Billing method
OpenSearch Recall Engine Edition is available with the following billing methods.
Billing method | Description |
Pay-as-you-go | Also known as post-paid. You are billed hourly based on instance specifications, and the charges are deducted from your Alibaba Cloud account. This method is ideal for short-term needs, like initial testing. You can release an instance at any time to save costs. |
Subscription | Also known as pre-paid. With this method, you pay for an instance when you create it. It is more cost-effective than pay-as-you-go. |
Pricing composition
Billing for OpenSearch - Recall Engine Edition instances consists of six components: the instance fee, query node fee, data node fee, storage space per data node, index storage fee, and data update resource fee.
Total cost = instance fee + query node fee + data node fee + data node storage fee + index storage fee + data update resources fee.
Instance, query, and data node fees
Pay-as-you-go
Instance rental fee = unit price * hours
Query node fee = unit price × hours × number of query nodes
Data node fee = Unit price * Hours * Number of data nodes
Region | Billable item | Instance family | Available specifications | Unit price | Unit |
China (Hangzhou), China (Shanghai), China (Beijing), China (Shenzhen), China (Zhangjiakou) | instance usage | - | - | 0.38 | CNY/hour/instance |
query node | local SSD disk type | 2 cores 8 GB | 0.58 | CNY/hour/instance | |
4 cores 16 GB | 1.16 | CNY/hour/instance | |||
8 cores 16 GB (China (Shanghai) only) | 2.03 | CNY/hour/instance | |||
8 cores 32 GB | 2.31 | CNY/hour/instance | |||
cloud disk type | 2 cores 8 GB | 0.38 | CNY/hour/instance | ||
4 cores 16 GB | 0.75 | CNY/hour/instance | |||
8 cores 32 GB | 1.51 | CNY/hour/instance | |||
data node | local SSD disk type | 2 cores 16 GB | 0.71 | CNY/hour/instance | |
4 cores 32 GB | 1.42 | CNY/hour/instance | |||
6 cores 48 GB | 2.13 | CNY/hour/instance | |||
8 cores 16 GB (China (Shanghai) only) | 2.03 | CNY/hour/instance | |||
8 cores 64 GB | 2.84 | CNY/hour/instance | |||
10 cores 80 GB | 3.55 | CNY/hour/instance | |||
12 cores 96 GB | 4.26 | CNY/hour/instance | |||
14 cores 112 GB | 4.97 | CNY/hour/instance | |||
16 cores 128 GB | 5.68 | CNY/hour/instance | |||
cloud disk type | 2 cores 16 GB | 0.50 | CNY/hour/instance | ||
4 cores 16 GB | 0.75 | CNY/hour/instance | |||
4 cores 32 GB | 1.00 | CNY/hour/instance | |||
8 cores 16 GB | 1.17 | CNY/hour/instance | |||
8 cores 32 GB | 1.51 | CNY/hour/instance | |||
8 cores 64 GB | 2.00 | CNY/hour/instance | |||
12 cores 24 GB | 1.76 | CNY/hour/instance | |||
12 cores 48 GB | 2.26 | CNY/hour/instance | |||
12 cores 96 GB | 2.99 | CNY/hour/instance | |||
16 cores 32 GB | 2.35 | CNY/hour/instance | |||
16 cores 64 GB | 3.01 | CNY/hour/instance | |||
16 cores 128 GB | 3.99 | CNY/hour/instance | |||
24 cores 48 GB | 3.52 | CNY/hour/instance | |||
Singapore, Germany (Frankfurt) | instance usage | - | - | 0.61 | CNY/hour/instance |
query node | local SSD disk type | 2 cores 8 GB | 0.69 | CNY/hour/instance | |
4 cores 16 GB | 1.39 | CNY/hour/instance | |||
8 cores 32 GB | 2.78 | CNY/hour/instance | |||
