Exclusive resource groups for Data Integration provide dedicated Elastic Compute Service (ECS) instances for data integration tasks with high concurrency requirements that cannot be staggered. They give you isolated computing resources for fast and stable data movement across complex network environments.
DataWorks no longer recommends exclusive resource groups for Data Integration. Use serverless resource groups instead. A single serverless resource group covers data synchronization, task scheduling and execution, and API operation calls and management — replacing exclusive resource groups for scheduling, Data Integration, and DataService Studio, as well as shared resource groups.
Capabilities
Exclusive resource groups for Data Integration support:
Data synchronization in complex network environments, including cross-cloud platforms such as Alibaba Finance Cloud and Alibaba Gov Cloud, across different Alibaba Cloud accounts, and with on-premises data centers.
Offline, real-time, full, and incremental data synchronization between a wide range of data sources. For supported data sources, see Supported data sources and synchronization solutions.
Exclusive resource groups for Data Integration guarantee the total number of concurrent threads for data synchronization instances, but not the number of concurrent data synchronization instances. To guarantee the latter, you must also purchase an exclusive resource group for scheduling.
Limitations
Billing model: Exclusive resource groups for Data Integration are billed by subscription. You cannot delete or release a resource group before its subscription expires. After a subscription expires, the service is suspended and the resource group is released after a specific period. After release, all tasks that relied on the resource group can no longer run.
Regional scope: Exclusive resource groups for Data Integration do not support cross-region use. For example, a resource group in the China (Shanghai) region can only be used by workspaces in China (Shanghai).
Network type: Exclusive resource groups do not support Alibaba Cloud classic networks. If your data source is in a classic network, migrate it to a virtual private cloud (VPC) or enable public network access.
Specifications and pricing
Exclusive resource groups for Data Integration are billed by subscription. Choose the specification that fits your workload.
Region | Specifications | Unit Price (USD/Month) |
China (Hangzhou) | 4 vCPU 8 GiB | 76.23 |
8 vCPU 16 GiB | 152.45 | |
12 vCPU 24 GiB | 228.68 | |
16 vCPU 32 GiB | 304.91 | |
24 vCPU 48 GiB | 457.36 | |
32 vCPU 64 GiB | 609.82 | |
China (Shanghai) | 4 vCPU 8 GiB | 76.23 |
8 vCPU 16 GiB | 152.45 | |
12 vCPU 24 GiB | 228.68 | |
16 vCPU 32 GiB | 304.91 | |
24 vCPU 48 GiB | 457.36 | |
32 vCPU 64 GiB | 609.82 | |
China (Shenzhen) | 4 vCPU 8 GiB | 76.23 |
8 vCPU 16 GiB | 152.45 | |
12 vCPU 24 GiB | 228.68 | |
16 vCPU 32 GiB | 304.91 | |
24 vCPU 48 GiB | 457.36 | |
32 vCPU 64 GiB | 609.82 | |
China (Beijing) | 4 vCPU 8 GiB | 76.23 |
8 vCPU 16 GiB | 152.45 | |
12 vCPU 24 GiB | 228.68 | |
16 vCPU 32 GiB | 304.91 | |
24 vCPU 48 GiB | 457.36 | |
32 vCPU 64 GiB | 609.82 | |
