Realtime Compute for Apache Flink:Create a workspace

Notes

• Creating a Flink workspace involves other Alibaba Cloud services, such as VPC, OSS, and monitoring-related services. For more information about the fees, see Billing items.

• Realtime Compute for Apache Flink uses a storage-compute disaggregation architecture when you attach an OSS Bucket, note the following:

  • If you have not activated OSS, see Activate OSS to activate it. The OSS bucket must be in the same region as the Flink workspace and must use the Standard storage class.

  • Do not set any policies, such as versioning or compliance retention, on the attached OSS bucket. This prevents Flink jobs from running correctly.

    If you enable versioning for your OSS bucket, an excessive number of delete markers can slow down List operations. Therefore, if you enable versioning, you should configure a lifecycle rule to periodically clean up delete markers and unneeded historical versions. This practice reduces storage costs and improves OSS performance. For more information about how to clean up delete markers, see Use a lifecycle rule based on the last modified time to reduce storage costs when versioning is used.

    Warning

    If versioning is disabled for your OSS bucket and you need to set a lifecycle rule, do not set a purge policy for Flink-related folders. These folders include artifacts, flink-jobs, flink-savepoints, sql-artifacts, plan, and flink-sessionclusters. A purge policy might delete Flink-related files and cause Flink jobs to fail.

  • When you activate a workspace and set the storage type to an OSS bucket, you must ensure that the attached bucket is of the Zone-redundant Storage class to ensure zone-redundant high availability for the Flink workspace. For more information, see Switch the storage redundancy type of a bucket. Otherwise, if a zone becomes unavailable, state data cannot be accessed, and Flink cannot guarantee stateful job startup.

    Note

    If a workspace contains both zone-redundant high-availability projects and single-zone projects, directly upgrading the attached bucket to zone-redundant storage may increase storage costs. To balance cost and reliability, you can activate a new workspace specifically for zone-redundant high availability and attach a zone-redundant OSS bucket to it. For more information about OSS storage fees, see Storage fees.

  • If the storage type is OSS Bucket, make sure to disable the Malicious File Detection SDK in Security Center. Enabling this detection can trigger OSS traffic control, which causes Flink jobs to become stuck in the initialization state.

Activate a subscription workspace

1. Log on to the Realtime Compute for Apache Flink console.

2. Next to Stream Compute Flink, click Buy Now.

3. If this is your first purchase, you must grant Flink the permissions to access the required cloud resources.

   Click Go to RAM for Authorization. On the Resource Access Authorization page, click Agree to Authorization. Flink is then granted the required permissions to access cloud resources.

4. On the purchase page, specify the configurations.

   