cloud disk type | 2 cores 8 GB | 0.60 | CNY/hour/instance | ||
4 cores 16 GB | 1.21 | CNY/hour/instance | |||
8 cores 32 GB | 2.41 | CNY/hour/instance | |||
data node | local SSD disk type | 2 cores 16 GB | 0.84 | CNY/hour/instance | |
4 cores 32 GB | 1.68 | CNY/hour/instance | |||
6 cores 48 GB | 2.52 | CNY/hour/instance | |||
8 cores 64 GB | 3.36 | CNY/hour/instance | |||
10 cores 80 GB | 4.19 | CNY/hour/instance | |||
12 cores 96 GB | 5.03 | CNY/hour/instance | |||
14 cores 112 GB | 5.87 | CNY/hour/instance | |||
16 cores 128 GB | 6.71 | CNY/hour/instance | |||
cloud disk type | 2 cores 16 GB | 0.74 | CNY/hour/instance | ||
4 cores 16 GB | 1.21 | CNY/hour/instance | |||
4 cores 32 GB | 1.49 | CNY/hour/instance | |||
8 cores 16 GB | 1.88 | CNY/hour/instance | |||
8 cores 32 GB | 2.41 | CNY/hour/instance | |||
8 cores 64 GB | 2.97 | CNY/hour/instance | |||
12 cores 24 GB | 2.81 | CNY/hour/instance | |||
12 cores 48 GB | 3.62 | CNY/hour/instance | |||
12 cores 96 GB | 4.46 | CNY/hour/instance | |||
16 cores 32 GB | 3.75 | CNY/hour/instance | |||
16 cores 64 GB | 4.83 | CNY/hour/instance | |||
16 cores 128 GB | 5.94 | CNY/hour/instance | |||
24 cores 48 GB | 5.63 | CNY/hour/instance | |||
China (Hong Kong) | instance usage | - | - | 0.61 | CNY/hour/instance |
query node | local SSD disk type | 2 cores 8 GB | 0.97 | CNY/hour/instance | |
4 cores 16 GB | 1.94 | CNY/hour/instance | |||
8 cores 32 GB | 3.88 | CNY/hour/instance | |||
cloud disk type | 2 cores 8 GB | 1.11 | CNY/hour/instance | ||
4 cores 16 GB | 2.23 | CNY/hour/instance | |||
8 cores 32 GB | 4.45 | CNY/hour/instance | |||
data node | local SSD disk type | 2 cores 16 GB | 1.34 | CNY/hour/instance | |
4 cores 32 GB | 2.69 | CNY/hour/instance | |||
6 cores 48 GB | 4.03 | CNY/hour/instance | |||
8 cores 64 GB | 5.36 | CNY/hour/instance | |||
10 cores 80 GB | 6.72 | CNY/hour/instance | |||
12 cores 96 GB | 8.06 | CNY/hour/instance | |||
14 cores 112 GB | 9.41 | CNY/hour/instance | |||
16 cores 128 GB | 10.73 | CNY/hour/instance | |||
cloud disk type | 2 cores 16 GB | 1.41 | CNY/hour/instance | ||
4 cores 16 GB | 2.23 | CNY/hour/instance | |||
4 cores 32 GB | 2.83 | CNY/hour/instance | |||
8 cores 16 GB | 3.35 | CNY/hour/instance | |||
8 cores 32 GB | 4.45 | CNY/hour/instance | |||
8 cores 64 GB | 5.65 | CNY/hour/instance | |||
12 cores 24 GB | 5.03 | CNY/hour/instance | |||
12 cores 48 GB | 6.68 | CNY/hour/instance | |||
12 cores 96 GB | 8.48 | CNY/hour/instance | |||
16 cores 32 GB | 6.70 | CNY/hour/instance | |||
16 cores 64 GB | 8.90 | CNY/hour/instance | |||
16 cores 128 GB | 11.30 | CNY/hour/instance | |||
24 cores 48 GB | 10.05 | CNY/hour/instance |
For low-QPS workloads with large indexes, use an instance type with local SSDs. For high-QPS workloads or if your index can fit into memory, use an instance type with cloud disks.
Data node allocation: After purchasing data nodes, you can configure the number of replicas and shards for your instance. The number of data nodes must be less than the number of replicas multiplied by the number of shards. You can change the replica count in the instance information and the shard count in the basic table information.
