All Products
Search
Document Center

OpenSearch:Billing overview of OpenSearch LLM-Based Conversational Search Edition

Last Updated:Sep 18, 2023

This topic provides an overview of the billing of OpenSearch LLM-Based Conversational Search Edition.

Billing method

Billing method

Description

Pay-as-you-go

The pay-as-you-go billing method allows you to use resources first and pay for them afterward. The system generates a bill per hour and deducts fees from the balance of your Alibaba Cloud account based on the hourly bill. All the hourly bills belong to a total bill.

Billable items

Billable item

Unit

Available range

Step

Description

Storage resources

GB

1 to 100

1 GB

A fixed storage fee is charged. The storage fee varies based on the storage quota that you select when you purchase an OpenSearch LLM-Based Conversational Search Edition instance.

Computing resources

CU

0 to 3600

N/A

You are charged based on the amount of computing resources that you use per hour. The more requests and larger queries per second (QPS), the more computing resources you consume.

Pricing

Pricing of storage resources

Region

Unit price (USD per GB per hour)

Example

Singapore

0.030

If you purchase an OpenSearch LLM-Based Conversational Search Edition instance and configure it with a storage quota of 10 GB, the hourly storage fee is calculated based on the following formula: 10 GB × USD 0.03 per GB per hour = USD 0.3.

Pricing of computing resources

Region

Unit price (USD per CU per hour)

Example

Singapore

0.086

If Instance A used 100 compute units (CUs) of computing resources for conversational search between 11 and 12 a.m., the computing resource fee within the 1 hour is calculated based on the following formula: 100 CUs × USD 0.086 per CU per hour = USD 8.6.

Important
  • The computing resource fee varies with the actual usage of computing resources that you consume in the console and for API requests.

  • On average, you can initiate 10 API requests for conversational search by using 1 CU of computing resources. The actual usage of computing resources varies based on the complexity of the conversation.

  • Each instance supports up to 10 QPS for conversational search. If the QPS of an instance exceeds the limit, the Service Level Agreement (SLA) cannot be guaranteed. To increase the QPS limit, apply in advance.

Billing example

User A purchases an OpenSearch LLM-Based Conversational Search Edition instance and configures it with a storage quota of 2 GB.

Between 6 and 7 a.m., User A does not initiate API requests for conversational search, and no computing resources are consumed.

The total fee within the 1 hour is calculated based on the following formula: 2 GB × USD 0.03 per GB per hour = USD 0.06.

Between 7 and 8 a.m., User A initiates 100 API requests for conversational search, and 10 CUs of computing resources are consumed.

The total fee within the 1 hour is calculated based on the following formula: 2 GB × USD 0.03 per GB per hour + 10 CUs × USD 0.086 per CU per hour = USD 0.92.

Between 8 and 9 a.m., User A initiates 1,000 API requests for conversational search, and 100 CUs of computing resources are consumed.

The total fee within the 1 hour is calculated based on the following formula: 2 GB × USD 0.03 per GB per hour + 100 CUs × USD 0.086 per CU per hour = USD 8.66.