you-go mode, and is recommended for long-term users. The fees are billed based on your subscription cycle.
l Pay-as-you-go
Pay-as-you-go billing is based on the billing cycle of the resources. When the billing cycle is reached, the system generates a bill and charges you accordingly. If your account balance is insufficient, the resources enter the grace period. If the payment is not completed during the grace period, the resources are stopped and reclaimed.
How Do I Choose a Billing Mode?
The following table shows the differences between the two billing modes:
Billing Mode | Yearly/Monthly subscription | Pay-As-You-Go |
Payment Methods | Prepaid and billed by your subscription cycle. | Postpaid and billed by service duration. |
Billing Cycle | Billed based on your subscription cycle. | Billed by hour. |
Instance Upgrade | Scale-up is supported. The new data takes effect after the ticket is completed. The service is not available when the ticket is being carried out. Scale-down is not supported. | Scale-up is supported. The new data takes effect after the ticket is completed. The service is not available when the ticket is being carried out. Scale-down is not supported. |
Changing Billing Mode | Can be changed to pay-as-you-go. | Can be changed to annual/monthly subscription. |
Changing Specifications | Instance type can be changed. | Instance type can be changed. |
Application Scenarios | This mode is ideal when you can estimate the resource usage period. The price is more favorable than that of the pay-as-you-go billing mode. This mode is recommended for long-term users. | This mode is ideal when you want more flexibility and control over resource usage. The service can be enabled and disabled anytime. |
You can change between subscription-based billing and pay-as-you-go. For details, see related sections.
What Are the Available Bandwidth and Disk Capacities?
Bandwidth for advanced version: 300 MB/s; bandwidth for basic version: 100 MB/s
Disk capacities for advanced version:1200 GB-6,000 GB; disk capacities for basic version: 600 GB-6,000 GB
Why Can't I Create a Pay-As-You-Go Kafka Instance When My Account Is Not in Arrears?
Please check if your account balance is less than RMB 100. If so, you cannot create a Kafka instance.
(1) Pre-deposit and post-payment: You can top up your eSurfing Cloud account and ensure that your account balance is not less than RMB 100. Payment is made according to your actual usage.
(2) Billing cycle: The service is charged by hour (based on Beijing time). The fees for one hour of service are charged if the service time is less than one hour. Fees are deducted from the cash balance of your account. We recommend you activate the service by hour to maximize cost effectiveness. The service is also charged by hour if you delete the instance in advance.
How Can I Reactivate the Service if My Account Is in Arrears?
If you pay off the arrears within 15 days, the service will be activated automatically.
If you do not pay off the arrears within the required period, the resource enters a retention period. During this period, your cloud services cannot be accessed or used, but your stored cloud data will be retained.
l If you top up your account during the retention period, the system will automatically deduct the amount owed.
l If you still do not complete the top-up after the retention period has ended, your stored cloud data will be deleted and the resources will be released.
How Do I Unfreeze Resources When My Subscription Expires?
If you renew your subscription within 15 days after the expiration date, the new cycle is calculated from your renewal time.
Table Dealing with resources in arrears
Product Type | Retention Period | Resources in Arrears After Retention Period Ends |
Computing, storage, network, and database products | 15 days | The stored data will be deleted and the cloud resources will be released. |
Terminal products | 7 days | The stored data will be deleted and the cloud resources will be released. |
Can I Unsubscribe From an Instance?
If you want to unsubscribe from the resources, you can log in to eSurfing Cloud Management Center or the product console for unsubscription. Currently, there are three unsubscription methods, 7-day full refund, non-full refund, and unconditional refund. For details, refer to Unsubscription Rules in Billing Center on the eSurfing Cloud official website.
Subject to compliance with the eSurfing Cloud unsubscription rules, you can unsubscribe from resources based on your needs. The following shows the details of different unsubscription methods:
l 7-day Full Refund You can enjoy a 7-day full refund for newly purchased resource instances (excluding resources that have undergone renewals, configuration upgrades, capacity expansions, and operating system changes) if all of the following conditions are met: The unsubscription is initiated within seven days after the resource is activated.
