Computing resources (vCPUs and memory) are core components that you must purchase when you order a CloudBox. This topic describes how computing resources in CloudBox are billed and lists the specifications of available compute SKUs.
Billing details
CloudBox offers computing resources as compute SKUs (Stock Keeping Units). Each SKU includes a specific number of vCPUs and a specific memory size. A CloudBox must include at least one compute SKU, and you can combine multiple SKUs to fit your workload. For more information, see Compute SKU specifications.
The fee for a compute SKU is calculated as: Fee = Unit Price × Quantity × Subscription Duration.
Unit price: The unit price of a compute SKU varies based on its vCPU count and memory size.
NoteYou can view the configurations and prices of each compute SKU in the CloudBox console when placing an order.
Quantity: You can purchase multiple units of each compute SKU.
Subscription duration: You can subscribe to resources for yearly terms. For an initial CloudBox purchase, the subscription duration can be three or five years.
Payment options
Fees for computing resources are bundled with other charges in your CloudBox order. CloudBox uses the subscription billing method. You can choose from the All Upfront, Partial Upfront, or No Upfront payment options.
All Upfront: Pay the full amount at the time of purchase.
Half Upfront: Pay 50% of the total cost at the time of purchase and pay the remaining balance in monthly installments.
After you confirm the order completion, Alibaba Cloud generates an installment bill on the first day of each month, starting from the month after your confirmation. The amount of each installment is calculated as: Remaining Balance / Number of Subscription Months. Ensure that your account has a sufficient balance to pay the bills on time. Failure to pay a bill within seven calendar days of receipt may result in service interruption.
Compute SKU specifications
A compute SKU is a basic unit of computing power in CloudBox, defining a server's vCPU and memory. Each compute SKU corresponds to a maximum instance type, which determines the instance family and the number of ECS instances you can create. The following table describes the supported maximum instance types.
For recommendations on how to configure CloudBox computing resources, see Best practices for configuring computing resources for CloudBox.
Maximum instance type | Type | Features | Other instance types |
General-purpose |
| ecs.g6.large, ecs.g6.xlarge, and more | |
Compute-optimized |
| ecs.c6.large, ecs.c6.xlarge, and more | |
Memory-optimized |
| ecs.r6.large, ecs.r6.xlarge, and more | |
GPU-accelerated compute-optimized |
| ecs.gn6i-c4g1.xlarge, ecs.gn6i-c8g1.2xlarge, and more | |
ecs.g6se.26xlarge | Storage-enhanced |
| ecs.g6se.large, ecs.g6se.xlarge, and more |
When creating an ECS instance, you can select any instance type from the same family as the maximum instance type. All instance types in the same instance family have the same vCPU-to-memory ratio. For example, if you purchase a compute SKU that supports ecs.g6.26xlarge as the maximum instance type, which has a vCPU-to-memory ratio of approximately 1:4, you can create ECS instances of types such as ecs.g6.large (2 vCPUs, 8 GiB) and ecs.g6.xlarge (4 vCPUs, 16 GiB).
Compute SKUs are grouped into the following categories based on their primary use case: general-purpose, compute-optimized, memory-optimized, and GPU-accelerated compute-optimized. You can purchase multiple compute SKUs to use a combination of different instance types.
If you cannot find a suitable configuration, contact your account manager to request a custom SKU configuration.
