Quick Deployment

This section helps users efficiently deploy the "Building a New API LLM Gateway" solution. When deploying this solution with one click, follow the steps and instructions in this section to complete the quick deployment.

1. Go to the solution details page, click "Start Deployment", select the corresponding deployment region, and jump to the RFS deployment page.
   Figure 1 RFS Purchase Page
   

2. Click "Next" to confirm the basic configuration and set the password.
   Figure 2 Basic Configuration Diagram

   

   Table 1 Community Standalone Edition Configuration Instructions

   Parameter	Type	Mandatory	Description	Default Value
   vpc_name	string	Mandatory	Virtual private cloud name.Value range:4-64 characters, supports lowercase letters, numbers, and hyphens; must start with a lowercase letter and cannot end with a hyphen.	newapi-llm-gateway-demo
   security_group_name	string	Mandatory	Security group name.Please refer to the deployment guide for configuring security group rules. Value range:4-64 characters, supports lowercase letters, numbers, and hyphens; must start with a lowercase letter and cannot end with a hyphen.	newapi-llm-gateway-demo
   ecs_name	string	Mandatory	Cloud server instance name, Value range:4-64 characters, supports lowercase letters, numbers, and hyphens; must start with a lowercase letter and cannot end with a hyphen.	newapi-llm-gateway-demo
   ecs_flavor	string	Mandatory	Cloud server instance specification. The naming rule for Flexus Cloud Server X instance specification IDs is x1.?u.?g. For example, the specification ID for 2vCPUs4GiB is x1.2u.4g. For specific Huawei Cloud Flexus Cloud Server X instance specifications, refer to the X instance console configuration. For Elastic Cloud Server (ECS) specifications, refer to the ECS console configuration.	x1.8u.16g
   ecs_password	string	Mandatory	Cloud server password. Length: 8-26 characters. The password must contain at least three of the following four types: uppercase letters, lowercase letters, digits, and special characters (!@$%^-_=+[{}]:,./?). The default administrator account is root.	root
   system_disk_size	number	Mandatory	System disk size(GB). The disk type defaults to High I/O.Value range: 40-1,024. Disk shrinking is not supported.	100
   bandwidth_size	number	Mandatory	Elastic IP bandwidth (Mbit/s), billed by traffic, value range: 1-300.	300
   charging_mode	string	Mandatory	Billing mode: postPaid (pay-as-you-go)or prePaid (yearly/monthly).	postPaid
   charging_unit	string	Mandatory	Subscription period type: month or year, effective only in prePaid mode.	month
   charging_period	number	Mandatory	Subscription period: 1-9 (months) or 1-3 (years), effective only in prePaid mode.	1

   Figure 3 Basic Configuration Diagram

   

   Table 2 CCE Container High Availability Edition Configuration Instructions

   Parameter	Type	Mandatory	Description	Default Value
   resource_name_prefix	string	Mandatory	Solution name.Value range:4-64 characters, supports lowercase letters, numbers, and hyphens; must start with a lowercase letter and cannot end with a hyphen.	newapi-llm-gateway-ha-demo
   cce_cluster_flavor	string	Mandatory	CCE Turbo cluster specifications. The specification cannot be changed after the cluster is created. Optional values: cce.s2.small, cce.s2.medium, cce.s2.large, cce.s2.xlarge. For specific specifications, please refer to the deployment guide configuration.	cce.s2.small
   cce_node_pool_password	string	Mandatory	CCE cluster node password, used for cluster node login. Valid range: 8-24 characters. The password must contain at least one uppercase letter, one lowercase letter, and include either a digit or a special character (~!^*-=_+,).	-
   cce_node_pool_flavor	string	Mandatory	Cloud server instance specification.The naming rule for Flexus Cloud Server X instance specification IDs is x1.?u.?g. For example, the specification ID for 2vCPUs4GiB is x1.2u.4g. Please use specifications of 3vCPUs6GiB or above. For specific Huawei Cloud Flexus X instance specifications, refer to the X instance console configuration. For Elastic Cloud Server (ECS) specifications, refer to the ECS console configuration.	x1.8u.16g
   cce_node_pool_count	number	Mandatory	The initial number of nodes in a CCE node pool has a minimum value of 3 and a maximum value of 10.	3
   rds_flavor	string	Mandatory	Cloud Database RDS for MySQL instance specification. This solution creates a primary/standby edition by default. For other specifications, please refer to the RDS console configuration.	rds.mysql.n1-v2.xlarge.2.ha(4U8G)
   mysql_password	string	Mandatory	The password for the MySQL database must be 8-24 characters long, including at least one uppercase letter, one lowercase letter, and one number or special character (~!^*-=_+,).	-
   mysql_user_password	string	Mandatory	The password for the database user of the MySQL database must be 8-24 characters long, including at least one uppercase letter, one lowercase letter, and one number or special character (~!^*-=_+,). It cannot be identical to the username or the reverse of the username.	-
   redis_capacity	number	Mandatory	Distributed Cache Service Redis Edition instance specifications. Available options: 1GB-64GB.	1GB
   redis_password	string	Mandatory	Redis database password. Valid range: 8-24 characters. The password must contain at least uppercase letters, lowercase letters, and include numbers or special characters (~!^*-=_+,).	-
   session_secret	string	Mandatory	Session key for the New API management backend. Valid range: 8-24 characters. The password must contain at least uppercase letters, lowercase letters, and include numbers or special characters (~!^*-=_+,).	-
   crypto_secret	string	Mandatory	Encryption key used for encrypting large model API keys. Valid range: 8-24 characters. The password must contain at least uppercase letters, lowercase letters, and include numbers or special characters (~!^*-=_+,).	-
   charging_mode	string	Mandatory	Billing mode: postPaid (pay-as-you-go)or prePaid (yearly/monthly).	postPaid
   charging_unit	string	Mandatory	Subscription period type: month or year, effective only in prePaid mode.	month
   charging_period	number	Mandatory	Subscription period: 1-9 (months) or 1-3 (years), effective only in prePaid mode.	1

3. (Optional) In the encryption activation interface, select "Custom Key" from the key dropdown menu, then click "OK".
   Figure 4 Encryption Settings
   

4. On the Configure Stack page, select "rf_admin_trust" from the Agency drop-down list and click "Next" (This step is optional if you use an account (HUAWEI ID) or use an IAM user in the admin user group.).
   Figure 5 Configuring a stack
   

5. On the Confirm Configurations page, confirm the configurations and click "Create Execution Plan".
   Figure 6 Confirming the configurations

   

6. In the displayed Create Execution Plan dialog box, enter an execution plan name and click "OK".
   Figure 7 Creating an execution plan

   

7. Wait until the status of the execution plan changes to Available and then click "Deploy" in the Operation column.
   Figure 8 Execution plan page

   

8. In the displayed dialog box, click "Execute".The system will complete the deduction through automatic payment. Please ensure that your account has sufficient balance.
   Figure 9 Confirming the execution plan

   

   

   • When creating a new account or when the account balance is insufficient, please go to the "Billing Center > Top-up" page to recharge. For details, please refer toAccount Top-up.
   • If you have selected a yearly/monthly billing mode and automatic payment fails due to insufficient account balance, please go to the "Billing Center > Unpaid Orders" page to complete the payment manually.

9. Click the Events tab and check whether the message "Apply required resource success." is displayed in the Description column. If it is, the solution is successfully deployed.
   Figure 10 Solution deployed

   

10. On the Outputs tab, view the access description of the platform. After the stack is successfully deployed, the application setup script starts to execute. Please wait patiently for about 20 minutes (subject to network fluctuations).
    Figure 11 Access description