System-reserved space: The system reserves a portion of disk space on data nodes of an instance type with cloud disks. The approximate reserved space per node is as follows: 2 GB for a 2 vCPU/16 GB specification, 2 GB for a 4 vCPU/32 GB specification, 3 GB for an 8 vCPU/64 GB specification, and 5 GB for a 16 vCPU/128 GB specification. This restriction does not apply to data nodes of an instance type with local SSDs.
Subscription
Region | Billable item | Instance family | Instance type | Unit price | Unit |
China (Hangzhou), China (Shanghai), China (Beijing), China (Shenzhen) | instance subscription | - | - | 248.48 | CNY/month |
query node | local SSD | 2 vCPUs 8 GB | 375.02 | CNY/month | |
4 vCPUs 16 GB | 750.04 | CNY/month | |||
8 vCPUs 16 GB (China (Shanghai) only) | 1,318.25 | CNY/month | |||
8 vCPUs 32 GB | 1,500.08 | CNY/month | |||
cloud disk | 2 vCPUs 8 GB | 244.08 | CNY/month | ||
4 vCPUs 16 GB | 488.16 | CNY/month | |||
8 vCPUs 32 GB | 976.32 | CNY/month | |||
data node | local SSD | 2 vCPUs 16 GB | 459.68 | CNY/month | |
4 vCPUs 32 GB | 919.37 | CNY/month | |||
6 vCPUs 48 GB | 1,379.05 | CNY/month | |||
8 vCPUs 16 GB (China (Shanghai) only) | 1,318.25 | CNY/month | |||
8 vCPUs 64 GB | 1,838.74 | CNY/month | |||
10 vCPUs 80 GB | 2,298.42 | CNY/month | |||
12 vCPUs 96 GB | 2,758.10 | CNY/month | |||
14 vCPUs 112 GB | 3,217.79 | CNY/month | |||
16 vCPUs 128 GB | 3,677.47 | CNY/month | |||
cloud disk | 2 vCPUs 16 GB | 323.41 | CNY/month | ||
4 vCPUs 16 GB | 488.16 | CNY/month | |||
4 vCPUs 32 GB | 646.81 | CNY/month | |||
8 vCPUs 16 GB | 760.72 | CNY/month | |||
8 vCPUs 32 GB | 976.32 | CNY/month | |||
8 vCPUs 64 GB | 1,293.62 | CNY/month | |||
12 vCPUs 24 GB | 1,141.07 | CNY/month | |||
12 vCPUs 48 GB | 1,464.48 | CNY/month | |||
12 vCPUs 96 GB | 1,940.44 | CNY/month | |||
16 vCPUs 32 GB | 1,521.43 | CNY/month | |||
16 vCPUs 64 GB | 1,952.64 | CNY/month | |||
16 vCPUs 128 GB | 2,587.25 | CNY/month | |||
24 vCPUs 48 GB | 2,282.15 | CNY/month | |||
Singapore, Germany (Frankfurt) | instance subscription | - | - | 397.47 | CNY/month |
query node | local SSD | 2 vCPUs 8 GB | 449.86 | CNY/month | |
4 vCPUs 16 GB | 899.73 | CNY/month | |||
8 vCPUs 32 GB | 1,799.46 | CNY/month | |||
cloud disk | 2 vCPUs 8 GB | 391.19 | CNY/month | ||
4 vCPUs 16 GB | 782.38 | CNY/month | |||
8 vCPUs 32 GB | 1,564.76 | CNY/month | |||
data node | local SSD | 2 vCPUs 16 GB | 543.59 | CNY/month | |
4 vCPUs 32 GB | 1,087.17 | CNY/month | |||
6 vCPUs 48 GB | 1,630.76 | CNY/month | |||