China (Zhangjiakou) | 4 vCPU 8 GiB | 57.27 |
8 vCPU 16 GiB | 114.34 | |
12 vCPU 24 GiB | 171.61 | |
16 vCPU 32 GiB | 228.68 | |
24 vCPU 48 GiB | 343.02 | |
32 vCPU 64 GiB | 457.36 | |
China (Chengdu) | 4 vCPU 8 GiB | 62.49 |
8 vCPU 16 GiB | 124.79 | |
12 vCPU 24 GiB | 187.09 | |
16 vCPU 32 GiB | 249.38 | |
24 vCPU 48 GiB | 374.17 | |
32 vCPU 64 GiB | 498.77 | |
US (Silicon Valley) | 4 vCPU 8 GiB | 149.26 |
8 vCPU 16 GiB | 298.53 | |
12 vCPU 24 GiB | 447.77 | |
16 vCPU 32 GiB | 597.06 | |
24 vCPU 48 GiB | 895.55 | |
32 vCPU 64 GiB | 1,195.00 | |
US (Virginia) | 4 vCPU 8 GiB | 110.63 |
8 vCPU 16 GiB | 221.26 | |
12 vCPU 24 GiB | 331.89 | |
16 vCPU 32 GiB | 442.53 | |
24 vCPU 48 GiB | 663.78 | |
32 vCPU 64 GiB | 885.06 | |
China (Hong Kong) | 4 vCPU 8 GiB | 138.73 |
8 vCPU 16 GiB | 276.58 | |
12 vCPU 24 GiB | 416.18 | |
16 vCPU 32 GiB | 553.16 | |
24 vCPU 48 GiB | 832.35 | |
32 vCPU 64 GiB | 1,106.32 | |
Singapore | 4 vCPU 8 GiB | 138.73 |
8 vCPU 16 GiB | 276.58 | |
12 vCPU 24 GiB | 416.18 | |
16 vCPU 32 GiB | 553.16 | |
24 vCPU 48 GiB | 832.35 | |
32 vCPU 64 GiB | 1,106.32 | |
Germany (Frankfurt) | 4 vCPU 8 GiB | 138.73 |
8 vCPU 16 GiB | 276.58 | |
12 vCPU 24 GiB | 408.96 | |
16 vCPU 32 GiB | 553.16 | |
24 vCPU 48 GiB | 817.91 | |
32 vCPU 64 GiB | 1,106.32 | |
Malaysia (Kuala Lumpur) | 4 vCPU 8 GiB | 130.01 |
8 vCPU 16 GiB | 259.06 | |
12 vCPU 24 GiB | 390.05 | |
16 vCPU 32 GiB | 517.92 | |
24 vCPU 48 GiB | 780.07 | |
32 vCPU 64 GiB | 1,035.84 | |
Indonesia (Jakarta) | 4 vCPU 8 GiB | 138.73 |
8 vCPU 16 GiB | 276.58 | |
12 vCPU 24 GiB | 416.18 | |
16 vCPU 32 GiB | 553.16 | |
24 vCPU 48 GiB | 832.35 | |
32 vCPU 64 GiB | 1,106.32 | |
Japan (Tokyo) | 4 vCPU 8 GiB | 158.92 |
8 vCPU 16 GiB | 316.97 | |
12 vCPU 24 GiB | 474.14 | |
16 vCPU 32 GiB | 633.94 | |
24 vCPU 48 GiB | 948.27 | |
32 vCPU 64 GiB | 1,268.75 | |
UK (London) | 4 vCPU 8 GiB | 104.67 |
8 vCPU 16 GiB | 209.33 | |
12 vCPU 24 GiB | 314.00 | |
16 vCPU 32 GiB | 418.67 | |
24 vCPU 48 GiB | 628.00 | |
32 vCPU 64 GiB | 837.14 | |
UAE (Dubai) | 4 vCPU 8 GiB | 166.38 |
8 vCPU 16 GiB | 331.80 | |
12 vCPU 24 GiB | 498.38 | |
16 vCPU 32 GiB | 663.80 | |
24 vCPU 48 GiB | 996.76 | |
32 vCPU 64 GiB | 1,327.59 | |
SAU (Riyadh - Partner Region) | 4 vCPU 8 GiB | 166.48 |
8 vCPU 16 GiB | 331.90 | |
12 vCPU 24 GiB | 499.42 | |
16 vCPU 32 GiB | 663.79 | |
24 vCPU 48 GiB | 998.82 | |
32 vCPU 64 GiB | 1,327.58 |
Offline synchronization: maximum concurrent threads by specification
| Specifications | Max concurrent threads | Typical use case |
|---|---|---|
| 4 vCPU 8 GiB | 8 (distributed processing disabled) | 1 MySQL source database for real-time sync; single Kafka source; other single-table real-time tasks |
| 8 vCPU 16 GiB | 16 | 2–5 MySQL source databases for real-time sync |
| 12 vCPU 24 GiB | 24 | 6 or more MySQL source databases for real-time sync; PolarDB-X source for real-time sync |
| 16 vCPU 32 GiB | 32 | — |
| 24 vCPU 48 GiB | 48 | — |
For the 4 vCPU 8 GiB specification, you can set the maximum concurrent threads to 8 only when distributed processing is disabled. When distributed processing is enabled, you cannot set the number of concurrent threads to 8.
