Compute plans

更新时间:
复制 MD 格式

ApsaraDB for MongoDB offers compute plans, which are subscription-based resource plans. You can use compute plans to offset the instance type fees of pay-as-you-go replica set instances. This topic describes compute plans, how to purchase them, and how to view their offset usage.

Overview of compute plans

The "pay-as-you-go + compute plan" billing method combines the flexibility of pay-as-you-go with the cost savings of a subscription.

Item

Description

Target Offset

Compute plans can offset the instance type fees of pay-as-you-go replica set instances across all regions under your Alibaba Cloud account. If a compute plan has insufficient capacity, the excess fees are deducted from your Alibaba Cloud account balance on a pay-as-you-go basis.

Offset order

You can purchase multiple compute plans. The plans are used to offset fees in the order of their expiration dates. The plan that expires first is used first.

Renewal and scaling

Not supported. If your compute plan has insufficient capacity, is about to expire, or has expired, you can purchase another plan to stack its usage.

Can I unsubscribe?

You can request a refund for a compute plan only if it has not been used.

Effective period

Each compute plan is valid for 6 months or 1 year. After a plan expires, its remaining capacity becomes invalid.

Compute plan specifications and pricing

ApsaraDB for MongoDB uses the unit CU (Capacity Unit) × H, or "unit" for short, to measure the resource capacity that a compute plan provides. The unit price is CNY 21 per unit. ApsaraDB for MongoDB offers compute plans ranging from 50 to 20,000 units across four different tiers. The higher the tier, the greater the discount. For plans in the same tier, a shorter effective period results in a greater discount.

Compute plan tier (units)

Effective period

Unit price (CNY/unit)

Discount (compared to pay-as-you-go)

50–100

6 months

18.90

10% off

1 year

19.95

5% off

100–1,000

6 months

17.85

15% off

1 year

18.90

10% off

1,000–10,000

6 months

16.80

20% off

1 year

17.85

15% off

10,000–20,000

6 months

15.75

25% off

1 year

16.80

20% off

Compute plan consumption rules

Consumption formula: Region coefficient × Instance node coefficient × Units consumed per hour by the instance type × Usage duration (hours)

Example

Instance environment

Offset coefficient

Compute plan units consumed by the instance

  • Region: China (Hong Kong).

  • Number of primary and secondary nodes: Three nodes.

  • Number of read-only nodes: 3 read-only nodes.

  • Storage type: SSD local disk.

  • Specifications: 1-core 2 GB.

  • Usage duration: 24 hours.

  • Region coefficient: 1.68.

  • Instance node coefficient: 2.

  • Units consumed per hour by the instance type: 0.030.

1.68 × 2 × 0.030 × 24 = 2.4192 (units).

The following tables list the region coefficients, instance node coefficients, and units consumed per hour by instance type.

Region coefficient

The following table lists the region coefficients for instances in different regions.

Instance region

Region coefficient

The Chinese mainland (excluding China (Hong Kong))

1.00

  • China (Hong Kong)

  • Singapore

  • UK (London)

  • Thailand (Bangkok)

  • South Korea (Seoul)

  • Germany (Frankfurt)

  • UAE (Dubai)

1.68

Japan (Tokyo)

1.9

  • Malaysia (Kuala Lumpur)

  • Indonesia (Jakarta)

  • Philippines (Manila)

1.5

  • US (Silicon Valley)

  • US (Virginia)

1.35

Instance node coefficient

The following table lists the node coefficients for instances with different numbers of nodes.

Number of primary and secondary nodes

Number of read-only nodes

Instance node coefficient

Three nodes

No read-only nodes

1

1 read-only node

1.33

2 read-only nodes

1.67

3 read-only nodes

2

4 read-only nodes

2.33

5 read-only nodes

2.67

Five nodes

No read-only nodes

1.67

1 read-only node

2

2 read-only nodes

2.33

3 read-only nodes

2.67

4 read-only nodes

3

5 read-only nodes

3.33

Seven nodes

No read-only nodes

2.33

1 read-only node

2.67

2 read-only nodes

3

3 read-only nodes

3.33

4 read-only nodes

3.67

5 read-only nodes

4

Units consumed per hour by instance type

The following table lists the units consumed per hour by different instance types.