8 vCPUs 64 GB | 2,174.35 | CNY/month | |||
10 vCPUs 80 GB | 2,717.93 | CNY/month | |||
12 vCPUs 96 GB | 3,261.52 | CNY/month | |||
14 vCPUs 112 GB | 3,805.11 | CNY/month | |||
16 vCPUs 128 GB | 4,347.68 | CNY/month | |||
cloud disk | 2 vCPUs 16 GB | 481.51 | CNY/month | ||
4 vCPUs 16 GB | 782.38 | CNY/month | |||
4 vCPUs 32 GB | 963.02 | CNY/month | |||
8 vCPUs 16 GB | 1,215.36 | CNY/month | |||
8 vCPUs 32 GB | 1,564.76 | CNY/month | |||
8 vCPUs 64 GB | 1,926.04 | CNY/month | |||
12 vCPUs 24 GB | 1,823.03 | CNY/month | |||
12 vCPUs 48 GB | 2,347.13 | CNY/month | |||
12 vCPUs 96 GB | 2,889.05 | CNY/month | |||
16 vCPUs 32 GB | 2,430.71 | CNY/month | |||
16 vCPUs 64 GB | 3,129.51 | CNY/month | |||
16 vCPUs 128 GB | 3,852.07 | CNY/month | |||
24 vCPUs 48 GB | 3,646.07 | CNY/month | |||
China (Hong Kong) | instance subscription | - | - | 12.51 | CNY/instance/month |
query node | local SSD | 2 vCPUs 8 GB | 19.78 | CNY/instance/month | |
4 vCPUs 16 GB | 39.56 | CNY/instance/month | |||
8 vCPUs 32 GB | 79.13 | CNY/instance/month | |||
cloud disk | 2 vCPUs 8 GB | 22.70 | CNY/instance/month | ||
4 vCPUs 16 GB | 45.39 | CNY/instance/month | |||
8 vCPUs 32 GB | 90.78 | CNY/instance/month | |||
data node | local SSD | 2 vCPUs 16 GB | 27.41 | CNY/instance/month | |
4 vCPUs 32 GB | 54.83 | CNY/instance/month | |||
6 vCPUs 48 GB | 82.24 | CNY/instance/month | |||
8 vCPUs 64 GB | 109.40 | CNY/instance/month | |||
10 vCPUs 80 GB | 137.06 | CNY/instance/month | |||
12 vCPUs 96 GB | 164.48 | CNY/instance/month | |||
14 vCPUs 112 GB | 191.89 | CNY/instance/month | |||
16 vCPUs 128 GB | 218.79 | CNY/instance/month | |||
cloud disk | 2 vCPUs 16 GB | 28.82 | CNY/instance/month | ||
4 vCPUs 16 GB | 45.39 | CNY/instance/month | |||
4 vCPUs 32 GB | 57.63 | CNY/instance/month | |||
8 vCPUs 16 GB | 68.34 | CNY/instance/month | |||
8 vCPUs 32 GB | 90.78 | CNY/instance/month | |||
8 vCPUs 64 GB | 115.26 | CNY/instance/month | |||
12 vCPUs 24 GB | 102.51 | CNY/instance/month | |||
12 vCPUs 48 GB | 136.17 | CNY/instance/month | |||
12 vCPUs 96 GB | 172.89 | CNY/instance/month | |||
16 vCPUs 32 GB | 136.68 | CNY/instance/month | |||
16 vCPUs 64 GB | 181.56 | CNY/instance/month | |||
16 vCPUs 128 GB | 230.52 | CNY/instance/month | |||
24 vCPUs 48 GB | 205.02 | CNY/instance/month |
Per-node storage cost
Pay-as-you-go
Storage cost for data nodes = (total storage space per data node - free storage space per data node) * unit price * hours * number of data nodes.