   Category	Configuration item	Description	Example
   Basic configurations	Billing method	Set this to Subscription. For more information about billing, see Subscription.	Subscription
   	Duration	Select the subscription duration.	1 Month
   	Auto-renewal	If you do not select Enable, you can later modify the renewal policy in the Alibaba Cloud Management Console by choosing Expenses > Renewal Management.	-
   	Processor architecture	The Intel x86 and YiTian ARM architectures are available to meet your requirements. Note Currently, the YiTian ARM architecture is supported only for subscription workspaces in the China (Beijing), China (Ulanqab), and China (Hangzhou) regions. The actual options on the console prevail.	Intel x86 architecture
   	Region	We recommend that you select the same region as your upstream and downstream storage.	Beijing
   Network configurations	Deployment mode	This option is available only for some subscription workspaces that use the Intel x86 architecture. The supported deployment modes are as follows: Single zone (cost-effective): Flink automatically allocates the optimal compute zone in the selected region and connects the resource inventory of the zone. This enables imperceptible resource scheduling across zones and improves elastic scaling. Cross-zone (high business availability): This mode provides zone-redundant high availability. When the zone where a cross-zone project job is running fails, the job is automatically rescheduled to another stable zone in the same region to resume. This effectively prevents service interruptions caused by single-zone failures and ensures job continuity and high availability. Important To ensure end-to-end high availability, you must make sure that your upstream and downstream systems are also highly available. If you do not set Deployment Mode to Cross-zone during activation, you can enable zone-redundant high availability later using other methods. However, you must ensure that the workspace has sufficient cross-zone CU resources. For more information, see Methods 2 and 3 in Zone-redundant high availability.	-
   	CU information	Deployment mode is Single Zone: You only need to configure Single Zone compute resources. If the deployment mode is Cross-zone, you can select or deselect Enable Single-zone Hybrid Purchase as needed. If you clear this option, you only need to configure cross-zone compute resources. If you select this option, you must set the Compute Resource Quota for both single-zone and cross-zone resources. Note Each Realtime Compute for Apache Flink development console and its necessary components require an additional 2 CUs of management resources. These resources are configured automatically. For more information, see Billing items.	-
   	VPC	Select a VPC in the corresponding region. You cannot change the VPC after the workspace is activated. To create a VPC, see Create a VPC and a vSwitch. If the target upstream and downstream storage is in a VPC, the Flink workspace must be in the same VPC. Only the 192.168.0.0/16, 172.16.0.0/12, and 10.0.0.0/8 standard private CIDR blocks and their subnets are supported.	flink-test-vpc
   	vSwitch	If Deployment Mode is set to Cross-zone, you must select two vSwitches in different zones. Each workspace requires two to three IP addresses. You must select at least one vSwitch for each zone. Plan your CIDR blocks based on the scale of your Flink jobs. Note Only created vSwitches that are in the target VPC and in zones supported by Flink are displayed. You can modify the vSwitches after activation. The zone of the selected vSwitch is independent of the zone where the job runs. The vSwitch is used only for network connectivity to ensure that your jobs can access network resources. Do not modify or delete the configurations of these vSwitches. This can affect network connectivity and normal job operation.	flinktest
   Workspace configuration	Workspace name	The name must start with a lowercase letter and can contain lowercase letters, digits, and hyphens (-). It cannot end with a hyphen. The name must be 1 to 60 characters in length. Important The workspace name must be globally unique, including across workspaces created by other users. If the name is a duplicate, the system prompts that the workspace name already exists. You must change the name. The name cannot be changed after the workspace is activated.	flinktestnamespace
   Storage configuration	OSS storage	The OSS Bucket is used to store information such as system checkpoints, job snapshots, logs, and JAR packages. For more information about the content stored in each folder of the OSS Bucket, see File management. Important When you attach an OSS Bucket, see Notes. If Deployment Mode is set to Cross-zone, the OSS configuration is verified. Only OSS Buckets of the Zone-redundant Storage class are displayed. You cannot change the OSS Bucket after the Flink workspace is created.	flink-test-oss
   Tags and tag values		A tag consists of a tag key and a tag value. Tags identify cloud resources. They help you classify, search for, and aggregate cloud resources that have the same characteristics from different dimensions. This simplifies resource management. You can manage tags in a centralized manner in Tag Management.	-
   Monitoring service	Monitoring service	You can select Free Monitoring Service or Pay-as-you-go Prometheus Service. For a comparison of their features, see Feature comparison between Cloud Monitor and ARMS alerting. Free Monitoring Service: After you create the Flink workspace, Cloud Monitor provides basic monitoring and alerting capabilities. For more information, see Free trial. Prometheus Service: After you create the Flink workspace, ARMS is automatically activated to provide commercial Prometheus monitoring and alerting features. For more information about billing, see Billing of Prometheus Service.	Free Monitoring Service

5. To activate the service, select Terms of Service, click Buy Now, and complete the payment.

   After the payment is complete, click Management Console to view the workspace. Workspace creation is typically complete within 5 to 10 minutes.

Activate a pay-as-you-go workspace

1. Log on to the Realtime Compute for Apache Flink console.

2. Click Buy Now next to Stream Compute Flink.

3. (Optional) If this is your first purchase, you must grant Flink the permissions to access the required cloud resources.