If your concurrency needs exceed the current specification, scale out the resource group or upgrade the specification. See Scale out or scale in a resource group and Change specifications.
Pricing
| Region | Specifications | Unit price (USD/month) |
|---|---|---|
| China (Hangzhou) | 4 vCPU 8 GiB | 76.23 |
| 8 vCPU 16 GiB | 152.45 | |
| 12 vCPU 24 GiB | 228.68 | |
| 16 vCPU 32 GiB | 304.91 | |
| 24 vCPU 48 GiB | 457.36 | |
| 32 vCPU 64 GiB | 609.82 | |
| China (Shanghai) | 4 vCPU 8 GiB | 76.23 |
| 8 vCPU 16 GiB | 152.45 | |
| 12 vCPU 24 GiB | 228.68 | |
| 16 vCPU 32 GiB | 304.91 | |
| 24 vCPU 48 GiB | 457.36 | |
| 32 vCPU 64 GiB | 609.82 | |
| China (Shenzhen) | 4 vCPU 8 GiB | 76.23 |
| 8 vCPU 16 GiB | 152.45 | |
| 12 vCPU 24 GiB | 228.68 | |
| 16 vCPU 32 GiB | 304.91 | |
| 24 vCPU 48 GiB | 457.36 | |
| 32 vCPU 64 GiB | 609.82 | |
| China (Beijing) | 4 vCPU 8 GiB | 76.23 |
| 8 vCPU 16 GiB | 152.45 | |
| 12 vCPU 24 GiB | 228.68 | |
| 16 vCPU 32 GiB | 304.91 | |
| 24 vCPU 48 GiB | 457.36 | |
| 32 vCPU 64 GiB | 609.82 | |
| China (Zhangjiakou) | 4 vCPU 8 GiB | 57.27 |
| 8 vCPU 16 GiB | 114.34 | |
| 12 vCPU 24 GiB | 171.61 | |
| 16 vCPU 32 GiB | 228.68 | |
| 24 vCPU 48 GiB | 343.02 | |
| 32 vCPU 64 GiB | 457.36 | |
| China (Chengdu) | 4 vCPU 8 GiB | 62.49 |
| 8 vCPU 16 GiB | 124.79 | |
| 12 vCPU 24 GiB | 187.09 | |
| 16 vCPU 32 GiB | 249.38 | |
| 24 vCPU 48 GiB | 374.17 | |
| 32 vCPU 64 GiB | 498.77 | |
| US (Silicon Valley) | 4 vCPU 8 GiB | 149.26 |
| 8 vCPU 16 GiB | 298.53 | |
| 12 vCPU 24 GiB | 447.77 | |
| 16 vCPU 32 GiB | 597.06 | |
| 24 vCPU 48 GiB | 895.55 | |
| 32 vCPU 64 GiB | 1,195.00 | |
| US (Virginia) | 4 vCPU 8 GiB | 110.63 |
| 8 vCPU 16 GiB | 221.26 | |
| 12 vCPU 24 GiB | 331.89 | |
| 16 vCPU 32 GiB | 442.53 | |
| 24 vCPU 48 GiB | 663.78 | |
| 32 vCPU 64 GiB | 885.06 | |
| China (Hong Kong) | 4 vCPU 8 GiB | 138.73 |
| 8 vCPU 16 GiB | 276.58 | |
| 12 vCPU 24 GiB | 416.18 | |
| 16 vCPU 32 GiB | 553.16 | |
| 24 vCPU 48 GiB | 832.35 | |
| 32 vCPU 64 GiB | 1,106.32 | |
| Singapore | 4 vCPU 8 GiB | 138.73 |
| 8 vCPU 16 GiB | 276.58 | |
| 12 vCPU 24 GiB | 416.18 | |
| 16 vCPU 32 GiB | 553.16 | |
| 24 vCPU 48 GiB | 832.35 | |
| 32 vCPU 64 GiB | 1,106.32 | |
| Germany (Frankfurt) | 4 vCPU 8 GiB | 138.73 |
| 8 vCPU 16 GiB | 276.58 | |
| 12 vCPU 24 GiB | 408.96 | |
| 16 vCPU 32 GiB | 553.16 | |
| 24 vCPU 48 GiB | 817.91 | |
| 32 vCPU 64 GiB | 1,106.32 | |
| Malaysia (Kuala Lumpur) | 4 vCPU 8 GiB | 130.01 |
| 8 vCPU 16 GiB | 259.06 | |
| 12 vCPU 24 GiB | 390.05 | |
| 16 vCPU 32 GiB | 517.92 | |
| 24 vCPU 48 GiB | 780.07 | |
| 32 vCPU 64 GiB | 1,035.84 | |
| Indonesia (Jakarta) | 4 vCPU 8 GiB | 138.73 |
| 8 vCPU 16 GiB | 276.58 | |
| 12 vCPU 24 GiB | 416.18 | |
| 16 vCPU 32 GiB | 553.16 | |
| 24 vCPU 48 GiB | 832.35 | |
| 32 vCPU 64 GiB | 1,106.32 | |
| Japan (Tokyo) | 4 vCPU 8 GiB | 158.92 |
| 8 vCPU 16 GiB | 316.97 | |