Region | Billable item | Instance family | Instance type | Free storage | Storage range | Increment | Unit price | Unit |
China (Hangzhou), China (Shanghai), China (Beijing), China (Shenzhen), China (Zhangjiakou) | storage per data node | local SSD | 2 vCPUs 16 GB | 100 GB | [100 GB, 200 GB] | 50 GB | 0.000588 | CNY/GB/hour |
4 vCPUs 32 GB | 200 GB | [200 GB, 600 GB] | ||||||
6 vCPUs 48 GB | 300 GB | [300 GB, 700 GB] | ||||||
8 vCPUs 16 GB (China (Shanghai) only) | 100 GB | [100 GB, 200 GB] | ||||||
8 vCPUs 64 GB | 400 GB | [400 GB, 800 GB] | ||||||
10 vCPUs 80 GB | 500 GB | [500 GB, 900 GB] | ||||||
12 vCPUs 96 GB | 600 GB | [600 GB, 1,000 GB] | ||||||
14 vCPUs 112 GB | 700 GB | [700 GB, 1,100 GB] | ||||||
16 vCPUs 128 GB | 800 GB | [800 GB, 1,200 GB] | ||||||
cloud disk | 2 vCPUs 16 GB | 0 GB | [100 GB, 200 GB] | 0.0014 | ||||
4 vCPUs 16 GB | [100 GB, 200 GB] | |||||||
4 vCPUs 32 GB | [100 GB, 600 GB] | |||||||
8 vCPUs 16 GB | [100 GB, 200 GB] | |||||||
8 vCPUs 32 GB | [100 GB, 600 GB] | |||||||
8 vCPUs 64 GB | [100 GB, 800 GB] | |||||||
12 vCPUs 24 GB | [100 GB, 500 GB] | |||||||
12 vCPUs 48 GB | [100 GB, 700 GB] | |||||||
12 vCPUs 96 GB | [100 GB, 1,000 GB] | |||||||
16 vCPUs 32 GB | [100 GB, 600 GB] | |||||||
16 vCPUs 64 GB | [100 GB, 800 GB] | |||||||
16 vCPUs 128 GB | [100 GB, 1,200 GB] | |||||||
24 vCPUs 48 GB | [100 GB, 700 GB] | |||||||
Singapore, Germany (Frankfurt) | storage per data node | local SSD | 2 vCPUs 16 GB | 100 GB | [100 GB, 200 GB] | 50 GB | 0.000588 | CNY/GB/hour |
4 vCPUs 32 GB | 200 GB | [200 GB, 600 GB] | ||||||
6 vCPUs 48 GB | 300 GB | [300 GB, 700 GB] | ||||||
8 vCPUs 64 GB | 400 GB | [400 GB, 800 GB] | ||||||
10 vCPUs 80 GB | 500 GB | [500 GB, 900 GB] | ||||||
12 vCPUs 96 GB | 600 GB | [600 GB, 1,000 GB] | ||||||
14 vCPUs 112 GB | 700 GB | [700 GB, 1,100 GB] | ||||||
16 vCPUs 128 GB | 800 GB | [800 GB, 1,200 GB] | ||||||
cloud disk | 2 vCPUs 16 GB | 0 GB | [100 GB, 200 GB] | 0.001421 | ||||
4 vCPUs 16 GB | [100 GB, 200 GB] | |||||||
4 vCPUs 32 GB | [100 GB, 600 GB] | |||||||
8 vCPUs 16 GB | [100 GB, 200 GB] | |||||||
8 vCPUs 32 GB | [100 GB, 600 GB] | |||||||
8 vCPUs 64 GB | [100 GB, 800 GB] | |||||||
12 vCPUs 24 GB | [100 GB, 500 GB] | |||||||
12 vCPUs 48 GB | [100 GB, 700 GB] | |||||||
12 vCPUs 96 GB | [100 GB, 1,000 GB] | |||||||
16 vCPUs 32 GB | [100 GB, 600 GB] | |||||||
16 vCPUs 64 GB | [100 GB, 800 GB] | |||||||
16 vCPUs 128 GB | [100 GB, 1,200 GB] | |||||||
24 vCPUs 48 GB | [100 GB, 700 GB] | |||||||
China (Hong Kong) | storage per data node | local SSD | 2 vCPUs 16 GB | 100 GB | [100 GB, 200 GB] | 50 GB | 0.000588 | CNY/GB/hour |
4 vCPUs 32 GB | 200 GB | [200 GB, 600 GB] | ||||||
6 vCPUs 48 GB | 300 GB | [300 GB, 700 GB] | ||||||
8 vCPUs 64 GB | 400 GB | [400 GB, 800 GB] | ||||||
10 vCPUs 80 GB | 500 GB | [500 GB, 900 GB] | ||||||
12 vCPUs 96 GB | 600 GB | [600 GB, 1,000 GB] | ||||||