Table 1. General-purpose compute units
SKU ID | vCPU | Memory (GiB) | Maximum instance type | Maximum instance type quantity | Use case |
OR-3EAB470501 | 208 | 768 | ecs.g6.26xlarge | 2 | A small general-purpose unit for applications in the retail and enterprise sectors. |
OR-2DDFCB757E | 312 | 1152 | ecs.g6.26xlarge | 3 | An enhanced small-to-medium general-purpose unit for I/O-intensive workloads, such as medium-to-large OLTP core databases. |
OR-E651CE3606 | 416 | 1536 | ecs.g6.26xlarge | 4 | A small-to-medium general-purpose unit for applications in the retail and enterprise sectors. |
OR-9E266697B6 | 520 | 1920 | ecs.g6.26xlarge | 5 | A medium general-purpose unit for applications such as point-of-sale systems or enterprise workspaces. |
OR-724E59CD09 | 624 | 2304 | ecs.g6.26xlarge | 6 | A medium general-purpose unit for applications such as sales systems or enterprise workspaces. |
OR-A8B62E9514 | 936 | 3456 | ecs.g6.26xlarge | 9 | A large general-purpose unit for Internet of Things (IoT) applications that store and manage sensor data. |
OR-68B18B81EE | 624 | 2688 | ecs.g6.26xlarge | 2 | A medium mixed-capacity unit for running applications from multiple teams on a single CloudBox. |
ecs.c6.26xlarge | 2 | ||||
ecs.r6.26xlarge | 2 | ||||
OR-218DCFD832 | 936 | 3648 | ecs.g6.26xlarge | 4 | A large mixed-capacity unit for running different applications with shared capacity on a single CloudBox. |
ecs.c6.26xlarge | 3 | ||||
ecs.r6.26xlarge | 2 | ||||
OR-B900CF0F79 | 624 | 1728 | ecs.g6.26xlarge | 3 | A medium general-purpose compute unit for applications that require high-performance computing. |
ecs.c6.26xlarge | 3 | ||||
OR-2603EC27DF | 936 | 2880 | ecs.g6.26xlarge | 6 | A large general-purpose compute unit for compute-intensive applications. |
ecs.c6.26xlarge | 3 | ||||
OR-6BF6AC9C02 | 416 | 1536 | ecs.g6se.26xlarge | 4 | A small-to-medium general-purpose compute unit for I/O-intensive workloads, such as medium-to-large OLTP core databases. |
OR-715AEEC381 | 936 | 3456 | ecs.g6se.26xlarge | 9 | A medium-to-large general-purpose compute unit for I/O-intensive workloads, such as medium-to-large OLTP core databases. |
Table 2. Compute-optimized units
SKU ID | vCPU | Memory (GiB) | Maximum instance type | Maximum instance type quantity | Use case |
OR-BC72F47D81 | 208 | 384 | ecs.c6.26xlarge | 2 | A small compute-optimized unit for compute-intensive applications. |
OR-8E676BE2D7 | 416 | 768 | ecs.c6.26xlarge | 4 | A small-to-medium compute-optimized unit for compute-intensive applications. |
OR-D4ADF7B9A7 | 936 | 1728 | ecs.c6.26xlarge | 9 | A medium compute-optimized unit for compute-intensive applications. |
OR-E05A0A94CB | 624 | 1536 | ecs.c6.26xlarge | 4 | A medium mixed-capacity unit for high-performance and compute-intensive applications. |
ecs.g6.26xlarge | 2 | ||||
OR-EFC36249B2 | 936 | 2304 | ecs.c6.26xlarge | 6 | A large mixed-capacity unit for high-performance and compute-intensive applications. |
ecs.g6.26xlarge | 3 |
Table 3. Memory-optimized units
SKU ID | vCPU | Memory (GiB) | Maximum instance type | Maximum instance type quantity | Use case |
OR-A6A14CF18D | 624 | 3840 | ecs.r6.26xlarge | 4 | A small memory-optimized unit for local databases with in-memory analytics capabilities. |
ecs.g6.26xlarge | 2 | ||||
OR-31D9EEAC15 | 624 | 4608 | ecs.r6.26xlarge | 6 | A medium memory-optimized unit for local databases with in-memory analytics capabilities. |
OR-D598F0C820 | 832 | 6144 | ecs.r6.26xlarge | 8 | A large memory-optimized unit for large-scale local databases. |
Table 4. GPU-accelerated compute-optimized units
SKU ID | vCPU | Memory (GiB) | GPU | Maximum instance type | Maximum instance type quantity | Use case |
OR-C051AB7D0D | 480 | 1920 | Intel® Server GPU × 5 | ecs.ebmgi6s.24xlarge | 5 | Ideal for AI deep learning tasks, such as image classification, object detection, video content moderation, optical character recognition (OCR), and speech recognition. Also suitable for professional graphics processing tasks, such as graphics and image rendering, industrial 3D design, and non-linear video editing. |
OR-FD8C1EA2E4 | 592 | 2256 | NVIDIA T4 × 16 | ecs.gn6i-c24g1.24xlarge | 4 |
|
ecs.g6.26xlarge | 2 | |||||
OR-48E9DF43BC | 304 | 1140 | NVIDIA T4 × 4 | ecs.gn6i-c24g1.24xlarge | 1 |
|
ecs.g6.26xlarge | 2 |
Related documents
To learn how to purchase computing resources for CloudBox, see Purchase CloudBox resources.