Dedicated disk edition

Specifications

Instance type code (InstanceClass)

Units consumed per hour (units/hour)

2-core 8 GB

mdb.shard.4x.large.d

0.123

2-core 16 GB

mdb.shard.8x.large.d

0.170

4-core 8 GB

mdb.shard.2x.xlarge.d

0.183

4-core 16 GB

mdb.shard.4x.xlarge.d

0.242

4-core 32 GB

mdb.shard.8x.xlarge.d

0.336

8-core 16 GB

mdb.shard.2x.2xlarge.d

0.362

8-core 32 GB

mdb.shard.4x.2xlarge.d

0.481

8-core 64 GB

mdb.shard.8x.2xlarge.d

0.669

16-core 32 GB

mdb.shard.2x.4xlarge.d

0.720

16-core 64 GB

mdb.shard.4x.4xlarge.d

0.958

16-core 128 GB

mdb.shard.8x.4xlarge.d

1.333

32-core 64 GB

mdb.shard.2x.8xlarge.d

1.436

32-core 128 GB

mdb.shard.4x.8xlarge.d

1.912

32-core 256 GB

mdb.shard.8x.8xlarge.d

2.663

General-purpose local disk edition

Specifications

Instance type code (InstanceClass)

Units consumed per hour (units/hour)

1-core 2 GB

dds.mongo.mid

0.030

2-core 4 GB

dds.mongo.standard

0.060

4-core 8 GB

dds.mongo.large

0.109

8-core 16 GB

dds.mongo.xlarge

0.209

8-core 32 GB

dds.mongo.2xlarge

0.397

16-core 64 GB

dds.mongo.4xlarge

0.794

Dedicated local disk edition

Specifications

Instance type code (InstanceClass)

Units consumed per hour (units/hour)

2-core 16 GB

mongo.x8.medium

0.159

4-core 32 GB

mongo.x8.large

0.308

8-core 64 GB

mongo.x8.xlarge

0.605

16-core 128 GB

mongo.x8.2xlarge

1.190

32-core 256 GB

mongo.x8.4xlarge

2.331

Dedicated physical machine

Specifications

Instance type code (InstanceClass)

Units consumed per hour (units/hour)

60-core 440 GB

dds.mongo.2xmonopolize

7.577

90-core 660 GB

dds.mongo.3xmonopolize

8.451

Purchase a compute plan

  1. Log on to the ApsaraDB for MongoDB console.

  2. In the navigation pane on the left, click Replica Set Instances.

  3. On the ReplicaSet Instances page, click Purchase Compute Plan.

  4. Set the following parameters to configure the compute plan.

    Parameter

    Description

    Compute Plan Tier

    Select the capacity range of the compute plan in units (CU×H).

    The larger the capacity, the greater the discount. For more information, see Compute plan specifications and pricing.

    Compute Plan Capacity

    The capacity of the compute plan in units (CU×H).

    Subscription Duration

    The effective period of the compute plan.

    The compute plan takes effect immediately after purchase. Each plan is valid for 6 months or 1 year. After a plan expires, its remaining capacity becomes invalid.

    Quantity

    The number of compute plans to purchase.

  5. Click Buy Now.

  6. Read the Terms of Service and click Pay.

View compute plan offset usage

Important

A compute plan takes effect immediately after purchase. However, the usage details are typically displayed with a 6-hour delay. Check the offset usage 6 hours after your purchase.

  1. Go to the Resource Plan page.

    You can go to the Resource Plan page in either of the following ways:

  2. On the Resource Plan page, set Resource Dimension to Resource Plan.

  3. On the Resource Instance Management page, set Resource Dimension to Resource Plan.

  4. View the details of the compute plan.

    You can view the following information on the Instance Summary or Usage Details tab:

    • Instance Summary: On this page, you can view the Status, Total Capacity, Remaining Capacity, Effective Time, and Expiration Time of the compute plan.

    • Usage Details: On this page, you can view the detailed usage of the compute plan. This includes Offset Time, Remaining Capacity Before Offset, Offset Amount, Remaining Capacity After Offset, Offset Instance ID, Offset Billable Item, and Original Usage of Resource Instance.

    Note

    For instances whose fees are offset by a compute plan, you can still view the pay-as-you-go billing records in Billing Details. However, the payable amount for these bills is 0 because the fees have been offset by the compute plan.