14 vCPUs 112 GB | 700 GB | [700 GB, 1,100 GB] | ||||||
16 vCPUs 128 GB | 800 GB | [800 GB, 1,200 GB] | ||||||
cloud disk | 2 vCPUs 16 GB | 0 GB | [100 GB, 200 GB] | 0.001574 | ||||
4 vCPUs 16 GB | [100 GB, 200 GB] | |||||||
4 vCPUs 32 GB | [100 GB, 600 GB] | |||||||
8 vCPUs 16 GB | [100 GB, 200 GB] | |||||||
8 vCPUs 32 GB | [100 GB, 600 GB] | |||||||
8 vCPUs 64 GB | [100 GB, 800 GB] | |||||||
12 vCPUs 24 GB | [100 GB, 500 GB] | |||||||
12 vCPUs 48 GB | [100 GB, 700 GB] | |||||||
12 vCPUs 96 GB | [100 GB, 1,000 GB] | |||||||
16 vCPUs 32 GB | [100 GB, 600 GB] | |||||||
16 vCPUs 64 GB | [100 GB, 800 GB] | |||||||
16 vCPUs 128 GB | [100 GB, 1,200 GB] | |||||||
24 vCPUs 48 GB | [100 GB, 700 GB] |
Subscription
Region | Billing item | Node specification | Unit price | Unit |
China (Hangzhou), China (Shanghai), China (Beijing), China (Shenzhen) | Storage space per data node | local SSD | 0.380700 | CNY/GB/month |
cloud disk | 0.907200 | CNY/GB/month | ||
Singapore, Germany (Frankfurt) | local SSD | 0.380700 | CNY/GB/month | |
cloud disk | 0.920842 | CNY/GB/month | ||
China (Hong Kong) | Storage space per data node | local SSD | 0.35955 | CNY/GB/month |
cloud disk | 0.96303 | CNY/GB/month |
Index storage cost
Pay-as-you-go
OpenSearch (Vector Search Edition) charges for index storage based on the actual index size.
Index storage cost = unit price × hours × (index size - 100).
The first 100 GB of total index size per instance is free; any usage beyond that is charged at the unit price.
Region | Billable item | Free storage quota | Unit price | Unit |
China (Hangzhou), China (Shanghai), China (Beijing), China (Shenzhen), China (Zhangjiakou) | index storage | 100 GB | 0.000415 | CNY/hour/GB |
Singapore, Germany (Frankfurt) | index storage | 100 GB | 0.000358 | CNY/hour/GB |
China (Hong Kong) | index storage | 100 GB | 0.000442 | CNY/hour/GB |
Data update fee
Pay-as-you-go
Alibaba Cloud OpenSearch - Vector Search Edition charges a data update fee based on the amount of resources consumed during data updates.
Each data source and table in an instance includes a free quota of two 4-core, 8 GB resources for data updates. You are charged only for additional quota.
Data update resource cost = (Total data update resources - free data update resources) * Unit price * Hours.
Region | Billing item | Unit price | Unit |
China (Hangzhou), China (Shanghai), China (Beijing), China (Shenzhen), China (Zhangjiakou) | number of data updates | 0.67 | CNY/update/hour |
Singapore, Germany (Frankfurt) | number of data updates | 0.74 | CNY/update/hour |
China (Hong Kong) | number of data updates | 0.76 | CNY/update/hour |
Resize
Scale up/down
Considerations | Category | Description |
Effective time | Immediately | Changes from an upgrade or downgrade take effect immediately. |
Configurable items | Availability edition | You can switch between the standard and high availability editions.