| 12 vCPU 24 GiB | 474.14 | |
| 16 vCPU 32 GiB | 633.94 | |
| 24 vCPU 48 GiB | 948.27 | |
| 32 vCPU 64 GiB | 1,268.75 | |
| UK (London) | 4 vCPU 8 GiB | 104.67 |
| 8 vCPU 16 GiB | 209.33 | |
| 12 vCPU 24 GiB | 314.00 | |
| 16 vCPU 32 GiB | 418.67 | |
| 24 vCPU 48 GiB | 628.00 | |
| 32 vCPU 64 GiB | 837.14 | |
| UAE (Dubai) | 4 vCPU 8 GiB | 166.38 |
| 8 vCPU 16 GiB | 331.80 | |
| 12 vCPU 24 GiB | 498.38 | |
| 16 vCPU 32 GiB | 663.80 | |
| 24 vCPU 48 GiB | 996.76 | |
| 32 vCPU 64 GiB | 1,327.59 | |
| SAU (Riyadh - Partner Region) | 4 vCPU 8 GiB | 166.48 |
| 8 vCPU 16 GiB | 331.90 | |
| 12 vCPU 24 GiB | 499.42 | |
| 16 vCPU 32 GiB | 663.79 | |
| 24 vCPU 48 GiB | 998.82 | |
| 32 vCPU 64 GiB | 1,327.58 |
Manage your subscription
Scale out or scale in: Purchase additional machines or reduce the machine count at any time. For the procedure and important considerations, see Scale out or scale in a resource group.
Change specifications: Upgrade or downgrade the specification for all machines in the resource group. Changing the specification affects every machine in the group, not individual machines. For the procedure and considerations, see Change specifications.
Renewal and release: Renew the subscription before it expires, or let it expire and the resource group will be automatically released after a specific period. For details, see Expiration and renewal.
After a resource group is released, all tasks that relied on it can no longer run.
Network connectivity
Exclusive resource groups for Data Integration are essentially a group of Alibaba Cloud ECS instances, similar to other types of exclusive resource groups. After purchase, you must attach the resource group to a VPC, then choose the connectivity solution that matches your data source's network environment.
| Data source network environment | Connectivity solution |
|---|---|
| Internet | Connect the data source directly to the VPC attached to the resource group. |
| VPC in the same region as the resource group | Attach the resource group to the same VPC as the data source. The system automatically adds the route needed for connectivity. |
| VPC in a different region from the resource group | Use Express Connect or VPN to connect the two VPCs. Add a route pointing to the destination database IP address. |
| On-premises data center | Use Express Connect or VPN to connect the data center to the resource group's VPC. Add a route pointing to the destination database IP address. |
| Classic network | Not supported. Migrate the data source to a VPC. |
For the full list of connectivity options, see Network connectivity solutions.
Whitelist configuration: If your data source restricts access by whitelist, add the resource group's endpoint to the whitelist. For steps, see Add a whitelist.
FAQ
For answers to common questions about exclusive resource groups, see Exclusive resource groups FAQ.