¡ Within the first 7 days of purchase, you can request a full refund and the full payment will be refunded. Each account can request a 7-day full refund for up to 3 times.
¡ The same user can request a 7-day full refund for up to 15 times accumulatively. According to the associated information of different accounts used for registration, login and operation, users with the same associated information are considered the same user. Examples of associated information: identical name, email address, certificate of the person in charge, mobile phone number, device, and IP address (including canceled accounts). The user hereby agrees that eSurfing Cloud can use the above information to verify if the user is the same.
¡ Unsubscribing from bundled resources (sold as a package) is considered unsubscribing from one resource instance and counted as one unsubscription.
1.1.2 FAQs for Purchasing
Which Versions Are Available?
You can purchase a Kafka engine and a native cloud engine and select the host type, node specifications, number of nodes, and storage space based on your demand.
How Do I Renew an Instance?
Click Renew in the cluster list. You can pay for required duration on the displayed page.
Manual renewal: For annual/monthly subscriptions of distributed cache services, if you renew a resource before expiration, you can extend the expiration time of the original resource and prevent it from being frozen after expiration or reclaimed by the system after the grace period. For details, please refer to Renewal Management - Manual Renewal in the Billing Center.
Automatic Renewal only applies to monthly/yearly resources. For details, please refer to Renewal Management - Automatic Renewal in Billing Center.
How Do I Unsubscribe From Instances?
Click Unsubscribe in the cluster list to access the unsubscription details page. If you want to unsubscribe from the resources, you can log in to eSurfing Cloud Management Center or the product console for unsubscription. Currently, there are three unsubscription methods, 7-day full refund, non-full refund, and unconditional refund. For details, refer to Unsubscription Rules in Billing Center on the eSurfing Cloud official website.
Can I Delete a Periodical Kafka Instance?
Yes. You can log in to the console, select the instance and choose More - Unsubscribe.
1.1.3 Does Kafka Support Multiple AZs?
Kafka nodes can be deployed and configured in an AZ or across multiple AZs.
l In single-AZ deployment mode, all Kafka nodes are located in the same AZ. This configuration mode applies to scenarios with low availability requirements, such as development or small-scale applications.
l In multi-AZ deployment mode, Kafka nodes are distributed across different AZs. This configuration mode provides higher availability and fault tolerance for AZ failures. Generally, a replication mechanism is needed in multi-AZ deployment for data redundancy and fault recovery.
The multi-AZ deployment mode generally uses three AZs, which provides better fault tolerance and availability. Each partition has a leader and two replicas spread across three different AZs.
Note:
Multi-AZ deployment increases network latency and complexity. Therefore, you need to carefully evaluate and plan network architecture, data replication strategies, and fault recovery mechanisms.
In summary, Kafka supports single- and multi-AZ deployment modes and you can select a mode based on your requirements for HA and fault tolerance, as well as available resources and network architecture.
1.1.4 How Do I Choose Between High I/O and Ultra-high I/O?
The choice of Kafka disk type mainly depends on your demands and budgets. Featuring faster random read and write and lower access latency, high I/O disks provide better performance and throughput
l and are preferred for Kafka. This feature is very important for Kafka, because its performance and throughput are limited by the read/write speed of disks. High I/O disks are suitable for scenarios where massive read/write operations are handled, such as frequent message transmission and data writes. They can process messages faster with lower latency, improving the system performance.
l Compared with traditional high I/O disks, ultra-high I/O disks are superior with faster random read/writes and lower latency. They apply to scenarios requiring extremely high performance, such as large-scale real-time data processing and message queues featuring high throughput. High I/O disks support faster data processing and higher system throughput with higher IOPS and lower latency.
In summary, high I/O disks are generally used for Kafka because of their satisfactory performance and throughput. If you have a generous budget, ultra-high I/O disks can also be used.