   Click Go to RAM for Authorization. On the Resource Access Authorization page, click Agree to Authorization. After the authorization is complete, Flink is granted the permissions to access the required cloud resources.

   Note

   If this is not your first purchase, you can skip this step.

4. On the purchase page, specify the configurations.

   

   Category	Configuration item	Description	Example
   Basic configurations	Billing method	Set this to Pay-as-you-go. For more information about billing, see Pay-as-you-go.	Pay-as-you-go
   	Region	We recommend that you select the same region as your upstream and downstream storage.	Beijing
   Network configurations	VPC	Select a VPC in the corresponding region. You cannot change the VPC after the workspace is activated. To create a VPC, see Create a VPC and a vSwitch. If the target upstream and downstream storage is in a VPC, the Flink workspace must be in the same VPC. Only the 192.168.0.0/16, 172.16.0.0/12, and 10.0.0.0/8 standard private CIDR blocks and their subnets are supported.	flink-test-vpc
   	vSwitch	If Deployment Mode is set to Cross-zone, you must select two vSwitches in different zones. Each workspace requires two to three IP addresses. You must select at least one vSwitch for each zone. Plan your CIDR blocks based on the scale of your Flink jobs. Note Only created vSwitches that are in the target VPC and in zones supported by Flink are displayed. You can modify the vSwitches after activation. The zone of the selected vSwitch is independent of the zone where the job runs. The vSwitch is used only for network connectivity to ensure that your jobs can access network resources. Do not modify or delete the configurations of these vSwitches. This can affect network connectivity and normal job operation.	flinktest
   Workspace configuration	Workspace name	The name must start with a lowercase letter and can contain lowercase letters, digits, and hyphens (-). It cannot end with a hyphen. The name must be 1 to 60 characters in length. Important The workspace name must be globally unique, including across workspaces created by other users. If the name is a duplicate, the system prompts that the workspace name already exists. You must change the name. The name cannot be changed after the workspace is activated.	flinktestnamespace
   	Compute resource Quota Limit	The default value is 1,000 CUs. You can change this value. The maximum value is 9,999,999. Note Each Realtime Compute for Apache Flink development console and its necessary components require an additional 2 CUs of management resources. These resources are configured automatically. For more information, see Billing items.	100
   Storage configuration	OSS storage	The OSS Bucket is used to store information such as system checkpoints, job snapshots, logs, and JAR packages. For more information about the content stored in each folder of the OSS Bucket, see File management. Important When you attach an OSS Bucket, see Notes. If Deployment Mode is set to Cross-zone, the OSS configuration is verified. Only OSS Buckets of the Zone-redundant Storage class are displayed. You cannot change the OSS Bucket after the Flink workspace is created.	flink-test-oss
   Tags and tag values		A tag consists of a tag key and a tag value. Tags identify cloud resources. They help you classify, search for, and aggregate cloud resources that have the same characteristics from different dimensions. This simplifies resource management. You can manage tags in a centralized manner in Tag Management.	-
   Monitoring service	Monitoring service	You can select Free Monitoring Service or Pay-as-you-go Prometheus Service. For a comparison of their features, see Feature comparison between Cloud Monitor and ARMS alerting. Free Monitoring Service: After you create the Flink workspace, Cloud Monitor provides basic monitoring and alerting capabilities. For more information, see Free trial. Prometheus Service: After you create the Flink workspace, ARMS is automatically activated to provide commercial Prometheus monitoring and alerting features. For more information about billing, see Billing of Prometheus Service.	Free Monitoring Service

5. To activate the service, select Terms of Service, click Buy Now, and complete the payment.

   After the payment is complete, click Management Console to view the workspace. Workspace creation is typically complete within 5 to 10 minutes.

What to do next

After you activate the Flink workspace, you can perform the following operations.

References

• Why is my activated Flink workspace not displayed in the Realtime Compute for Apache Flink console?

• How do I view information such as the workspace ID?

• Can I change the workspace name?

• Can I change the VPC and vSwitch?