|
Number of query nodes | You can increase or decrease the number of query nodes. If you decrease the number of query nodes, the new count must be at least the number of nodes in use. Example: If an instance has 10 query nodes and 4 are in use, you can reduce the total to a minimum of 4 query nodes. | |
Query node specifications | Only scale-up is supported for query nodes. This means the core count and memory of the new specifications must be greater than or equal to those of the current specifications. | |
Number of data nodes | You can increase or decrease the number of data nodes. If you decrease the number of data nodes, the new count must be at least the number of nodes in use. Example: If an instance has 10 data nodes and 6 are in use, you can reduce the total to a minimum of 6 data nodes. | |
Data node specifications | Only scale-up is supported for data nodes. This means the core count and memory of the new specifications must be greater than or equal to those of the current specifications. | |
Billing rules | Pay-as-you-go | After you apply the new configuration, billing is based on the new specifications. |
Procedure
On the Instance List page, find the instance you want to upgrade or downgrade. In the Actions column, click upgrade/downgrade.
On the Change Configuration page, select the desired availability version, number of query nodes, query node specification, number of data nodes, and data node specification. Select the Service Agreement checkbox, then click Buy Now.
For the query node specification family, you can select the cloud disk type or local SSD type.
1. You can either scale out or scale in resources during a single resize operation. You cannot scale out some resources while scaling in others.
Example: Your instance has 2 query nodes and 2 data nodes.
Not supported: Change to 1 query node (scale-in) and 3 data nodes (scale-out).
Supported: Change to 1 query node and 1 data node (scale-in only).
Supported: Change to 1 query node and 2 data nodes (a partial scale-in).
Supported: Change to 3 query nodes and 3 data nodes (scale-out only).
2. When you scale in query nodes or data nodes, the new count must be at least the number of nodes currently in use. The console validates this requirement and prompts you if it is not met. Changes to the number of data nodes take effect immediately.
Example: If an instance has 10 query nodes and 4 are in use, you can reduce the total to a minimum of 4 query nodes.
3. Changing the specifications of query nodes or data nodes triggers a background update process. This process blocks other resize or editing operations on the instance.
4. After you change the number of query nodes or data nodes, you must manually apply this change in the cluster. For more information, see deployment management.
5. The total number of query nodes in a cluster equals: Number of query nodes × replica count. The total number of data nodes in a cluster equals: replica count × number of data shards. When you scale out, ensure that the number of data nodes you add is a multiple of the shard count. A data shard is the smallest unit of data.
6. Changing the instance from the High-Availability Edition to the Standard Edition deletes the secondary cluster by default. If your dual-availability-zone clusters have inconsistent configurations, changing to a single-availability-zone deployment can affect your service. Proceed with caution.
7. After a resize, the number of data nodes in each availability zone must be at least the number of nodes previously in use in that zone.
Overdue payments and refunds
Impact of expiration or overdue payments
Instance type | Instance status | Actions |
pay-as-you-go instance | On the first day your payment is overdue, you receive SMS and email reminders to top up your account. Your service is unaffected during this period. | Top up your Alibaba Cloud account as soon as possible. |
From day 2 to 14 after your payment is overdue, all pay-as-you-go instances in your account are frozen. | Top up your Alibaba Cloud account to immediately restore the instances. | |
On day 15 after your payment is overdue, the system automatically releases the instances and permanently deletes their data. | The instances cannot be recovered. | |
subscription instance | From day 1 to 14 after expiration, the instance is frozen and unavailable. | Manually renew the instance to restore service immediately. |
On day 15 after expiration, the system automatically releases the instance and permanently deletes its data. | The instance and its data cannot be recovered. |
To prevent service disruption, renew your subscription instance before it expires or enable auto-renewal.
Ensure your account has a sufficient balance.
Refund
OpenSearch - Recall Engine Edition Pay-as-you-go refund policy: All charges for pay-as-you-go instances are non-refundable.
Billing details
Use the Alibaba Cloud management console to view bills for your Alibaba Cloud OpenSearch instance, which include separate charges for the base instance fee, query nodes, and data nodes.
Steps
Log on to the Alibaba Cloud console.
In the upper-right corner of the page, choose Billing > Billing Management.
In the left-side navigation pane, choose Billing Management > Billing Details.
On the Billing Details tab, use Instance Name or Instance ID to view detailed bills.